title
stringlengths 6
244
| abstract
stringlengths 19
5.09k
| label
stringclasses 10
values |
|---|---|---|
Bayesian Semiparametric Modeling of Response Mechanism for Nonignorable
Missing Data | Statistical inference with nonresponse is quite challenging, especially when
the response mechanism is nonignorable. In this case, the validity of
statistical inference depends on untestable correct specification of the
response model. To avoid the misspecification, we propose semiparametric
Bayesian estimation in which an outcome model is parametric, but the response
model is semiparametric in that we do not assume any parametric form for the
nonresponse variable. We adopt penalized spline methods to estimate the unknown
function. We also consider a fully nonparametric approach to modeling the
response mechanism by using radial basis function methods. Using Polya-gamma
data augmentation, we developed an efficient posterior computation algorithm
via Gibbs sampling in which most full conditional distributions can be obtained
in familiar forms. The performance of the proposed method is demonstrated in
simulation studies and an application to longitudinal data.
| stat |
How Does Batch Normalization Help Optimization? | Batch Normalization (BatchNorm) is a widely adopted technique that enables
faster and more stable training of deep neural networks (DNNs). Despite its
pervasiveness, the exact reasons for BatchNorm's effectiveness are still poorly
understood. The popular belief is that this effectiveness stems from
controlling the change of the layers' input distributions during training to
reduce the so-called "internal covariate shift". In this work, we demonstrate
that such distributional stability of layer inputs has little to do with the
success of BatchNorm. Instead, we uncover a more fundamental impact of
BatchNorm on the training process: it makes the optimization landscape
significantly smoother. This smoothness induces a more predictive and stable
behavior of the gradients, allowing for faster training.
| stat |
Model selection for ecological community data using tree shrinkage
priors | Researchers and managers model ecological communities to infer the biotic and
abiotic variables that shape species' ranges, habitat use, and co-occurrence
which, in turn, are used to support management decisions and test ecological
theories. Recently, species distribution models were developed for and applied
to data from ecological communities. Model development and selection for
ecological community data is difficult because a high level of complexity is
desired and achieved by including numerous parameters, which can degrade
predictive accuracy and be challenging to interpret and communicate. Like other
statistical models, multi-species distribution models can be overparameterized.
Regularization is a technique that optimizes predictive accuracy by shrinking
or eliminating model parameters. For Bayesian models, the prior distribution
automatically regularizes parameters. We propose a tree shrinkage prior for
Bayesian multi-species distributions models that performs regularization and
reduces the number of regression coefficients associated with predictor
variables. Using this prior, the number of regression coefficients in
multi-species distributions models is reduced by estimation of unique
regression coefficients for a smaller number of guilds rather than a larger
number of species. We demonstrated our tree shrinkage prior using examples of
presence-absence data for six species of aquatic vegetation and relative
abundance data for 15 species of fish. Our results show that the tree shrinkage
prior can increase the predictive accuracy of multi-species distribution models
and enable researchers to infer the number and species composition of guilds
from ecological community data.
| stat |
Uncertainty Quantification for High Dimensional Sparse Nonparametric
Additive Models | Statistical inference in high dimensional settings has recently attracted
enormous attention within the literature. However, most published work focuses
on the parametric linear regression problem. This paper considers an important
extension of this problem: statistical inference for high dimensional sparse
nonparametric additive models. To be more precise, this paper develops a
methodology for constructing a probability density function on the set of all
candidate models. This methodology can also be applied to construct confidence
intervals for various quantities of interest (such as noise variance) and
confidence bands for the additive functions. This methodology is derived using
a generalized fiducial inference framework. It is shown that results produced
by the proposed methodology enjoy correct asymptotic frequentist properties.
Empirical results obtained from numerical experimentation verify this
theoretical claim. Lastly, the methodology is applied to a gene expression data
set and discovers new findings for which most existing methods based on
parametric linear modeling failed to observe.
| stat |
Component-wise Adaptive Trimming For Robust Mixture Regression | Parameter estimation of mixture regression model using the expectation
maximization (EM) algorithm is highly sensitive to outliers. Here we propose a
fast and efficient robust mixture regression algorithm, called Component-wise
Adaptive Trimming (CAT) method. We consider simultaneous outlier detection and
robust parameter estimation to minimize the effect of outlier contamination.
Robust mixture regression has many important applications including in human
cancer genomics data, where the population often displays strong heterogeneity
added by unwanted technological perturbations. Existing robust mixture
regression methods suffer from outliers as they either conduct parameter
estimation in the presence of outliers, or rely on prior knowledge of the level
of outlier contamination. CAT was implemented in the framework of
classification expectation maximization, under which a natural definition of
outliers could be derived. It implements a least trimmed squares (LTS) approach
within each exclusive mixing component, where the robustness issue could be
transformed from the mixture case to simple linear regression case. The high
breakdown point of the LTS approach allows us to avoid the pre-specification of
trimming parameter. Compared with multiple existing algorithms, CAT is the most
competitive one that can handle and adaptively trim off outliers as well as
heavy tailed noise, in different scenarios of simulated data and real genomic
data. CAT has been implemented in an R package `RobMixReg' available in CRAN.
| stat |
The Shortest Confidence Interval for the Ratio of Quantiles of the Dagum
Distribution | J\k{e}drzejczak et al. (2018) constructed a confidence interval for a ratio
of quantiles coming from the Dagum distribution, which is frequently applied as
a theoretical model in numerous income distribution analyses. The proposed
interval is symmetric with respect to the ratio of sample quantiles, which
result may be unsatisfactory in many practical applications. The search for a
confidence interval with a smaller length led to the derivation of the shortest
interval with the ends being asymmetric relative to the ratio of sample
quantiles. In the paper, the existence of the shortest confidence interval is
shown and the method of obtaining such an interval is presented. The results of
the calculation show a reduction in the length of the confidence intervals by
several percent in relation to the symmetric confidence interval.
| stat |
Nonlinear spectral analysis: A local Gaussian approach | The spectral distribution $f(\omega)$ of a stationary time series
$\{Y_t\}_{t\in\mathbb{Z}}$ can be used to investigate whether or not periodic
structures are present in $\{Y_t\}_{t\in\mathbb{Z}}$, but $f(\omega)$ has some
limitations due to its dependence on the autocovariances $\gamma(h)$. For
example, $f(\omega)$ can not distinguish white i.i.d. noise from GARCH-type
models (whose terms are dependent, but uncorrelated), which implies that
$f(\omega)$ can be an inadequate tool when $\{Y_t\}_{t\in\mathbb{Z}}$ contains
asymmetries and nonlinear dependencies.
Asymmetries between the upper and lower tails of a time series can be
investigated by means of the local Gaussian autocorrelations introduced in
Tj{\o}stheim and Hufthammer (2013), and these local measures of dependence can
be used to construct the local Gaussian spectral density presented in this
paper. A key feature of the new local spectral density is that it coincides
with $f(\omega)$ for Gaussian time series, which implies that it can be used to
detect non-Gaussian traits in the time series under investigation. In
particular, if $f(\omega)$ is flat, then peaks and troughs of the new local
spectral density can indicate nonlinear traits, which potentially might
discover local periodic phenomena that remain undetected in an ordinary
spectral analysis.
| stat |
Margin-free classification and new class detection using finite
Dirichlet mixtures | We present a margin-free finite mixture model which allows us to
simultaneously classify objects into known classes and to identify possible new
object types using a set of continuous attributes. This application is
motivated by the needs of identifying and possibly detecting new types of a
particular kind of stars known as variable stars. We first suitably transform
the physical attributes of the stars onto the simplex to achieve scale
invariance while maintaining their dependence structure. This allows us to
compare data collected by different sky surveys which can have different
scales. The model hence combines a mixture of Dirichlet mixtures to represent
the known classes with the semi-supervised classification strategy of Vatanen
et al. (2012) for outlier detection. In line with previous work on
semiparametric model-based clustering, the single Dirichlet distributions can
be seen as providing the baseline pattern of the data. These are then combined
to effectively model the complex distributions of the attributes for the
different classes. The model is estimated using a hierarchical two-step
procedure which combines a suitably adapted version of the
Expectation-Maximization (EM) algorithm with Bayes' rule. We validate our model
on a reliable sample of periodic variable stars available in the literature
(Dubath et al., 2011) achieving an overall classification accuracy of 71.95%, a
sensitivity of 86.11% and a specificity of 99.79% for new class detection.
| stat |
Efficient Estimation For The Cox Proportional Hazards Cure Model | While analysing time-to-event data, it is possible that a certain fraction of
subjects will never experience the event of interest and they are said to be
cured. When this feature of survival models is taken into account, the models
are commonly referred to as cure models. In the presence of covariates, the
conditional survival function of the population can be modelled by using cure
model which depends on the probability of being uncured (incidence) and the
conditional survival function of the uncured subjects (latency), and a
combination of logistic regression and Cox proportional hazards (PH) regression
is used to model the incidence and latency respectively. In this paper, we have
shown the asymptotic normality of the profile likelihood estimator via
asymptotic expansion of the profile likelihood and obtain the explicit form of
the variance estimator with an implicit function in the profile likelihood. We
have also shown the efficient score function based on projection theory and the
profile likelihood score function are equal. Our contribution in this paper is
that we have expressed the efficient information matrix as the variance of the
profile likelihood score function. A simulation study suggests that the
estimated standard errors from bootstrap samples (SMCURE package) and the
profile likelihood score function (our approach) are providing similar and
comparable results. The numerical result of our proposed method is also shown
by using the melanoma data from SMCURE R-package (Cai et al., 2012) and we
compare the results with the output obtained from SMCURE package.
| stat |
Information Assisted Dictionary Learning for fMRI data analysis | In this paper, the task-related fMRI problem is treated in its matrix
factorization formulation, focused on the Dictionary Learning (DL) approach.
The new method allows the incorporation of a priori knowledge associated both
with the experimental design as well as with available brain Atlases. Moreover,
the proposed method can efficiently cope with uncertainties related to the HRF
modeling. In addition, the proposed method bypasses one of the major drawbacks
that are associated with DL methods; that is, the selection of the
sparsity-related regularization parameters. In our formulation, an alternative
sparsity promoting constraint is employed, that bears a direct relation to the
number of voxels in the spatial maps. Hence, the related parameters can be
tuned using information that is available from brain atlases. The proposed
method is evaluated against several other popular techniques, including GLM.
The obtained performance gains are reported via a novel realistic synthetic
fMRI dataset as well as real data that are related to a challenging
experimental design.
| stat |
Oil and Gas Reservoirs Parameters Analysis Using Mixed Learning of
Bayesian Networks | In this paper, a multipurpose Bayesian-based method for data analysis, causal
inference and prediction in the sphere of oil and gas reservoir development is
considered. This allows analysing parameters of a reservoir, discovery
dependencies among parameters (including cause and effects relations), checking
for anomalies, prediction of expected values of missing parameters, looking for
the closest analogues, and much more. The method is based on extended algorithm
MixLearn@BN for structural learning of Bayesian networks. Key ideas of
MixLearn@BN are following: (1) learning the network structure on homogeneous
data subsets, (2) assigning a part of the structure by an expert, and (3)
learning the distribution parameters on mixed data (discrete and continuous).
Homogeneous data subsets are identified as various groups of reservoirs with
similar features (analogues), where similarity measure may be based on several
types of distances. The aim of the described technique of Bayesian network
learning is to improve the quality of predictions and causal inference on such
networks. Experimental studies prove that the suggested method gives a
significant advantage in missing values prediction and anomalies detection
accuracy. Moreover, the method was applied to the database of more than a
thousand petroleum reservoirs across the globe and allowed to discover novel
insights in geological parameters relationships.
| stat |
Adversarial Training Makes Weight Loss Landscape Sharper in Logistic
Regression | Adversarial training is actively studied for learning robust models against
adversarial examples. A recent study finds that adversarially trained models
degenerate generalization performance on adversarial examples when their weight
loss landscape, which is loss changes with respect to weights, is sharp.
Unfortunately, it has been experimentally shown that adversarial training
sharpens the weight loss landscape, but this phenomenon has not been
theoretically clarified. Therefore, we theoretically analyze this phenomenon in
this paper. As a first step, this paper proves that adversarial training with
the L2 norm constraints sharpens the weight loss landscape in the linear
logistic regression model. Our analysis reveals that the sharpness of the
weight loss landscape is caused by the noise aligned in the direction of
increasing the loss, which is used in adversarial training. We theoretically
and experimentally confirm that the weight loss landscape becomes sharper as
the magnitude of the noise of adversarial training increases in the linear
logistic regression model. Moreover, we experimentally confirm the same
phenomena in ResNet18 with softmax as a more general case.
| stat |
Probabilistic performance estimators for computational chemistry
methods: Systematic Improvement Probability and Ranking Probability Matrix.
I. Theory | The comparison of benchmark error sets is an essential tool for the
evaluation of theories in computational chemistry. The standard ranking of
methods by their Mean Unsigned Error is unsatisfactory for several reasons
linked to the non-normality of the error distributions and the presence of
underlying trends. Complementary statistics have recently been proposed to
palliate such deficiencies, such as quantiles of the absolute errors
distribution or the mean prediction uncertainty. We introduce here a new score,
the systematic improvement probability (SIP), based on the direct system-wise
comparison of absolute errors. Independently of the chosen scoring rule, the
uncertainty of the statistics due to the incompleteness of the benchmark data
sets is also generally overlooked. However, this uncertainty is essential to
appreciate the robustness of rankings. In the present article, we develop two
indicators based on robust statistics to address this problem: P_{inv}, the
inversion probability between two values of a statistic, and \mathbf{P}_{r},
the ranking probability matrix. We demonstrate also the essential contribution
of the correlations between error sets in these scores comparisons.
| stat |
A Multi-Quantile Regression Time Series Model with Interquantile
Lipschitz Regularization for Wind Power Probabilistic Forecasting | Modern decision-making processes require uncertainty-aware models, especially
those relying on non-symmetric costs and risk-averse profiles. The objective of
this work is to propose a dynamic model for the conditional non-parametric
distribution function (CDF) to generate probabilistic forecasts for a renewable
generation time series. To do that, we propose an adaptive non-parametric
time-series model driven by a regularized multiple-quantile-regression (MQR)
framework. In our approach, all regression models are jointly estimated through
a single linear optimization problem that finds the global-optimal parameters
in polynomial time. An innovative feature of our work is the consideration of a
Lipschitz regularization of the first derivative of coefficients in the
quantile space, which imposes coefficient smoothness. The proposed
regularization induces a coupling effect among quantiles creating a single
non-parametric CDF model with improved out-of-sample performance. A case study
with realistic wind-power generation data from the Brazilian system shows: 1)
the regularization model is capable to improve the performance of MQR
probabilistic forecasts, and 2) our MQR model outperforms five relevant
benchmarks: two based on the MQR framework, and three based on parametric
models, namely, SARIMA, and GAS with Beta and Weibull CDF.
| stat |
Inferring Heterogeneous Causal Effects in Presence of Spatial
Confounding | We address the problem of inferring the causal effect of an exposure on an
outcome across space, using observational data. The data is possibly subject to
unmeasured confounding variables which, in a standard approach, must be
adjusted for by estimating a nuisance function. Here we develop a method that
eliminates the nuisance function, while mitigating the resulting
errors-in-variables. The result is a robust and accurate inference method for
spatially varying heterogeneous causal effects. The properties of the method
are demonstrated on synthetic as well as real data from Germany and the US.
| stat |
Massively Parallel Implicit Equal-Weights Particle Filter for Ocean
Drift Trajectory Forecasting | Forecasting ocean drift trajectories are important for many applications,
including search and rescue operations, oil spill cleanup and iceberg risk
mitigation. In an operational setting, forecasts of drift trajectories are
produced based on computationally demanding forecasts of three-dimensional
ocean currents. Herein, we investigate a complementary approach for shorter
time scales by using a recent state-of-the-art implicit equal-weights particle
filter applied to a simplified ocean model. To achieve this, we present a new
algorithmic design for a data-assimilation system in which all components -
including the model, model errors, and particle filter - take advantage of
massively parallel compute architectures, such as graphical processing units.
Faster computations can enable in-situ and ad-hoc model runs for emergency
management, and larger ensembles for better uncertainty quantification. Using a
challenging test case with near-realistic chaotic instabilities, we run
data-assimilation experiments based on synthetic observations from drifting and
moored buoys, and analyse the trajectory forecasts for the drifters. Our
results show that even sparse drifter observations are sufficient to
significantly improve short-term drift forecasts up to twelve hours. With
equidistant moored buoys observing only 0.1% of the state space, the ensemble
gives an accurate description of the true state after data assimilation
followed by a high-quality probabilistic forecast.
| stat |
evgam: An R package for Generalized Additive Extreme Value Models | This article introduces the R package evgam. The package provides functions
for fitting extreme value distributions. These include the generalized extreme
value and generalized Pareto distributions. The former can also be fitted
through a point process representation. evgam supports quantile regression via
the asymmetric Laplace distribution, which can be useful for estimating high
thresholds, sometimes used to discriminate between extreme and non-extreme
values. The main addition of evgam is to let extreme value distribution
parameters have generalized additive model forms, which can be objectively
estimated using Laplace's method. Illustrative examples fitting various
distributions with various specifications are given. These include daily
precipitation accumulations for part of Colorado, US, used to illustrate
spatial models, and daily maximum temperatures for Fort Collins, Colorado, US,
used to illustrate temporal models.
| stat |
Cognitive Constructivism and the Epistemic Significance of Sharp
Statistical Hypotheses in Natural Sciences | This book presents our case in defense of a constructivist epistemological
framework and the use of compatible statistical theory and inference tools. The
basic metaphor of decision theory is the maximization of a gambler's expected
fortune, according to his own subjective utility, prior beliefs an learned
experiences. This metaphor has proven to be very useful, leading the
development of Bayesian statistics since its XX-th century revival, rooted on
the work of de Finetti, Savage and others. The basic metaphor presented in this
text, as a foundation for cognitive constructivism, is that of an
eigen-solution, and the verification of its objective epistemic status. The
FBST - Full Bayesian Significance Test - is the cornerstone of a set of
statistical tolls conceived to assess the epistemic value of such
eigen-solutions, according to their four essential attributes, namely,
sharpness, stability, separability and composability. We believe that this
alternative perspective, complementary to the one ofered by decision theory,
can provide powerful insights and make pertinent contributions in the context
of scientific research.
| stat |
Calibrated Prediction Intervals for Neural Network Regressors | Ongoing developments in neural network models are continually advancing the
state of the art in terms of system accuracy. However, the predicted labels
should not be regarded as the only core output; also important is a
well-calibrated estimate of the prediction uncertainty. Such estimates and
their calibration are critical in many practical applications. Despite their
obvious aforementioned advantage in relation to accuracy, contemporary neural
networks can, generally, be regarded as poorly calibrated and as such do not
produce reliable output probability estimates. Further, while post-processing
calibration solutions can be found in the relevant literature, these tend to be
for systems performing classification. In this regard, we herein present two
novel methods for acquiring calibrated predictions intervals for neural network
regressors: empirical calibration and temperature scaling. In experiments using
different regression tasks from the audio and computer vision domains, we find
that both our proposed methods are indeed capable of producing calibrated
prediction intervals for neural network regressors with any desired confidence
level, a finding that is consistent across all datasets and neural network
architectures we experimented with. In addition, we derive an additional
practical recommendation for producing more accurate calibrated prediction
intervals. We release the source code implementing our proposed methods for
computing calibrated predicted intervals. The code for computing calibrated
predicted intervals is publicly available.
| stat |
Derivative-based global sensitivity analysis for models with
high-dimensional inputs and functional outputs | We present a framework for derivative-based global sensitivity analysis (GSA)
for models with high-dimensional input parameters and functional outputs. We
combine ideas from derivative-based GSA, random field representation via
Karhunen--Lo\`{e}ve expansions, and adjoint-based gradient computation to
provide a scalable computational framework for computing the proposed
derivative-based GSA measures. We illustrate the strategy for a nonlinear ODE
model of cholera epidemics and for elliptic PDEs with application examples from
geosciences and biotransport.
| stat |
Decision rules for identifying combination therapies in open-entry,
randomized controlled platform trials | The design and conduct of platform trials have become increasingly popular
for drug development programs, attracting interest from statisticians,
clinicians and regulatory agencies. Many statistical questions related to
designing platform trials - such as the impact of decision rules, sharing of
information across cohorts, and allocation ratios on operating characteristics
and error rates - remain unanswered. In many platform trials, the definition of
error rates is not straightforward as classical error rate concepts are not
applicable. In particular, the strict control of the family-wise Type I error
rate often seems unreasonably rigid. For an open-entry, exploratory platform
trial design comparing combination therapies to the respective monotherapies
and standard-of-care, we define a set of error rates and operating
characteristics and then use these to compare a set of design parameters under
a range of simulation assumptions. When setting up the simulations, we aimed
for realistic trial trajectories, e.g. in case one compound is found to be
superior to standard-of-care, it could become the new standard-of-care in
future cohorts. Our results indicate that the method of data sharing, exact
specification of decision rules and quality of the biomarker used to make
interim decisions all strongly contribute to the operating characteristics of
the platform trial. Together with the potential flexibility and complexity of a
platform trial, which also impact the achieved operating characteristics, this
implies that utmost care needs to be given to evaluation of different
assumptions and design parameters at the design stage.
| stat |
Rapid evaluation of the spectral signal detection threshold and
Stieltjes transform | Accurate detection of signal components is a frequently-encountered challenge
in statistical applications with low signal-to-noise ratio. This problem is
particularly challenging in settings with heteroscedastic noise. In certain
signal-plus-noise models of data, such as the classical spiked covariance model
and its variants, there are closed formulas for the spectral signal detection
threshold (the largest sample eigenvalue attributable solely to noise) in the
isotropic noise setting. However, existing methods for numerically evaluating
the threshold for more general noise models remain unsatisfactory.
In this work, we introduce a rapid algorithm for evaluating the spectral
signal detection threshold. We consider noise matrices with a separable
variance profile, as these arise often in applications. The solution is based
on nested applications of Newton's method. We also devise a new algorithm for
evaluating the Stieltjes transform of the spectral distribution at real values
exceeding the threshold. The Stieltjes transform on this domain is known to be
a key quantity in parameter estimation for spectral denoising methods. The
correctness of both algorithms is proven from a detailed analysis of the master
equations characterizing the Stieltjes transform, and their performance is
demonstrated in numerical experiments.
| stat |
Intact-VAE: Estimating Treatment Effects under Unobserved Confounding | As an important problem of causal inference, we discuss the identification
and estimation of treatment effects under unobserved confounding. Representing
the confounder as a latent variable, we propose Intact-VAE, a new variant of
variational autoencoder (VAE), motivated by the prognostic score that is
sufficient for identifying treatment effects. We theoretically show that, under
certain settings, treatment effects are identified by our model, and further,
based on the identifiability of our model (i.e., determinacy of
representation), our VAE is a consistent estimator with representation balanced
for treatment groups. Experiments on (semi-)synthetic datasets show
state-of-the-art performance under diverse settings.
| stat |
Efficient Sampling and Structure Learning of Bayesian Networks | Bayesian networks are probabilistic graphical models widely employed to
understand dependencies in high dimensional data, and even to facilitate causal
discovery. Learning the underlying network structure, which is encoded as a
directed acyclic graph (DAG) is highly challenging mainly due to the vast
number of possible networks. Efforts have focussed on two fronts:
constraint-based methods that perform conditional independence tests to exclude
edges and score and search approaches which explore the DAG space with greedy
or MCMC schemes. Here we synthesise these two fields in a novel hybrid method
which reduces the complexity of MCMC approaches to that of a constraint-based
method. Individual steps in the MCMC scheme only require simple table lookups
so that very long chains can be efficiently obtained. Furthermore, the scheme
includes an iterative procedure to correct for errors from the conditional
independence tests. The algorithm offers markedly superior performance to
alternatives, particularly because DAGs can also be sampled from the posterior
distribution, enabling full Bayesian model averaging for much larger Bayesian
networks.
| stat |
Neural Decomposition: Functional ANOVA with Variational Autoencoders | Variational Autoencoders (VAEs) have become a popular approach for
dimensionality reduction. However, despite their ability to identify latent
low-dimensional structures embedded within high-dimensional data, these latent
representations are typically hard to interpret on their own. Due to the
black-box nature of VAEs, their utility for healthcare and genomics
applications has been limited. In this paper, we focus on characterising the
sources of variation in Conditional VAEs. Our goal is to provide a
feature-level variance decomposition, i.e. to decompose variation in the data
by separating out the marginal additive effects of latent variables z and fixed
inputs c from their non-linear interactions. We propose to achieve this through
what we call Neural Decomposition - an adaptation of the well-known concept of
functional ANOVA variance decomposition from classical statistics to deep
learning models. We show how identifiability can be achieved by training models
subject to constraints on the marginal properties of the decoder networks. We
demonstrate the utility of our Neural Decomposition on a series of synthetic
examples as well as high-dimensional genomics data.
| stat |
The Effect of Real Estate Auction Events on Mortality Rate | This study has investigated the mortality rate of parties at real estate
auctions compared to that of the overall population in South Korea by using
various variables, including age, real estate usage, cumulative number of real
estate auction events, disposal of real estate, and appraisal price. In each
case, there has been a significant difference between mortality rate of parties
at real estate auctions and that of the overall population, which provides a
new insight regarding utilization of the information on real estate auctions.
Despite the need for further detailed analysis on the correlation between real
estate auction events and death, because the result from this study is still
meaningful, the result is summarized for informational purposes.
| stat |
Variational inference for the multi-armed contextual bandit | In many biomedical, science, and engineering problems, one must sequentially
decide which action to take next so as to maximize rewards. One general class
of algorithms for optimizing interactions with the world, while simultaneously
learning how the world operates, is the multi-armed bandit setting and, in
particular, the contextual bandit case. In this setting, for each executed
action, one observes rewards that are dependent on a given 'context', available
at each interaction with the world. The Thompson sampling algorithm has
recently been shown to enjoy provable optimality properties for this set of
problems, and to perform well in real-world settings. It facilitates generative
and interpretable modeling of the problem at hand. Nevertheless, the design and
complexity of the model limit its application, since one must both sample from
the distributions modeled and calculate their expected rewards. We here show
how these limitations can be overcome using variational inference to
approximate complex models, applying to the reinforcement learning case
advances developed for the inference case in the machine learning community
over the past two decades. We consider contextual multi-armed bandit
applications where the true reward distribution is unknown and complex, which
we approximate with a mixture model whose parameters are inferred via
variational inference. We show how the proposed variational Thompson sampling
approach is accurate in approximating the true distribution, and attains
reduced regrets even with complex reward distributions. The proposed algorithm
is valuable for practical scenarios where restrictive modeling assumptions are
undesirable.
| stat |
On the Minimal Supervision for Training Any Binary Classifier from Only
Unlabeled Data | Empirical risk minimization (ERM), with proper loss function and
regularization, is the common practice of supervised classification. In this
paper, we study training arbitrary (from linear to deep) binary classifier from
only unlabeled (U) data by ERM. We prove that it is impossible to estimate the
risk of an arbitrary binary classifier in an unbiased manner given a single set
of U data, but it becomes possible given two sets of U data with different
class priors. These two facts answer a fundamental question---what the minimal
supervision is for training any binary classifier from only U data. Following
these findings, we propose an ERM-based learning method from two sets of U
data, and then prove it is consistent. Experiments demonstrate the proposed
method could train deep models and outperform state-of-the-art methods for
learning from two sets of U data.
| stat |
Longitudinal mediation analysis of time-to-event endpoints in the
presence of competing risks | This proposal is motivated by an analysis of the English Longitudinal Study
of Ageing (ELSA), which aims to investigate the role of loneliness in
explaining the negative impact of hearing loss on dementia. The methodological
challenges that complicate this mediation analysis include the use of a
time-to-event endpoint subject to competing risks, as well as the presence of
feedback relationships between the mediator and confounders that are both
repeatedly measured over time. To account for these challenges, we introduce
natural effect proportional (cause-specific) hazard models. These extend
marginal structural proportional (cause-specific) hazard models to enable
effect decomposition. We show that under certain causal assumptions, the
path-specific direct and indirect effects indexing this model are identifiable
from the observed data. We next propose an inverse probability weighting
approach to estimate these effects. On the ELSA data, this approach reveals
little evidence that the total efect of hearing loss on dementia is mediated
through the feeling of loneliness, with a non-statistically significant
indirect effect equal to 1.012 (hazard ratio (HR) scale; 95% confidence
interval (CI) 0.986 to 1.053).
| stat |
Fine-scale spatiotemporal air pollution analysis using mobile monitors
on Google Street View vehicles | People are increasingly concerned with understanding their personal
environment, including possible exposure to harmful air pollutants. In order to
make informed decisions on their day-to-day activities, they are interested in
real-time information on a localized scale. Publicly available, fine-scale,
high-quality air pollution measurements acquired using mobile monitors
represent a paradigm shift in measurement technologies. A methodological
framework utilizing these increasingly fine-scale measurements to provide
real-time air pollution maps and short-term air quality forecasts on a
fine-resolution spatial scale could prove to be instrumental in increasing
public awareness and understanding. The Google Street View study provides a
unique source of data with spatial and temporal complexities, with the
potential to provide information about commuter exposure and hot spots within
city streets with high traffic. We develop a computationally efficient
spatiotemporal model for these data and use the model to make short-term
forecasts and high-resolution maps of current air pollution levels. We also
show via an experiment that mobile networks can provide more nuanced
information than an equally-sized fixed-location network. This modeling
framework has important real-world implications in understanding citizens'
personal environments, as data production and real-time availability continue
to be driven by the ongoing development and improvement of mobile measurement
technologies.
| stat |
Combining Model and Parameter Uncertainty in Bayesian Neural Networks | Bayesian neural networks (BNNs) have recently regained a significant amount
of attention in the deep learning community due to the development of scalable
approximate Bayesian inference techniques. There are several advantages of
using Bayesian approach: Parameter and prediction uncertainty become easily
available, facilitating rigid statistical analysis. Furthermore, prior
knowledge can be incorporated. However so far there have been no scalable
techniques capable of combining both model (structural) and parameter
uncertainty. In this paper we introduce the concept of model uncertainty in
BNNs and hence make inference in the joint space of models and parameters.
Moreover, we suggest an adaptation of a scalable variational inference approach
with reparametrization of marginal inclusion probabilities to incorporate the
model space constraints. Finally, we show that incorporating model uncertainty
via Bayesian model averaging and Bayesian model selection allows to drastically
sparsify the structure of BNNs.
| stat |
Calibration methods for spatial Data | In an environmental framework, extreme values of certain spatio-temporal
processes, for example wind speeds, are the main cause of severe damage in
property, such as electrical networks, transport and agricultural
infrastructures. Therefore, availability of accurate data on such processes is
highly important in risk analysis, and in particular in producing probability
maps showing the spatial distribution of damage risks. Typically, as is the
case of wind speeds, data are available at few stations with many missing
observations and consequently simulated data are often used to augment
information, due to simulated environmental data being available at high
spatial and temporal resolutions. However, simulated data often mismatch
observed data, particularly at tails, therefore calibrating and bringing it in
line with observed data may offer practitioners more reliable and richer data
sources. Although the calibration methods that we describe in this manuscript
may equally apply to other environmental variables, we describe the methods
specifically with reference to wind data and its consequences. Since most
damages are caused by extreme winds, it is particularly important to calibrate
the right tail of simulated data based on observations. Response relationships
between the extremes of simulated and observed data are by nature highly
non-linear and non-Gaussian, therefore data fusion techniques available for
spatial data may not be adequate for this purpose. After giving a brief
description of standard calibration and data fusion methods to update simulated
data based on the observed data, we propose and describe in detail a specific
conditional quantile matching calibration method and show how our wind speed
data can be calibrated using this method.
| stat |
Interpretable Machine Learning for Privacy-Preserving Pervasive Systems | Our everyday interactions with pervasive systems generate traces that capture
various aspects of human behavior and enable machine learning algorithms to
extract latent information about users. In this paper, we propose a machine
learning interpretability framework that enables users to understand how these
generated traces violate their privacy.
| stat |
A Time To Event Framework For Multi-touch Attribution | Multi-touch attribution (MTA) estimates the relative contributions of the
multiple ads a user may see prior to any observed conversions. Increasingly,
advertisers also want to base budget and bidding decisions on these
attributions, spending more on ads that drive more conversions. We describe two
requirements for an MTA system to be suitable for this application: First, it
must be able to handle continuously updated and incomplete data. Second, it
must be sufficiently flexible to capture that an ad's effect will change over
time. We describe an MTA system, consisting of a model for user conversion
behavior and a credit assignment algorithm, that satisfies these requirements.
Our model for user conversion behavior treats conversions as occurrences in an
inhomogeneous Poisson process, while our attribution algorithm is based on
iteratively removing the last ad in the path.
| stat |
The AL-Gaussian Distribution as the Descriptive Model for the Internal
Proactive Inhibition in the Standard Stop Signal Task | Measurements of response inhibition components of reactive inhibition and
proactive inhibition within the stop signal paradigm have been of special
interest for researchers since the 1980s. While frequentist nonparametric and
Bayesian parametric methods have been proposed to precisely estimate the entire
distribution of reactive inhibition, quantified by stop signal reaction
times(SSRT), there is no method yet in the stop-signal task literature to
precisely estimate the entire distribution of proactive inhibition. We
introduce an Asymmetric Laplace Gaussian (ALG) model to describe the
distribution of proactive inhibition. The proposed method is based on two
assumptions of independent trial type(go/stop) reaction times, and Ex-Gaussian
(ExG) models for them. Results indicated that the four parametric, ALG model
uniquely describes the proactive inhibition distribution and its key shape
features; and, its hazard function is monotonically increasing as are its three
parametric ExG components. In conclusion, both response inhibition components
can be uniquely modeled via variations of the four parametric ALG model
described with their associated similar distributional features.
| stat |
Clustering high dimensional meteorological scenarios: results and
performance index | The Reseau de Transport d'Electricit\'e (RTE) is the French main electricity
network operational manager and dedicates large number of resources and efforts
towards understanding climate time series data. We discuss here the problem and
the methodology of grouping and selecting representatives of possible climate
scenarios among a large number of climate simulations provided by RTE. The data
used is composed of temperature times series for 200 different possible
scenarios on a grid of geographical locations in France. These should be
clustered in order to detect common patterns regarding temperatures curves and
help to choose representative scenarios for network simulations, which in turn
can be used for energy optimisation. We first show that the choice of the
distance used for the clustering has a strong impact on the meaning of the
results: depending on the type of distance used, either spatial or temporal
patterns prevail. Then we discuss the difficulty of fine-tuning the distance
choice (combined with a dimension reduction procedure) and we propose a
methodology based on a carefully designed index.
| stat |
Nearest-Neighbor Neural Networks for Geostatistics | Kriging is the predominant method used for spatial prediction, but relies on
the assumption that predictions are linear combinations of the observations.
Kriging often also relies on additional assumptions such as normality and
stationarity. We propose a more flexible spatial prediction method based on the
Nearest-Neighbor Neural Network (4N) process that embeds deep learning into a
geostatistical model. We show that the 4N process is a valid stochastic process
and propose a series of new ways to construct features to be used as inputs to
the deep learning model based on neighboring information. Our model framework
outperforms some existing state-of-art geostatistical modelling methods for
simulated non-Gaussian data and is applied to a massive forestry dataset.
| stat |
Statistical learning and cross-validation for point processes | This paper presents the first general (supervised) statistical learning
framework for point processes in general spaces. Our approach is based on the
combination of two new concepts, which we define in the paper: i) bivariate
innovations, which are measures of discrepancy/prediction-accuracy between two
point processes, and ii) point process cross-validation (CV), which we here
define through point process thinning. The general idea is to carry out the
fitting by predicting CV-generated validation sets using the corresponding
training sets; the prediction error, which we minimise, is measured by means of
bivariate innovations. Having established various theoretical properties of our
bivariate innovations, we study in detail the case where the CV procedure is
obtained through independent thinning and we apply our statistical learning
methodology to three typical spatial statistical settings, namely parametric
intensity estimation, non-parametric intensity estimation and Papangelou
conditional intensity fitting. Aside from deriving theoretical properties
related to these cases, in each of them we numerically show that our
statistical learning approach outperforms the state of the art in terms of mean
(integrated) squared error.
| stat |
ssMousetrack: Analysing computerized tracking data via Bayesian
state-space models in {R} | Recent technological advances have provided new settings to enhance
individual-based data collection and computerized-tracking data have became
common in many behavioral and social research. By adopting instantaneous
tracking devices such as computer-mouse, wii, and joysticks, such data provide
new insights for analysing the dynamic unfolding of response process.
ssMousetrack is a R package for modeling and analysing computerized-tracking
data by means of a Bayesian state-space approach. The package provides a set of
functions to prepare data, fit the model, and assess results via simple
diagnostic checks. This paper describes the package and illustrates how it can
be used to model and analyse computerized-tracking data. A case study is also
included to show the use of the package in empirical case studies.
| stat |
Continuous-time state-space modelling of the hot hand in basketball | We investigate the hot hand phenomenon using data on 110,513 free throws
taken in the National Basketball Association (NBA). As free throws occur at
unevenly spaced time points within a game, we consider a state-space model
formulated in continuous time to investigate serial dependence in players'
success probabilities. In particular, the underlying state process can be
interpreted as a player's (latent) varying form and is modelled using the
Ornstein-Uhlenbeck process. Our results support the existence of the hot hand,
but the magnitude of the estimated effect is rather small.
| stat |
Optimizing and accelerating space-time Ripley's K function based on
Apache Spark for distributed spatiotemporal point pattern analysis | With increasing point of interest (POI) datasets available with fine-grained
spatial and temporal attributes, space-time Ripley's K function has been
regarded as a powerful approach to analyze spatiotemporal point process.
However, space-time Ripley's K function is computationally intensive for
point-wise distance comparisons, edge correction and simulations for
significance testing. Parallel computing technologies like OpenMP, MPI and CUDA
have been leveraged to accelerate the K function, and related experiments have
demonstrated the substantial acceleration. Nevertheless, previous works have
not extended optimization of Ripley's K function from space dimension to
space-time dimension. Without sophisticated spatiotemporal query and
partitioning mechanisms, extra computational overhead can be problematic.
Meanwhile, these researches were limited by the restricted scalability and
relative expensive programming cost of parallel frameworks and impeded their
applications for large POI dataset and Ripley's K function variations. This
paper presents a distributed computing method to accelerate space-time Ripley's
K function upon state-of-the-art distributed computing framework Apache Spark,
and four strategies are adopted to simplify calculation procedures and
accelerate distributed computing respectively. Based on the optimized method, a
web-based visual analytics framework prototype has been developed. Experiments
prove the feasibility and time efficiency of the proposed method, and also
demonstrate its value on promoting applications of space-time Ripley's K
function in ecology, geography, sociology, economics, urban transportation and
other fields.
| stat |
Output-weighted optimal sampling for Bayesian regression and rare event
statistics using few samples | For many important problems the quantity of interest is an unknown function
of the parameters, which is a random vector with known statistics. Since the
dependence of the output on this random vector is unknown, the challenge is to
identify its statistics, using the minimum number of function evaluations. This
problem can been seen in the context of active learning or optimal experimental
design. We employ Bayesian regression to represent the derived model
uncertainty due to finite and small number of input-output pairs. In this
context we evaluate existing methods for optimal sample selection, such as
model error minimization and mutual information maximization. We show that for
the case of known output variance, the commonly employed criteria in the
literature do not take into account the output values of the existing
input-output pairs, while for the case of unknown output variance this
dependence can be very weak. We introduce a criterion that takes into account
the values of the output for the existing samples and adaptively selects inputs
from regions of the parameter space which have important contribution to the
output. The new method allows for application to high-dimensional inputs,
paving the way for optimal experimental design in high-dimensions.
| stat |
Filtering ASVs/OTUs via Mutual Information-Based Microbiome Network
Analysis | Microbial communities are widely studied using high-throughput sequencing
techniques, such as 16S rRNA gene sequencing. These techniques have attracted
biologists as they offer powerful tools to explore microbial communities and
investigate their patterns of diversity in biological and biomedical samples at
remarkable resolution. However, the accuracy of these methods can negatively
affected by the presence of contamination. Several studies have recognized that
contamination is a common problem in microbial studies and have offered
promising computational and laboratory-based approaches to assess and remove
contaminants. Here we propose a novel strategy, MI-based (mutual information
based) filtering method, which uses information theoretic functionals and graph
theory to identify and remove contaminants. We applied MI-based filtering
method to a mock community data set and evaluated the amount of information
loss due to filtering taxa. We also compared our method to commonly practice
traditional filtering methods. In a mock community data set, MI-based filtering
approach maintained the true bacteria in the community without significant loss
of information. Our results indicate that MI-based filtering method effectively
identifies and removes contaminants in microbial communities and hence it can
be beneficial as a filtering method to microbiome studies. We believe our
filtering method has two advantages over traditional filtering methods. First,
it does not required an arbitrary choice of threshold and second, it is able to
detect true taxa with low abundance.
| stat |
A generalized information criterion for high-dimensional PCA rank
selection | Principal component analysis (PCA) is the most commonly used statistical
procedure for dimension reduction. An important issue for applying PCA is to
determine the rank, which is the number of dominant eigenvalues of the
covariance matrix. The Akaike information criterion (AIC) and Bayesian
information criterion (BIC) are among the most widely used rank selection
methods. Both use the number of free parameters for assessing model complexity.
In this work, we adopt the generalized information criterion (GIC) to propose a
new method for PCA rank selection under the high-dimensional framework. The GIC
model complexity takes into account the sizes of covariance eigenvalues and can
be better adaptive to practical applications. Asymptotic properties of GIC are
derived and the selection consistency is established under the generalized
spiked covariance model.
| stat |
Change point localization in dependent dynamic nonparametric random dot
product graphs | In this paper, we study the change point localization problem in a sequence
of dependent nonparametric random dot product graphs. To be specific, assume
that at every time point, a network is generated from a nonparametric random
dot product graph model (see e.g. Athreya et al., 2017), where the latent
positions are generated from unknown underlying distributions. The underlying
distributions are piecewise constant in time and change at unknown locations,
called change points. Most importantly, we allow for dependence among networks
generated between two consecutive change points. This setting incorporates
edge-dependence within networks and temporal dependence between networks, which
is the most flexible setting in the published literature.
To accomplish the task of consistently localizing change points, we propose a
novel change point detection algorithm, consisting of two steps. First, we
estimate the latent positions of the random dot product model, our theoretical
result being a refined version of the state-of-the-art results, allowing the
dimension of the latent positions to grow unbounded. Subsequently, we construct
a nonparametric version of the CUSUM statistic (Page, 1954, Padilla et al.,
2019) that allows for temporal dependence. Consistent localization is proved
theoretically and supported by extensive numerical experiments, which
illustrate state-of-the-art performance. We also provide in depth discussion of
possible extensions to give more understanding and insights.
| stat |
Reconstructing Sparse Signals via Greedy Monte-Carlo Search | We propose a Monte-Carlo-based method for reconstructing sparse signals in
the formulation of sparse linear regression in a high-dimensional setting. The
basic idea of this algorithm is to explicitly select variables or covariates to
represent a given data vector or responses and accept randomly generated
updates of that selection if and only if the energy or cost function decreases.
This algorithm is called the greedy Monte-Carlo (GMC) search algorithm. Its
performance is examined via numerical experiments, which suggests that in the
noiseless case, GMC can achieve perfect reconstruction in undersampling
situations of a reasonable level: it can outperform the $\ell_1$ relaxation but
does not reach the algorithmic limit of MC-based methods theoretically
clarified by an earlier analysis. The necessary computational time is also
examined and compared with that of an algorithm using simulated annealing.
Additionally, experiments on the noisy case are conducted on synthetic datasets
and on a real-world dataset, supporting the practicality of GMC.
| stat |
Labeled Optimal Partitioning | In data sequences measured over space or time, an important problem is
accurate detection of abrupt changes. In partially labeled data, it is
important to correctly predict presence/absence of changes in positive/negative
labeled regions, in both the train and test sets. One existing dynamic
programming algorithm is designed for prediction in unlabeled test regions (and
ignores the labels in the train set); another is for accurate fitting of train
labels (but does not predict changepoints in unlabeled test regions). We
resolve these issues by proposing a new optimal changepoint detection model
that is guaranteed to fit the labels in the train data, and can also provide
predictions of unlabeled changepoints in test data. We propose a new dynamic
programming algorithm, Labeled Optimal Partitioning (LOPART), and we provide a
formal proof that it solves the resulting non-convex optimization problem. We
provide theoretical and empirical analysis of the time complexity of our
algorithm, in terms of the number of labels and the size of the data sequence
to segment. Finally, we provide empirical evidence that our algorithm is more
accurate than the existing baselines, in terms of train and test label error.
| stat |
Buffered environmental contours | The main idea of this paper is to use the notion of buffered failure
probability from probabilistic structural design, to introduce buffered
environmental contours. Classical environmental contours are used in structural
design in order to obtain upper bounds on the failure probabilities of a large
class of designs. The purpose of buffered failure probabilities is the same.
However, in constrast to classical environmental contours, this new concept
does not just take into account failure vs. functioning, but also to which
extent the system is failing. For example, this is relevant when considering
the risk of flooding: We are not just interested in knowing whether a river has
flooded. The damages caused by the flooding greatly depends on how much the
water has risen above the standard level.
| stat |
Semi-intrusive uncertainty propagation for multiscale models | A family of semi-intrusive uncertainty propagation (UP) methods for
multiscale models is introduced. The methods are semi-intrusive in the sense
that inspection of the model is limited up to the level of the single scale
systems, and viewing these single scale components as black-boxes. The goal is
to estimate uncertainty in the result of multiscale models at a reduced amount
of time as compared to black-box Monte Carlo (MC). In the resulting
semi-intrusive MC method, the required number of samples of an expensive single
scale model is minimized in order to reduce the execution time for the overall
UP. In the metamodeling approach, the expensive model component is replaced
completely by a computationally much cheaper surrogate model. These
semi-intrusive algorithms have been tested on two case studies based on
reaction-diffusion dynamics. The results demonstrate that the proposed
semi-intrusive methods can result in a significant reduction of the
computational time for multiscale UP, while still computing accurately the
estimates of uncertainties. The semi-intrusive methods can, therefore, be a
valid alternative, when uncertainties of a multiscale model cannot be estimated
by the black-box MC methods in a feasible amount of time.
| stat |
Kernel Change-point Detection with Auxiliary Deep Generative Models | Detecting the emergence of abrupt property changes in time series is a
challenging problem. Kernel two-sample test has been studied for this task
which makes fewer assumptions on the distributions than traditional parametric
approaches. However, selecting kernels is non-trivial in practice. Although
kernel selection for two-sample test has been studied, the insufficient samples
in change point detection problem hinder the success of those developed kernel
selection algorithms. In this paper, we propose KL-CPD, a novel kernel learning
framework for time series CPD that optimizes a lower bound of test power via an
auxiliary generative model. With deep kernel parameterization, KL-CPD endows
kernel two-sample test with the data-driven kernel to detect different types of
change-points in real-world applications. The proposed approach significantly
outperformed other state-of-the-art methods in our comparative evaluation of
benchmark datasets and simulation studies.
| stat |
Exploratory Data Analysis for Airline Disruption Management | Reliable platforms for data collation during airline schedule operations have
significantly increased the quality and quantity of available information for
effectively managing airline schedule disruptions. To that effect, this paper
applies macroscopic and microscopic techniques by way of basic statistics and
machine learning, respectively, to analyze historical scheduling and operations
data from a major airline in the United States. Macroscopic results reveal that
majority of irregular operations in airline schedule that occurred over a
one-year period stemmed from disruptions due to flight delays, while
microscopic results validate different modeling assumptions about key drivers
for airline disruption management like turnaround as a Gaussian process.
| stat |
A new measure for the analysis of epidemiological associations: Cannabis
use disorder examples | Analyses of population-based surveys are instrumental to research on
prevention and treatment of mental and substance use disorders.
Population-based data provides descriptive characteristics of multiple
determinants of public health and are typically available to researchers as an
annual data release. To provide trends in national estimates or to update the
existing ones, a meta-analytical approach to year-by-year data is typically
employed with ORs as effect sizes. However, if the estimated ORs exhibit
different patterns over time, some normalization of ORs may be warranted. We
propose a new normalized measure of effect size and derive an asymptotic
distribution for the respective test statistic. The normalization constant is
based on the maximum range of the standardized log(OR), for which we establish
a connection to the Laplace Limit Constant. Furthermore, we propose to employ
standardized log(OR) in a novel way to obtain accurate posterior inference.
Through simulation studies, we show that our new statistic is more powerful
than the traditional one for testing the hypothesis OR=1. We then applied it to
the United States population estimates of co-occurrence of side effect
problem-experiences (SEPE) among newly incident cannabis users, based on the
the National Survey on Drug Use and Health (NSDUH), 2004-2014.
| stat |
Deep Curvature Suite | We present MLRG Deep Curvature suite, a PyTorch-based, open-source package
for analysis and visualisation of neural network curvature and loss landscape.
Despite of providing rich information into properties of neural network and
useful for a various designed tasks, curvature information is still not made
sufficient use for various reasons, and our method aims to bridge this gap. We
present a primer, including its main practical desiderata and common
misconceptions, of \textit{Lanczos algorithm}, the theoretical backbone of our
package, and present a series of examples based on synthetic toy examples and
realistic modern neural networks tested on CIFAR datasets, and show the
superiority of our package against existing competing approaches for the
similar purposes.
| stat |
Rank-one Convexification for Sparse Regression | Sparse regression models are increasingly prevalent due to their ease of
interpretability and superior out-of-sample performance. However, the exact
model of sparse regression with an $\ell_0$ constraint restricting the support
of the estimators is a challenging (\NP-hard) non-convex optimization problem.
In this paper, we derive new strong convex relaxations for sparse regression.
These relaxations are based on the ideal (convex-hull) formulations for
rank-one quadratic terms with indicator variables. The new relaxations can be
formulated as semidefinite optimization problems in an extended space and are
stronger and more general than the state-of-the-art formulations, including the
perspective reformulation and formulations with the reverse Huber penalty and
the minimax concave penalty functions. Furthermore, the proposed rank-one
strengthening can be interpreted as a \textit{non-separable, non-convex,
unbiased} sparsity-inducing regularizer, which dynamically adjusts its penalty
according to the shape of the error function without inducing bias for the
sparse solutions. In our computational experiments with benchmark datasets, the
proposed conic formulations are solved within seconds and result in
near-optimal solutions (with 0.4\% optimality gap) for non-convex
$\ell_0$-problems. Moreover, the resulting estimators also outperform
alternative convex approaches from a statistical perspective, achieving high
prediction accuracy and good interpretability.
| stat |
Generalizing Convolutional Neural Networks for Equivariance to Lie
Groups on Arbitrary Continuous Data | The translation equivariance of convolutional layers enables convolutional
neural networks to generalize well on image problems. While translation
equivariance provides a powerful inductive bias for images, we often
additionally desire equivariance to other transformations, such as rotations,
especially for non-image data. We propose a general method to construct a
convolutional layer that is equivariant to transformations from any specified
Lie group with a surjective exponential map. Incorporating equivariance to a
new group requires implementing only the group exponential and logarithm maps,
enabling rapid prototyping. Showcasing the simplicity and generality of our
method, we apply the same model architecture to images, ball-and-stick
molecular data, and Hamiltonian dynamical systems. For Hamiltonian systems, the
equivariance of our models is especially impactful, leading to exact
conservation of linear and angular momentum.
| stat |
Improve Unsupervised Domain Adaptation with Mixup Training | Unsupervised domain adaptation studies the problem of utilizing a relevant
source domain with abundant labels to build predictive modeling for an
unannotated target domain. Recent work observe that the popular adversarial
approach of learning domain-invariant features is insufficient to achieve
desirable target domain performance and thus introduce additional training
constraints, e.g. cluster assumption. However, these approaches impose the
constraints on source and target domains individually, ignoring the important
interplay between them. In this work, we propose to enforce training
constraints across domains using mixup formulation to directly address the
generalization performance for target data. In order to tackle potentially huge
domain discrepancy, we further propose a feature-level consistency regularizer
to facilitate the inter-domain constraint. When adding intra-domain mixup and
domain adversarial learning, our general framework significantly improves
state-of-the-art performance on several important tasks from both image
classification and human activity recognition.
| stat |
Kernels for time series with irregularly-spaced multivariate
observations | Time series are an interesting frontier for kernel-based methods, for the
simple reason that there is no kernel designed to represent them and their
unique characteristics in full generality. Existing sequential kernels ignore
the time indices, with many assuming that the series must be regularly-spaced;
some such kernels are not even psd. In this manuscript, we show that a "series
kernel" that is general enough to represent irregularly-spaced multivariate
time series may be built out of well-known "vector kernels". We also show that
all series kernels constructed using our methodology are psd, and are thus
widely applicable. We demonstrate this point by formulating a Gaussian
process-based strategy - with our series kernel at its heart - to make
predictions about test series when given a training set. We validate the
strategy experimentally by estimating its generalisation error on multiple
datasets and comparing it to relevant baselines. We also demonstrate that our
series kernel may be used for the more traditional setting of time series
classification, where its performance is broadly in line with alternative
methods.
| stat |
Systematic statistical analysis of microbial data from dilution series | In microbial studies, samples are often treated under different experimental
conditions and then tested for microbial survival. A technique, dating back to
the 1880's, consists of diluting the samples several times and incubating each
dilution to verify the existence of microbial Colony Forming Units or CFU's,
seen by the naked eye. The main problem in the dilution series data analysis is
the uncertainty quantification of the simple point estimate of the original
number of CFU's in the sample (i.e., at dilution zero). Common approaches such
as log-normal or Poisson models do not seem to handle well extreme cases with
low or high counts, among other issues. We build a novel binomial model, based
on the actual design of the experimental procedure including the dilution
series. For repetitions we construct a hierarchical model for experimental
results from a single lab and in turn a higher hierarchy for inter-lab
analyses. Results seem promising, with a systematic treatment of all data
cases, including zeros, censored data, repetitions, intra and inter-laboratory
studies. Using a Bayesian approach, a robust and efficient MCMC method is used
to analyze several real data sets.
| stat |
A Closed-Form Filter for Binary Time Series | Non-Gaussian state-space models arise in several applications. Within this
framework, the binary time series setting is a source of constant interest due
to its relevance in many studies. However, unlike Gaussian state-space models,
where filtering, predictive and smoothing distributions are available in
closed-form, binary state-space models require approximations or sequential
Monte Carlo strategies for inference and prediction. This is due to the
apparent absence of conjugacy between the Gaussian states and the likelihood
induced by the observation equation for the binary data. In this article we
prove that the filtering, predictive and smoothing distributions in dynamic
probit models with Gaussian state variables are, in fact, available and belong
to a class of unified skew-normals (SUN) whose parameters can be updated
recursively in time via analytical expressions. Also the functionals of these
distributions depend on known functions, but their calculation requires
intractable numerical integration. Leveraging the SUN properties, we address
this point via new Monte Carlo methods based on independent and identically
distributed samples from the smoothing distribution, which can naturally be
adapted to the filtering and predictive case, thereby improving
state-of-the-art approximate or sequential Monte Carlo inference in
small-to-moderate dimensional studies. A scalable and optimal particle filter
which exploits the SUN properties is also developed to deal with online
inference in high dimensions. Performance gains over competitors are outlined
in a real-data financial application.
| stat |
Inference and Sampling of $K_{33}$-free Ising Models | We call an Ising model tractable when it is possible to compute its partition
function value (statistical inference) in polynomial time. The tractability
also implies an ability to sample configurations of this model in polynomial
time. The notion of tractability extends the basic case of planar zero-field
Ising models. Our starting point is to describe algorithms for the basic case
computing partition function and sampling efficiently. To derive the
algorithms, we use an equivalent linear transition to perfect matching counting
and sampling on an expanded dual graph. Then, we extend our tractable inference
and sampling algorithms to models, whose triconnected components are either
planar or graphs of $O(1)$ size. In particular, it results in a polynomial-time
inference and sampling algorithms for $K_{33}$ (minor) free topologies of
zero-field Ising models - a generalization of planar graphs with a potentially
unbounded genus.
| stat |
Comment on "Hidden truncation hyperbolic distributions, finite mixtures
thereof and their application for clustering" Murray, Browne, and \McNicholas | We comment on the paper of Murray, Browne, and McNicholas (2017), who
proposed mixtures of skew distributions, which they termed hidden truncation
hyperbolic (HTH). They recently made a clarification (Murray, Browne,
McNicholas, 2019) concerning their claim that the so-called CFUST distribution
is a special case of the HTH distribution. There are also some other matters in
the original version of the paper that were in need of clarification as
discussed here.
| stat |
Inference on extremal dependence in the domain of attraction of a
structured H\"usler-Reiss distribution motivated by a Markov tree with latent
variables | A Markov tree is a probabilistic graphical model for a random vector indexed
by the nodes of an undirected tree encoding conditional independence relations
between variables. One possible limit distribution of partial maxima of samples
from such a Markov tree is a max-stable H\"usler-Reiss distribution whose
parameter matrix inherits its structure from the tree, each edge contributing
one free dependence parameter. Our central assumption is that, upon marginal
standardization, the data-generating distribution is in the max-domain of
attraction of the said H\"usler-Reiss distribution, an assumption much weaker
than the one that data are generated according to a graphical model. Even if
some of the variables are unobservable (latent), we show that the underlying
model parameters are still identifiable if and only if every node corresponding
to a latent variable has degree at least three. Three estimation procedures,
based on the method of moments, maximum composite likelihood, and pairwise
extremal coefficients, are proposed for usage on multivariate peaks over
thresholds data when some variables are latent. A typical application is a
river network in the form of a tree where, on some locations, no data are
available. We illustrate the model and the identifiability criterion on a data
set of high water levels on the Seine, France, with two latent variables. The
structured H\"usler-Reiss distribution is found to fit the observed extremal
dependence patterns well. The parameters being identifiable we are able to
quantify tail dependence between locations for which there are no data.
| stat |
Early Detection of Influenza outbreaks in the United States | Public health surveillance systems often fail to detect emerging infectious
diseases, particularly in resource limited settings. By integrating relevant
clinical and internet-source data, we can close critical gaps in coverage and
accelerate outbreak detection. Here, we present a multivariate algorithm that
uses freely available online data to provide early warning of emerging
influenza epidemics in the US. We evaluated 240 candidate predictors and found
that the most predictive combination does \textit{not} include surveillance or
electronic health records data, but instead consists of eight Google search and
Wikipedia pageview time series reflecting changing levels of interest in
influenza-related topics. In cross validation on 2010-2016 data, this algorithm
sounds alarms an average of 16.4 weeks prior to influenza activity reaching the
Center for Disease Control and Prevention (CDC) threshold for declaring the
start of the season. In an out-of-sample test on data from the rapidly-emerging
fall wave of the 2009 H1N1 pandemic, it recognized the threat five weeks in
advance of this surveillance threshold. Simpler algorithms, including fixed
week-of-the-year triggers, lag the optimized alarms by only a few weeks when
detecting seasonal influenza, but fail to provide early warning in the 2009
pandemic scenario. This demonstrates a robust method for designing next
generation outbreak detection algorithms. By combining scan statistics with
machine learning, it identifies tractable combinations of data sources (from
among thousands of candidates) that can provide early warning of emerging
infectious disease threats worldwide.
| stat |
Model-data-driven constitutive responses: application to a multiscale
computational framework | Computational multiscale methods for analyzing and deriving constitutive
responses have been used as a tool in engineering problems because of their
ability to combine information at different length scales. However, their
application in a nonlinear framework can be limited by high computational
costs, numerical difficulties, and/or inaccuracies. In this paper, a hybrid
methodology is presented which combines classical constitutive laws
(model-based), a data-driven correction component, and computational multiscale
approaches. A model-based material representation is locally improved with data
from lower scales obtained by means of a nonlinear numerical homogenization
procedure leading to a model-data-driven approach. Therefore, macroscale
simulations explicitly incorporate the true microscale response, maintaining
the same level of accuracy that would be obtained with online micro-macro
simulations but with a computational cost comparable to classical model-driven
approaches. In the proposed approach, both model and data play a fundamental
role allowing for the synergistic integration between a physics-based response
and a machine learning black-box. Numerical applications are implemented in two
dimensions for different tests investigating both material and structural
responses in large deformation.
| stat |
Fair Marriage Principle and Initialization Map for the EM Algorithm | The popular convergence theory of the EM algorithm explains that the observed
incomplete data log-likelihood L and the complete data log-likelihood Q are
positively correlated, and we can maximize L by maximizing Q. The Deterministic
Annealing EM (DAEM) algorithm was hence proposed for avoiding locally maximal
Q. This paper provides different conclusions: 1) The popular convergence theory
is wrong; 2) The locally maximal Q can affect the convergent speed, but cannot
block the global convergence; 3) Like marriage competition, unfair competition
between two components may vastly decrease the globally convergent speed; 4)
Local convergence exists because the sample is too small, and unfair
competition exists; 5) An improved EM algorithm, called the Channel Matching
(CM) EM algorithm, can accelerate the global convergence. This paper provides
an initialization map with two means as two axes for the example of a binary
Gaussian mixture studied by the authors of DAEM algorithm. This map can tell
how fast the convergent speeds are for different initial means and why points
in some areas are not suitable as initial points. A two-dimensional example
indicates that the big sample or the fair initialization can avoid global
convergence. For more complicated mixture models, we need further study to
convert the fair marriage principle to specific methods for the
initializations.
| stat |
Kriging in Tensor Train data format | Combination of low-tensor rank techniques and the Fast Fourier transform
(FFT) based methods had turned out to be prominent in accelerating various
statistical operations such as Kriging, computing conditional covariance,
geostatistical optimal design, and others. However, the approximation of a full
tensor by its low-rank format can be computationally formidable. In this work,
we incorporate the robust Tensor Train (TT) approximation of covariance
matrices and the efficient TT-Cross algorithm into the FFT-based Kriging. It is
shown that here the computational complexity of Kriging is reduced to
$\mathcal{O}(d r^3 n)$, where $n$ is the mode size of the estimation grid, $d$
is the number of variables (the dimension), and $r$ is the rank of the TT
approximation of the covariance matrix. For many popular covariance functions
the TT rank $r$ remains stable for increasing $n$ and $d$. The advantages of
this approach against those using plain FFT are demonstrated in synthetic and
real data examples.
| stat |
Fast calculation of the variance of edge crossings | The crossing number, i.e. the minimum number of edge crossings arising when
drawing a graph on a certain surface, is a very important problem of graph
theory. The opposite problem, i.e. the maximum crossing number, is receiving
growing attention. Here we consider a complementary problem of the distribution
of the number of edge crossings, namely the variance of the number of
crossings, when embedding the vertices of an arbitrary graph in some space at
random. In his pioneering research, Moon derived that variance on random linear
arrangements of complete unipartite and bipartite graphs. Given the need of
efficient algorithms to support this sort of research and given also the
growing interest of the number of edge crossings in spatial networks, networks
where vertices are embedded in some space, here we derive algorithms to
calculate the variance in arbitrary graphs in $o(nm^2)$-time, and in forests in
$O(n)$-time. These algorithms work on a wide range of random layouts (not only
on Moon's) and are based on novel arithmetic expressions for the calculation of
the variance that we develop from previous theoretical work. This paves the way
for many applications that rely on a fast but exact calculation of the
variance.
| stat |
Airflow recovery from thoracic and abdominal movements using
Synchrosqueezing Transform and Locally Stationary Gaussian Process Regression | Airflow signal encodes rich information about respiratory system. While the
gold standard for measuring airflow is to use a spirometer with an occlusive
seal, this is not practical for ambulatory monitoring of patients. Advances in
sensor technology have made measurement of motion of the thorax and abdomen
feasible with small inexpensive devices, but estimation of airflow from these
time series is challenging. We propose to use the nonlinear-type time-frequency
analysis tool, synchrosqueezing transform, to properly represent the thoracic
and abdominal movement signals as the features, which are used to recover the
airflow by the locally stationary Gaussian process. We show that, using a
dataset that contains respiratory signals under normal sleep conditions, an
accurate prediction can be achieved by fitting the proposed model in the
feature space both in the intra- and inter-subject setups. We also apply our
method to a more challenging case, where subjects under general anesthesia
underwent transitions from pressure support to unassisted ventilation to
further demonstrate the utility of the proposed method.
| stat |
Ensemble Rejection Sampling | We introduce Ensemble Rejection Sampling, a scheme for exact simulation from
the posterior distribution of the latent states of a class of non-linear
non-Gaussian state-space models. Ensemble Rejection Sampling relies on a
proposal for the high-dimensional state sequence built using ensembles of state
samples. Although this algorithm can be interpreted as a rejection sampling
scheme acting on an extended space, we show under regularity conditions that
the expected computational cost to obtain an exact sample increases cubically
with the length of the state sequence instead of exponentially for standard
rejection sampling. We demonstrate this methodology by sampling exactly state
sequences according to the posterior distribution of a stochastic volatility
model and a non-linear autoregressive process. We also present an application
to rare event simulation.
| stat |
Skew selection for factor stochastic volatility models | This paper proposes factor stochastic volatility models with skew error
distributions. The generalized hyperbolic skew t-distribution is employed for
common-factor processes and idiosyncratic shocks. Using a Bayesian sparsity
modeling strategy for the skewness parameter provides a parsimonious skew
structure for possibly high-dimensional stochastic volatility models. Analyses
of daily stock returns are provided. Empirical results show that the skewness
is important for common-factor processes but less for idiosyncratic shocks. The
sparse skew structure improves prediction and portfolio performance.
| stat |
COVID-19 case data for Italy stratified by age class | The dataset described in this paper contains daily data about COVID-19 cases
that occurred in Italy over the period from Jan. 28, 2020 to March 20, 2021,
divided into ten age classes of the population, the first class being 0-9
years, the tenth class being 90 years and over. The dataset contains eight
columns, namely: date (day), age class, number of new cases, number of newly
hospitalized patients, number of patients entering intensive care, number of
deceased patients, number of recovered patients, number of active infected
patients. This data has been officially released for research purposes by the
Italian authority for COVID-19 epidemiologic surveillance (Istituto Superiore
di Sanit\`a - ISS), upon formal request by the authors, in accordance with the
Ordonnance of the Chief of the Civil Protection Department n. 691 dated Aug. 4
2020. A separate file contains the numerosity of the population in each age
class, according to the National Institute of Statistics (ISTAT) data of the
resident population of Italy as of Jan. 2020. This data has potential use, for
instance, in epidemiologic studies of the effects of the COVID-19 contagion in
Italy, in mortality analysis by age class, and in the development and testing
of dynamical models of the contagion.
| stat |
Invariant-equivariant representation learning for multi-class data | Representations learnt through deep neural networks tend to be highly
informative, but opaque in terms of what information they learn to encode. We
introduce an approach to probabilistic modelling that learns to represent data
with two separate deep representations: an invariant representation that
encodes the information of the class from which the data belongs, and an
equivariant representation that encodes the symmetry transformation defining
the particular data point within the class manifold (equivariant in the sense
that the representation varies naturally with symmetry transformations). This
approach is based primarily on the strategic routing of data through the two
latent variables, and thus is conceptually transparent, easy to implement, and
in-principle generally applicable to any data comprised of discrete classes of
continuous distributions (e.g. objects in images, topics in language,
individuals in behavioural data). We demonstrate qualitatively compelling
representation learning and competitive quantitative performance, in both
supervised and semi-supervised settings, versus comparable modelling approaches
in the literature with little fine tuning.
| stat |
Unbiased Subdata Selection for Fair Classification: A Unified Framework
and Scalable Algorithms | As an important problem in modern data analytics, classification has
witnessed varieties of applications from different domains. Different from
conventional classification approaches, fair classification concerns the issues
of unintentional biases against the sensitive features (e.g., gender, race).
Due to high nonconvexity of fairness measures, existing methods are often
unable to model exact fairness, which can cause inferior fair classification
outcomes. This paper fills the gap by developing a novel unified framework to
jointly optimize accuracy and fairness. The proposed framework is versatile and
can incorporate different fairness measures studied in literature precisely as
well as can be applicable to many classifiers including deep classification
models. Specifically, in this paper, we first prove Fisher consistency of the
proposed framework. We then show that many classification models within this
framework can be recast as mixed-integer convex programs, which can be solved
effectively by off-the-shelf solvers when the instance sizes are moderate and
can be used as benchmarks to compare the efficiency of approximation
algorithms. We prove that in the proposed framework, when the classification
outcomes are known, the resulting problem, termed "unbiased subdata selection,"
is strongly polynomial-solvable and can be used to enhance the classification
fairness by selecting more representative data points. This motivates us to
develop an iterative refining strategy (IRS) to solve the large-scale
instances, where we improve the classification accuracy and conduct the
unbiased subdata selection in an alternating fashion. We study the convergence
property of IRS and derive its approximation bound. More broadly, this
framework can be leveraged to improve classification models with unbalanced
data by taking F1 score into consideration.
| stat |
Network classification with applications to brain connectomics | While statistical analysis of a single network has received a lot of
attention in recent years, with a focus on social networks, analysis of a
sample of networks presents its own challenges which require a different set of
analytic tools. Here we study the problem of classification of networks with
labeled nodes, motivated by applications in neuroimaging. Brain networks are
constructed from imaging data to represent functional connectivity between
regions of the brain, and previous work has shown the potential of such
networks to distinguish between various brain disorders, giving rise to a
network classification problem. Existing approaches tend to either treat all
edge weights as a long vector, ignoring the network structure, or focus on
graph topology as represented by summary measures while ignoring the edge
weights. Our goal is to design a classification method that uses both the
individual edge information and the network structure of the data in a
computationally efficient way, and that can produce a parsimonious and
interpretable representation of differences in brain connectivity patterns
between classes. We propose a graph classification method that uses edge
weights as predictors but incorporates the network nature of the data via
penalties that promote sparsity in the number of nodes, in addition to the
usual sparsity penalties that encourage selection of edges. We implement the
method via efficient convex optimization and provide a detailed analysis of
data from two fMRI studies of schizophrenia.
| stat |
Bayesian Modeling of COVID-19 Positivity Rate -- the Indiana experience | In this short technical report we model, within the Bayesian framework, the
rate of positive tests reported by the the State of Indiana, accounting also
for the substantial variability (and overdispeartion) in the daily count of the
tests performed. The approach we take, results with a simple procedure for
prediction, a posteriori, of this rate of 'positivity' and allows for an easy
and a straightforward adaptation by any agency tracking daily results of
COVID-19 tests. The numerical results provided herein were obtained via an
updatable R Markdown document.
| stat |
Demystifying MMD GANs | We investigate the training and performance of generative adversarial
networks using the Maximum Mean Discrepancy (MMD) as critic, termed MMD GANs.
As our main theoretical contribution, we clarify the situation with bias in GAN
loss functions raised by recent work: we show that gradient estimators used in
the optimization process for both MMD GANs and Wasserstein GANs are unbiased,
but learning a discriminator based on samples leads to biased gradients for the
generator parameters. We also discuss the issue of kernel choice for the MMD
critic, and characterize the kernel corresponding to the energy distance used
for the Cramer GAN critic. Being an integral probability metric, the MMD
benefits from training strategies recently developed for Wasserstein GANs. In
experiments, the MMD GAN is able to employ a smaller critic network than the
Wasserstein GAN, resulting in a simpler and faster-training algorithm with
matching performance. We also propose an improved measure of GAN convergence,
the Kernel Inception Distance, and show how to use it to dynamically adapt
learning rates during GAN training.
| stat |
Selective Inference for Testing Trees and Edges in Phylogenetics | Selective inference is considered for testing trees and edges in phylogenetic
tree selection from molecular sequences. This improves the previously proposed
approximately unbiased test by adjusting the selection bias when testing many
trees and edges at the same time. The newly proposed selective inference
$p$-value is useful for testing selected edges to claim that they are
significantly supported if $p>1-\alpha$, whereas the non-selective $p$-value is
still useful for testing candidate trees to claim that they are rejected if
$p<\alpha$. The selective $p$-value controls the type-I error conditioned on
the selection event, whereas the non-selective $p$-value controls it
unconditionally. The selective and non-selective approximately unbiased
$p$-values are computed from two geometric quantities called signed distance
and mean curvature of the region representing tree or edge of interest in the
space of probability distributions. These two geometric quantities are
estimated by fitting a model of scaling-law to the non-parametric multiscale
bootstrap probabilities. Our general method is applicable to a wider class of
problems; phylogenetic tree selection is an example of model selection, and it
is interpreted as the variable selection of multiple regression, where each
edge corresponds to each predictor. Our method is illustrated in a previously
controversial phylogenetic analysis of human, rabbit and mouse.
| stat |
An Encoding Approach for Stable Change Point Detection | Without imposing prior distributional knowledge underlying multivariate time
series of interest, we propose a nonparametric change-point detection approach
to estimate the number of change points and their locations along the temporal
axis. We develop a structural subsampling procedure such that the observations
are encoded into multiple sequences of Bernoulli variables. A maximum
likelihood approach in conjunction with a newly developed searching algorithm
is implemented to detect change points on each Bernoulli process separately.
Then, aggregation statistics are proposed to collectively synthesize
change-point results from all individual univariate time series into consistent
and stable location estimations. We also study a weighting strategy to measure
the degree of relevance for different subsampled groups. Simulation studies are
conducted and shown that the proposed change-point methodology for multivariate
time series has favorable performance comparing with currently popular
nonparametric methods under various settings with different degrees of
complexity. Real data analyses are finally performed on categorical, ordinal,
and continuous time series taken from fields of genetics, climate, and finance.
| stat |
Individualized Multi-directional Variable Selection | In this paper we propose a heterogeneous modeling framework which achieves
individual-wise feature selection and individualized covariates' effects
subgrouping simultaneously. In contrast to conventional model selection
approaches, the new approach constructs a separation penalty with
multi-directional shrinkages, which facilitates individualized modeling to
distinguish strong signals from noisy ones and selects different relevant
variables for different individuals. Meanwhile, the proposed model identifies
subgroups among which individuals share similar covariates' effects, and thus
improves individualized estimation efficiency and feature selection accuracy.
Moreover, the proposed model also incorporates within-individual correlation
for longitudinal data to gain extra efficiency. We provide a general
theoretical foundation under a double-divergence modeling framework where the
number of individuals and the number of individual-wise measurements can both
diverge, which enables inference on both an individual level and a population
level. In particular, we establish strong oracle property for the
individualized estimator to ensure its optimal large sample property under
various conditions. An efficient ADMM algorithm is developed for computational
scalability. Simulation studies and applications to post-trauma mental disorder
analysis with genetic variation and an HIV longitudinal treatment study are
illustrated to compare the new approach to existing methods.
| stat |
Data-driven Uncertainty Quantification for Systematic Coarse-grained
Models | In this work, we present methodologies for the quantification of confidence
in bottom-up coarse-grained models for molecular and macromolecular systems.
Coarse-graining methods have been extensively used in the past decades in order
to extend the length and time scales accessible by simulation methodologies.
The quantification, though, of induced errors due to the limited availability
of fine-grained data is not yet established. Here, we employ rigorous
statistical methods to deduce guarantees for the optimal coarse models obtained
via approximations of the multi-body potential of mean force, with the relative
entropy, the relative entropy rate minimization, and the force matching
methods. Specifically, we present and apply statistical approaches, such as
bootstrap and jackknife, to infer confidence sets for a limited number of
samples, i.e., molecular configurations. Moreover, we estimate asymptotic
confidence intervals assuming adequate sampling of the phase space. We
demonstrate the need for non-asymptotic methods and quantify confidence sets
through two applications. The first is a two-scale fast/slow diffusion process
projected on the slow process. With this benchmark example, we establish the
methodology for both independent and time-series data. Second, we apply these
uncertainty quantification approaches on a polymeric bulk system. We consider
an atomistic polyethylene melt as the prototype system for developing
coarse-graining tools for macromolecular systems. For this system, we estimate
the coarse-grained force field and present confidence levels with respect to
the number of available microscopic data.
| stat |
Sample-efficient reinforcement learning using deep Gaussian processes | Reinforcement learning provides a framework for learning to control which
actions to take towards completing a task through trial-and-error. In many
applications observing interactions is costly, necessitating sample-efficient
learning. In model-based reinforcement learning efficiency is improved by
learning to simulate the world dynamics. The challenge is that model
inaccuracies rapidly accumulate over planned trajectories. We introduce deep
Gaussian processes where the depth of the compositions introduces model
complexity while incorporating prior knowledge on the dynamics brings
smoothness and structure. Our approach is able to sample a Bayesian posterior
over trajectories. We demonstrate highly improved early sample-efficiency over
competing methods. This is shown across a number of continuous control tasks,
including the half-cheetah whose contact dynamics have previously posed an
insurmountable problem for earlier sample-efficient Gaussian process based
models.
| stat |
A quantile-based g-computation approach to addressing the effects of
exposure mixtures | Exposure mixtures frequently occur in data across many domains, particularly
in the fields of environmental and nutritional epidemiology. Various strategies
have arisen to answer questions about mixtures, including methods such as
weighted quantile sum (WQS) regression that estimate a joint effect of the
mixture components.We demonstrate a new approach to estimating the joint
effects of a mixture: quantile g-computation. This approach combines the
inferential simplicity of WQS regression with the flexibility of g-computation,
a method of causal effect estimation. We use simulations to examine whether
quantile g-computation and WQS regression can accurately and precisely estimate
effects of mixtures in common scenarios. We examine the bias, confidence
interval coverage, and bias-variance tradeoff of quantile g-computation and WQS
regression, and how these quantities are impacted by the presence of non-causal
exposures, exposure correlation, unmeasured confounding, and non-linear
effects. Quantile g-computation, unlike WQS regression allows inference on
mixture effects that is unbiased with appropriate confidence interval coverage
at sample sizes typically encountered in epidemiologic studies and when the
assumptions of WQS regression are not met. Further, WQS regression can magnify
bias from unmeasured confounding that might occur if important components of
the mixture are omitted. Unlike inferential approaches that examine effects of
individual exposures, methods like quantile g-computation that can estimate the
effect of a mixture are essential for understanding effects of potential public
health actions that act on exposure sources. Our approach may serve to help
bridge gaps between epidemiologic analysis and interventions such as
regulations on industrial emissions or mining processes, dietary changes, or
consumer behavioral changes that act on multiple exposures simultaneously.
| stat |
Bayesian Batch Active Learning as Sparse Subset Approximation | Leveraging the wealth of unlabeled data produced in recent years provides
great potential for improving supervised models. When the cost of acquiring
labels is high, probabilistic active learning methods can be used to greedily
select the most informative data points to be labeled. However, for many
large-scale problems standard greedy procedures become computationally
infeasible and suffer from negligible model change. In this paper, we introduce
a novel Bayesian batch active learning approach that mitigates these issues.
Our approach is motivated by approximating the complete data posterior of the
model parameters. While naive batch construction methods result in correlated
queries, our algorithm produces diverse batches that enable efficient active
learning at scale. We derive interpretable closed-form solutions akin to
existing active learning procedures for linear models, and generalize to
arbitrary models using random projections. We demonstrate the benefits of our
approach on several large-scale regression and classification tasks.
| stat |
A Dynamic Separable Network Model with Actor Heterogeneity: An
Application to Global Weapons Transfers | In this paper we propose to extend the separable temporal exponential random
graph model (STERGM) to account for time-varying network- and actor-specific
effects. Our application case is the network of international major
conventional weapons transfers, based on data from the Stockholm International
Peace Research Institute (SIPRI). The application is particularly suitable
since it allows to distinguish the potentially differing driving forces for
creating new trade relationships and for the endurance of existing ones. In
accordance with political economy models we expect security- and
network-related covariates to be most important for the formation of transfers,
whereas repeated transfers should prevalently be determined by the receivers'
market size and military spending. Our proposed modelling approach corroborates
the hypothesis and quantifies the corresponding effects. Additionally, we
subject the time-varying heterogeneity effects to a functional principal
component analysis. This serves as exploratory tool and allows to identify
countries that stand out by exceptional increases or decreases of their
tendency to import and export weapons.
| stat |
A Cross-validated Ensemble Approach to Robust Hypothesis Testing of
Continuous Nonlinear Interactions: Application to Nutrition-Environment
Studies | Gene-environment and nutrition-environment studies often involve testing of
high-dimensional interactions between two sets of variables, each having
potentially complex nonlinear main effects on an outcome. Construction of a
valid and powerful hypothesis test for such an interaction is challenging, due
to the difficulty in constructing an efficient and unbiased estimator for the
complex, nonlinear main effects. In this work we address this problem by
proposing a Cross-validated Ensemble of Kernels (CVEK) that learns the space of
appropriate functions for the main effects using a cross-validated ensemble
approach. With a carefully chosen library of base kernels, CVEK flexibly
estimates the form of the main-effect functions from the data, and encourages
test power by guarding against over-fitting under the alternative. The method
is motivated by a study on the interaction between metal exposures in utero and
maternal nutrition on children's neurodevelopment in rural Bangladesh. The
proposed tests identified evidence of an interaction between minerals and
vitamins intake and arsenic and manganese exposures. Results suggest that the
detrimental effects of these metals are most pronounced at low intake levels of
the nutrients, suggesting nutritional interventions in pregnant women could
mitigate the adverse impacts of in utero metal exposures on children's
neurodevelopment.
| stat |
Prediction intervals for random-effects meta-analysis: a confidence
distribution approach | For the inference of random-effects models in meta-analysis, the prediction
interval was proposed as a summary measure of the treatment effects that
explains the heterogeneity in the target population. While the
Higgins-Thompson-Spiegelhalter (HTS) plug-in-type prediction interval has been
widely used, in which the heterogeneity parameter is replaced with its point
estimate, its validity depends on a large sample approximation. Most
meta-analyses, however, include less than 20 studies. It has been revealed that
the validity of the HTS method is not assured under realistic situations, but
no solution to this problem has been proposed in literature. Therefore, in this
article, we describe our proposed prediction interval. Instead of using the
plug-in scheme, we developed a bootstrap approach using an exact confidence
distribution to account for the uncertainty in estimation of the heterogeneity
parameter. Compared to the HTS method, the proposed method provides an accurate
prediction interval that adequately explains the heterogeneity of treatment
effects and the statistical error. Simulation studies demonstrated that the HTS
method had poor coverage performance; by contrast, the coverage probabilities
for the proposed method satisfactorily retained the nominal level. Applications
to three published random-effects meta-analyses are presented.
| stat |
Non-decimated Quaternion Wavelet Spectral Tools with Applications | Quaternion wavelets are redundant wavelet transforms generalizing
complex-valued non-decimated wavelet transforms. In this paper we propose a
matrix-formulation for non-decimated quaternion wavelet transforms and define
spectral tools for use in machine learning tasks. Since quaternionic algebra is
an extension of complex algebra, quaternion wavelets bring redundancy in the
components that proves beneficial in wavelet based tasks. Specifically, the
wavelet coefficients in the decomposition are quaternion-valued numbers that
define the modulus and three phases.
The novelty of this paper is definition of non-decimated quaternion wavelet
spectra based on the modulus and phase-dependent statistics as low-dimensional
summaries for 1-D signals or 2-D images. A structural redundancy in
non-decimated wavelets and a componential redundancy in quaternion wavelets are
linked to extract more informative features. In particular, we suggest an
improved way of classifying signals and images based on their scaling indices
in terms of spectral slopes and information contained in the three quaternionic
phases. We show that performance of the proposed method significantly improves
when compared to the standard versions of wavelets including the complex-valued
wavelets.
To illustrate performance of the proposed spectral tools we provide two
examples of application on real-data problems: classification of sounds using
scaling in high-frequency recordings over time and monitoring of steel rolling
process using the fractality of captured digitized images. The proposed tools
are compared with the counterparts based on standard wavelet transforms.
| stat |
Querying multiple sets of $p$-values through composed hypothesis testing | Motivation: Combining the results of different experiments to exhibit complex
patterns or to improve statistical power is a typical aim of data integration.
The starting point of the statistical analysis often comes as sets of
$p$-values resulting from previous analyses, that need to be combined in a
flexible way to explore complex hypotheses, while guaranteeing a low proportion
of false discoveries.
Results: We introduce the generic concept of {\sl composed hypothesis}, which
corresponds to an arbitrary complex combination of simple hypotheses. We
rephrase the problem of testing a composed hypothesis as a classification task,
and show that finding items for which the composed null hypothesis is rejected
boils down to fitting a mixture model and classify the items according to their
posterior probabilities. We show that inference can be efficiently performed
and provide a thorough classification rule to control for type I error. The
performance and the usefulness of the approach are illustrated on simulations
and on two different applications combining data from different types. The
method is scalable, does not require any parameter tuning, and provided
valuable biological insight on the considered application cases.
Availability: We implement the QCH methodology in the \texttt{qch} R package
hosted on CRAN.
| stat |
Constructing Confidence Intervals for the Signals in Sparse Phase
Retrieval | In this paper, we provide a general methodology to draw statistical
inferences on individual signal coordinates or linear combinations of them in
sparse phase retrieval. Given an initial estimator for the targeting parameter
(some simple function of the signal), which is generated by some existing
algorithm, we can modify it in a way that the modified version is
asymptotically normal and unbiased. Then confidence intervals and hypothesis
testings can be constructed based on this asymptotic normality. For
conciseness, we focus on confidence intervals in this work, while a similar
procedure can be adopted for hypothesis testings. Under some mild assumptions
on the signal and sample size, we establish theoretical guarantees for the
proposed method. These assumptions are generally weak in the sense that the
dimension could exceed the sample size and many non-zero small coordinates are
allowed. Furthermore, theoretical analysis reveals that the modified estimators
for individual coordinates have uniformly bounded variance, and hence
simultaneous interval estimation is possible. Numerical simulations in a wide
range of settings are supportive of our theoretical results.
| stat |
Asymptotic Unbiasedness of the Permutation Importance Measure in Random
Forest Models | Variable selection in sparse regression models is an important task as
applications ranging from biomedical research to econometrics have shown.
Especially for higher dimensional regression problems, for which the link
function between response and covariates cannot be directly detected, the
selection of informative variables is challenging. Under these circumstances,
the Random Forest method is a helpful tool to predict new outcomes while
delivering measures for variable selection. One common approach is the usage of
the permutation importance. Due to its intuitive idea and flexible usage, it is
important to explore circumstances, for which the permutation importance based
on Random Forest correctly indicates informative covariates. Regarding the
latter, we deliver theoretical guarantees for the validity of the permutation
importance measure under specific assumptions and prove its (asymptotic)
unbiasedness. An extensive simulation study verifies our findings.
| stat |
Does Covid-19 Mass Testing Work? The Importance of Accounting for the
Epidemic Dynamics | Mass antigen testing has been proposed as a possible cost-effective tool to
contain the Covid-19 pandemic. We test the impact of a voluntary mass testing
campaign implemented in the Italian region of South Tyrol on the spread of the
virus in the following months. We do so by using an innovative empirical
approach which embeds a semi-parametric growth model - where Covid-19
transmission dynamics are allowed to vary across regions and to be impacted by
the implementation of the mass testing campaign - into a synthetic control
framework which creates an appropriate control group of other Italian regions.
We find that the mass test campaign decreased the growth rate of Covid-19 by
39% which corresponds to a reduction in the total additional cases of 18%, 30%
and 56% within ten, twenty and forty days from the intervention date,
respectively. Our results suggest that mass testing campaigns are useful
instruments for mitigating the pandemic.
| stat |
Model fitting in Multiple Systems Analysis for the quantification of
Modern Slavery: Classical and Bayesian approaches | Multiple systems estimation is a key approach for quantifying hidden
populations such as the number of victims of modern slavery. The UK Government
published an estimate of 10,000 to 13,000 victims, constructed by the present
author, as part of the strategy leading to the Modern Slavery Act 2015. This
estimate was obtained by a stepwise multiple systems method based on six lists.
Further investigation shows that a small proportion of the possible models give
rather different answers, and that other model fitting approaches may choose
one of these. Three data sets collected in the field of modern slavery,
together with a data set about the death toll in the Kosovo conflict, are used
to investigate the stability and robustness of various multiple systems
estimate methods. The crucial aspect is the way that interactions between lists
are modelled, because these can substantially affect the results. Model
selection and Bayesian approaches are considered in detail, in particular to
assess their stability and robustness when applied to real modern slavery data.
A new Markov Chain Monte Carlo Bayesian approach is developed; overall, this
gives robust and stable results at least for the examples considered. The
software and datasets are freely and publicly available to facilitate wider
implementation and further research.
| stat |
Goodness-of-fit Test for Latent Block Models | Latent block models are used for probabilistic biclustering, which is shown
to be an effective method for analyzing various relational data sets. However,
there has been no statistical test method for determining the row and column
cluster numbers of latent block models. Recent studies have constructed
statistical-test-based methods for stochastic block models, which assume that
the observed matrix is a square symmetric matrix and that the cluster
assignments are the same for rows and columns. In this study, we developed a
new goodness-of-fit test for latent block models to test whether an observed
data matrix fits a given set of row and column cluster numbers, or it consists
of more clusters in at least one direction of the row and the column. To
construct the test method, we used a result from the random matrix theory for a
sample covariance matrix. We experimentally demonstrated the effectiveness of
the proposed method by showing the asymptotic behavior of the test statistic
and measuring the test accuracy.
| stat |
Stay Ahead of Poachers: Illegal Wildlife Poaching Prediction and Patrol
Planning Under Uncertainty with Field Test Evaluations | Illegal wildlife poaching threatens ecosystems and drives endangered species
toward extinction. However, efforts for wildlife protection are constrained by
the limited resources of law enforcement agencies. To help combat poaching, the
Protection Assistant for Wildlife Security (PAWS) is a machine learning
pipeline that has been developed as a data-driven approach to identify areas at
high risk of poaching throughout protected areas and compute optimal patrol
routes. In this paper, we take an end-to-end approach to the data-to-deployment
pipeline for anti-poaching. In doing so, we address challenges including
extreme class imbalance (up to 1:200), bias, and uncertainty in wildlife
poaching data to enhance PAWS, and we apply our methodology to three national
parks with diverse characteristics. (i) We use Gaussian processes to quantify
predictive uncertainty, which we exploit to improve robustness of our
prescribed patrols and increase detection of snares by an average of 30%. We
evaluate our approach on real-world historical poaching data from Murchison
Falls and Queen Elizabeth National Parks in Uganda and, for the first time,
Srepok Wildlife Sanctuary in Cambodia. (ii) We present the results of
large-scale field tests conducted in Murchison Falls and Srepok Wildlife
Sanctuary which confirm that the predictive power of PAWS extends promisingly
to multiple parks. This paper is part of an effort to expand PAWS to 800 parks
around the world through integration with SMART conservation software.
| stat |
Optimum Testing Time of Software using Size-Biased Concepts | Optimum software release time problem has been an interesting area of
research for several decades now. We introduce here a new concept of
size-biased modelling to solve for the optimum software release time. Bayesian
approach is used to solve the problem. We also discuss about the applicability
of the model for a specific data set, though we believe that the model is
applicable to all kind of software reliability data collected in a discrete
framework. It has applications in other fields like oil exploration also.
Finally, we compare favourably our model with another similar model published
recently. We also provide in this article some future possibilities of research
work.
| stat |
Multiple-bias sensitivity analysis using bounds | Unmeasured confounding, selection bias, and measurement error are well-known
sources of bias in epidemiologic research. Methods for assessing these biases
have their own limitations. Many quantitative sensitivity analysis approaches
consider each type of bias individually, while more complex approaches are
harder to implement or require numerous assumptions. By failing to consider
multiple biases at once, researchers can underestimate -- or overestimate --
their joint impact. We show that it is possible to bound the total composite
bias due to these three sources, and to use that bound to assess the
sensitivity of a risk ratio to any combination of these biases. We derive
bounds for the total composite bias under a variety of scenarios, providing
researchers with tools to assess their total potential impact. We apply this
technique to a study where unmeasured confounding and selection bias are both
concerns, and to another study in which possible differential exposure
misclassification and unmeasured confounding are concerns. We also show that a
"multi-bias E-value" can describe the minimal strength of joint bias-parameter
association necessary for an observed risk ratio to be compatible with a null
causal effect (or with other pre-specified effect sizes). This may provide
intuition about the relative impacts of each type of bias. The approach we
describe is easy to implement with minimal assumptions, and we provide R
functions to do so.
| stat |
Modelling publication bias and p-hacking | Publication bias and p-hacking are two well-known phenomena that strongly
affect the scientific literature and cause severe problems in meta-analyses.
Due to these phenomena, the assumptions of meta-analyses are seriously violated
and the results of the studies cannot be trusted. While publication bias is
almost perfectly captured by the weighting function selection model, p-hacking
is much harder to model and no definitive solution has been found yet. In this
paper we propose to model both publication bias and p-hacking with selection
models. We derive some properties for these models, and we compare them
formally and through simulations. Finally, two real data examples are used to
show how the models work in practice.
| stat |
LSMI-Sinkhorn: Semi-supervised Squared-Loss Mutual Information
Estimation with Optimal Transport | Estimating mutual information is an important machine learning and statistics
problem. To estimate the mutual information from data, a common practice is
preparing a set of paired samples. However, in some cases, it is difficult to
obtain a large number of data pairs. To address this problem, we propose
squared-loss mutual information (SMI) estimation using a small number of paired
samples and the available unpaired ones. We first represent SMI through the
density ratio function, where the expectation is approximated by the samples
from marginals and its assignment parameters. The objective is formulated using
the optimal transport problem and quadratic programming. Then, we introduce the
least-square mutual information-Sinkhorn algorithm (LSMI-Sinkhorn) for
efficient optimization. Through experiments, we first demonstrate that the
proposed method can estimate the SMI without a large number of paired samples.
We also evaluate and show the effectiveness of the proposed LSMI-Sinkhorn on
various types of machine learning problems such as image matching and photo
album summarization.
| stat |
Distribution-free uncertainty quantification for classification under
label shift | Trustworthy deployment of ML models requires a proper measure of uncertainty,
especially in safety-critical applications. We focus on uncertainty
quantification (UQ) for classification problems via two avenues -- prediction
sets using conformal prediction and calibration of probabilistic predictors by
post-hoc binning -- since these possess distribution-free guarantees for i.i.d.
data. Two common ways of generalizing beyond the i.i.d. setting include
handling covariate and label shift. Within the context of distribution-free UQ,
the former has already received attention, but not the latter. It is known that
label shift hurts prediction, and we first argue that it also hurts UQ, by
showing degradation in coverage and calibration. Piggybacking on recent
progress in addressing label shift (for better prediction), we examine the
right way to achieve UQ by reweighting the aforementioned conformal and
calibration procedures whenever some unlabeled data from the target
distribution is available. We examine these techniques theoretically in a
distribution-free framework and demonstrate their excellent practical
performance.
| stat |
Sensitivity analyses for effect modifiers not observed in the target
population when generalizing treatment effects from a randomized controlled
trial: Assumptions, models, effect scales, data scenarios, and implementation
details | Background: Randomized controlled trials are often used to inform policy and
practice for broad populations. The average treatment effect (ATE) for a target
population, however, may be different from the ATE observed in a trial if there
are effect modifiers whose distribution in the target population is different
that from that in the trial. Methods exist to use trial data to estimate the
target population ATE, provided the distributions of treatment effect modifiers
are observed in both the trial and target population -- an assumption that may
not hold in practice.
Methods: The proposed sensitivity analyses address the situation where a
treatment effect modifier is observed in the trial but not the target
population. These methods are based on an outcome model or the combination of
such a model and weighting adjustment for observed differences between the
trial sample and target population. They accommodate several types of outcome
models: linear models (including single time outcome and pre- and
post-treatment outcomes) for additive effects, and models with log or logit
link for multiplicative effects. We clarify the methods' assumptions and
provide detailed implementation instructions.
Illustration: We illustrate the methods using an example generalizing the
effects of an HIV treatment regimen from a randomized trial to a relevant
target population.
Conclusion: These methods allow researchers and decision-makers to have more
appropriate confidence when drawing conclusions about target population
effects.
| stat |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.