title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Introduction to Clustering Algorithms and Applications | Data clustering is the process of identifying natural groupings or clusters
within multidimensional data based on some similarity measure. Clustering is a
fundamental process in many different disciplines. Hence, researchers from
different fields are actively working on the clustering problem. This paper
provides an overview of the different representative clustering methods. In
addition, application of clustering in different field is briefly introduced.
|
A new integral loss function for Bayesian optimization | We consider the problem of maximizing a real-valued continuous function $f$
using a Bayesian approach. Since the early work of Jonas Mockus and Antanas
\v{Z}ilinskas in the 70's, the problem of optimization is usually formulated by
considering the loss function $\max f - M_n$ (where $M_n$ denotes the best
function value observed after $n$ evaluations of $f$). This loss function puts
emphasis on the value of the maximum, at the expense of the location of the
maximizer. In the special case of a one-step Bayes-optimal strategy, it leads
to the classical Expected Improvement (EI) sampling criterion. This is a
special case of a Stepwise Uncertainty Reduction (SUR) strategy, where the risk
associated to a certain uncertainty measure (here, the expected loss) on the
quantity of interest is minimized at each step of the algorithm. In this
article, assuming that $f$ is defined over a measure space $(\mathbb{X},
\lambda)$, we propose to consider instead the integral loss function
$\int_{\mathbb{X}} (f - M_n)_{+}\, d\lambda$, and we show that this leads, in
the case of a Gaussian process prior, to a new numerically tractable sampling
criterion that we call $\rm EI^2$ (for Expected Integrated Expected
Improvement). A numerical experiment illustrates that a SUR strategy based on
this new sampling criterion reduces the error on both the value and the
location of the maximizer faster than the EI-based strategy.
|
AFP Algorithm and a Canonical Normal Form for Horn Formulas | AFP Algorithm is a learning algorithm for Horn formulas. We show that it does
not improve the complexity of AFP Algorithm, if after each negative
counterexample more that just one refinements are performed. Moreover, a
canonical normal form for Horn formulas is presented, and it is proved that the
output formula of AFP Algorithm is in this normal form.
|
Conic Multi-Task Classification | Traditionally, Multi-task Learning (MTL) models optimize the average of
task-related objective functions, which is an intuitive approach and which we
will be referring to as Average MTL. However, a more general framework,
referred to as Conic MTL, can be formulated by considering conic combinations
of the objective functions instead; in this framework, Average MTL arises as a
special case, when all combination coefficients equal 1. Although the advantage
of Conic MTL over Average MTL has been shown experimentally in previous works,
no theoretical justification has been provided to date. In this paper, we
derive a generalization bound for the Conic MTL method, and demonstrate that
the tightest bound is not necessarily achieved, when all combination
coefficients equal 1; hence, Average MTL may not always be the optimal choice,
and it is important to consider Conic MTL. As a byproduct of the generalization
bound, it also theoretically explains the good experimental results of previous
relevant works. Finally, we propose a new Conic MTL model, whose conic
combination coefficients minimize the generalization bound, instead of choosing
them heuristically as has been done in previous methods. The rationale and
advantage of our model is demonstrated and verified via a series of experiments
by comparing with several other methods.
|
Diffusion Fingerprints | We introduce, test and discuss a method for classifying and clustering data
modeled as directed graphs. The idea is to start diffusion processes from any
subset of a data collection, generating corresponding distributions for
reaching points in the network. These distributions take the form of
high-dimensional numerical vectors and capture essential topological properties
of the original dataset. We show how these diffusion vectors can be
successfully applied for getting state-of-the-art accuracies in the problem of
extracting pathways from metabolic networks. We also provide a guideline to
illustrate how to use our method for classification problems, and discuss
important details of its implementation. In particular, we present a simple
dimensionality reduction technique that lowers the computational cost of
classifying diffusion vectors, while leaving the predictive power of the
classification process substantially unaltered. Although the method has very
few parameters, the results we obtain show its flexibility and power. This
should make it helpful in many other contexts.
|
Caffe: Convolutional Architecture for Fast Feature Embedding | Caffe provides multimedia scientists and practitioners with a clean and
modifiable framework for state-of-the-art deep learning algorithms and a
collection of reference models. The framework is a BSD-licensed C++ library
with Python and MATLAB bindings for training and deploying general-purpose
convolutional neural networks and other deep models efficiently on commodity
architectures. Caffe fits industry and internet-scale media needs by CUDA GPU
computation, processing over 40 million images a day on a single K40 or Titan
GPU ($\approx$ 2.5 ms per image). By separating model representation from
actual implementation, Caffe allows experimentation and seamless switching
among platforms for ease of development and deployment from prototyping
machines to cloud environments. Caffe is maintained and developed by the
Berkeley Vision and Learning Center (BVLC) with the help of an active community
of contributors on GitHub. It powers ongoing research projects, large-scale
industrial applications, and startup prototypes in vision, speech, and
multimedia.
|
Uniform Sampling for Matrix Approximation | Random sampling has become a critical tool in solving massive matrix
problems. For linear regression, a small, manageable set of data rows can be
randomly selected to approximate a tall, skinny data matrix, improving
processing time significantly. For theoretical performance guarantees, each row
must be sampled with probability proportional to its statistical leverage
score. Unfortunately, leverage scores are difficult to compute.
A simple alternative is to sample rows uniformly at random. While this often
works, uniform sampling will eliminate critical row information for many
natural instances. We take a fresh look at uniform sampling by examining what
information it does preserve. Specifically, we show that uniform sampling
yields a matrix that, in some sense, well approximates a large fraction of the
original. While this weak form of approximation is not enough for solving
linear regression directly, it is enough to compute a better approximation.
This observation leads to simple iterative row sampling algorithms for matrix
approximation that run in input-sparsity time and preserve row structure and
sparsity at all intermediate steps. In addition to an improved understanding of
uniform sampling, our main proof introduces a structural result of independent
interest: we show that every matrix can be made to have low coherence by
reweighting a small subset of its rows.
|
A two-stage architecture for stock price forecasting by combining SOM
and fuzzy-SVM | This paper proposed a model to predict the stock price based on combining
Self-Organizing Map (SOM) and fuzzy-Support Vector Machines (f-SVM). Extraction
of fuzzy rules from raw data based on the combining of statistical machine
learning models is base of this proposed approach. In the proposed model, SOM
is used as a clustering algorithm to partition the whole input space into the
several disjoint regions. For each partition, a set of fuzzy rules is extracted
based on a f-SVM combining model. Then fuzzy rules sets are used to predict the
test data using fuzzy inference algorithms. The performance of the proposed
approach is compared with other models using four data sets
|
Improving the Interpretability of Support Vector Machines-based Fuzzy
Rules | Support vector machines (SVMs) and fuzzy rule systems are functionally
equivalent under some conditions. Therefore, the learning algorithms developed
in the field of support vector machines can be used to adapt the parameters of
fuzzy systems. Extracting fuzzy models from support vector machines has the
inherent advantage that the model does not need to determine the number of
rules in advance. However, after the support vector machine learning, the
complexity is usually high, and interpretability is also impaired. This paper
not only proposes a complete framework for extracting interpretable SVM-based
fuzzy modeling, but also provides optimization issues of the models.
Simulations examples are given to embody the idea of this paper.
|
Nonconvex Statistical Optimization: Minimax-Optimal Sparse PCA in
Polynomial Time | Sparse principal component analysis (PCA) involves nonconvex optimization for
which the global solution is hard to obtain. To address this issue, one popular
approach is convex relaxation. However, such an approach may produce suboptimal
estimators due to the relaxation effect. To optimally estimate sparse principal
subspaces, we propose a two-stage computational framework named "tighten after
relax": Within the 'relax' stage, we approximately solve a convex relaxation of
sparse PCA with early stopping to obtain a desired initial estimator; For the
'tighten' stage, we propose a novel algorithm called sparse orthogonal
iteration pursuit (SOAP), which iteratively refines the initial estimator by
directly solving the underlying nonconvex problem. A key concept of this
two-stage framework is the basin of attraction. It represents a local region
within which the `tighten' stage has desired computational and statistical
guarantees. We prove that, the initial estimator obtained from the 'relax'
stage falls into such a region, and hence SOAP geometrically converges to a
principal subspace estimator which is minimax-optimal within a certain model
class. Unlike most existing sparse PCA estimators, our approach applies to the
non-spiked covariance models, and adapts to non-Gaussianity as well as
dependent data settings. Moreover, through analyzing the computational
complexity of the two stages, we illustrate an interesting phenomenon that
larger sample size can reduce the total iteration complexity. Our framework
motivates a general paradigm for solving many complex statistical problems
which involve nonconvex optimization with provable guarantees.
|
Computing Multi-Relational Sufficient Statistics for Large Databases | Databases contain information about which relationships do and do not hold
among entities. To make this information accessible for statistical analysis
requires computing sufficient statistics that combine information from
different database tables. Such statistics may involve any number of {\em
positive and negative} relationships. With a naive enumeration approach,
computing sufficient statistics for negative relationships is feasible only for
small databases. We solve this problem with a new dynamic programming algorithm
that performs a virtual join, where the requisite counts are computed without
materializing join tables. Contingency table algebra is a new extension of
relational algebra, that facilitates the efficient implementation of this
M\"obius virtual join operation. The M\"obius Join scales to large datasets
(over 1M tuples) with complex schemas. Empirical evaluation with seven
benchmark datasets showed that information about the presence and absence of
links can be exploited in feature selection, association rule mining, and
Bayesian network learning.
|
Hierarchical Adaptive Structural SVM for Domain Adaptation | A key topic in classification is the accuracy loss produced when the data
distribution in the training (source) domain differs from that in the testing
(target) domain. This is being recognized as a very relevant problem for many
computer vision tasks such as image classification, object detection, and
object category recognition. In this paper, we present a novel domain
adaptation method that leverages multiple target domains (or sub-domains) in a
hierarchical adaptation tree. The core idea is to exploit the commonalities and
differences of the jointly considered target domains.
Given the relevance of structural SVM (SSVM) classifiers, we apply our idea
to the adaptive SSVM (A-SSVM), which only requires the target domain samples
together with the existing source-domain classifier for performing the desired
adaptation. Altogether, we term our proposal as hierarchical A-SSVM (HA-SSVM).
As proof of concept we use HA-SSVM for pedestrian detection and object
category recognition. In the former we apply HA-SSVM to the deformable
part-based model (DPM) while in the latter HA-SSVM is applied to multi-category
classifiers. In both cases, we show how HA-SSVM is effective in increasing the
detection/recognition accuracy with respect to adaptation strategies that
ignore the structure of the target data. Since, the sub-domains of the target
data are not always known a priori, we shown how HA-SSVM can incorporate
sub-domain structure discovery for object category recognition.
|
A Case Study in Text Mining: Interpreting Twitter Data From World Cup
Tweets | Cluster analysis is a field of data analysis that extracts underlying
patterns in data. One application of cluster analysis is in text-mining, the
analysis of large collections of text to find similarities between documents.
We used a collection of about 30,000 tweets extracted from Twitter just before
the World Cup started. A common problem with real world text data is the
presence of linguistic noise. In our case it would be extraneous tweets that
are unrelated to dominant themes. To combat this problem, we created an
algorithm that combined the DBSCAN algorithm and a consensus matrix. This way
we are left with the tweets that are related to those dominant themes. We then
used cluster analysis to find those topics that the tweets describe. We
clustered the tweets using k-means, a commonly used clustering algorithm, and
Non-Negative Matrix Factorization (NMF) and compared the results. The two
algorithms gave similar results, but NMF proved to be faster and provided more
easily interpreted results. We explored our results using two visualization
tools, Gephi and Wordle.
|
Stretchy Polynomial Regression | This article proposes a novel solution for stretchy polynomial regression
learning. The solution comes in primal and dual closed-forms similar to that of
ridge regression. Essentially, the proposed solution stretches the covariance
computation via a power term thereby compresses or amplifies the estimation.
Our experiments on both synthetic data and real-world data show effectiveness
of the proposed method for compressive learning.
|
Interpreting Tree Ensembles with inTrees | Tree ensembles such as random forests and boosted trees are accurate but
difficult to understand, debug and deploy. In this work, we provide the inTrees
(interpretable trees) framework that extracts, measures, prunes and selects
rules from a tree ensemble, and calculates frequent variable interactions. An
rule-based learner, referred to as the simplified tree ensemble learner (STEL),
can also be formed and used for future prediction. The inTrees framework can
applied to both classification and regression problems, and is applicable to
many types of tree ensembles, e.g., random forests, regularized random forests,
and boosted trees. We implemented the inTrees algorithms in the "inTrees" R
package.
|
To lie or not to lie in a subspace | Give deterministic necessary and sufficient conditions to guarantee that if a
subspace fits certain partially observed data from a union of subspaces, it is
because such data really lies in a subspace.
Furthermore, Give deterministic necessary and sufficient conditions to
guarantee that if a subspace fits certain partially observed data, such
subspace is unique.
Do this by characterizing when and only when a set of incomplete vectors
behaves as a single but complete one.
|
Supervised Hashing Using Graph Cuts and Boosted Decision Trees | Embedding image features into a binary Hamming space can improve both the
speed and accuracy of large-scale query-by-example image retrieval systems.
Supervised hashing aims to map the original features to compact binary codes in
a manner which preserves the label-based similarities of the original data.
Most existing approaches apply a single form of hash function, and an
optimization process which is typically deeply coupled to this specific form.
This tight coupling restricts the flexibility of those methods, and can result
in complex optimization problems that are difficult to solve. In this work we
proffer a flexible yet simple framework that is able to accommodate different
types of loss functions and hash functions. The proposed framework allows a
number of existing approaches to hashing to be placed in context, and
simplifies the development of new problem-specific hashing methods. Our
framework decomposes the into two steps: binary code (hash bits) learning, and
hash function learning. The first step can typically be formulated as a binary
quadratic problem, and the second step can be accomplished by training standard
binary classifiers. For solving large-scale binary code inference, we show how
to ensure that the binary quadratic problems are submodular such that an
efficient graph cut approach can be used. To achieve efficiency as well as
efficacy on large-scale high-dimensional data, we propose to use boosted
decision trees as the hash functions, which are nonlinear, highly descriptive,
and very fast to train and evaluate. Experiments demonstrate that our proposed
method significantly outperforms most state-of-the-art methods, especially on
high-dimensional data.
|
An application of topological graph clustering to protein function
prediction | We use a semisupervised learning algorithm based on a topological data
analysis approach to assign functional categories to yeast proteins using
similarity graphs. This new approach to analyzing biological networks yields
results that are as good as or better than state of the art existing
approaches.
|
Asymptotic Accuracy of Bayesian Estimation for a Single Latent Variable | In data science and machine learning, hierarchical parametric models, such as
mixture models, are often used. They contain two kinds of variables: observable
variables, which represent the parts of the data that can be directly measured,
and latent variables, which represent the underlying processes that generate
the data. Although there has been an increase in research on the estimation
accuracy for observable variables, the theoretical analysis of estimating
latent variables has not been thoroughly investigated. In a previous study, we
determined the accuracy of a Bayes estimation for the joint probability of the
latent variables in a dataset, and we proved that the Bayes method is
asymptotically more accurate than the maximum-likelihood method. However, the
accuracy of the Bayes estimation for a single latent variable remains unknown.
In the present paper, we derive the asymptotic expansions of the error
functions, which are defined by the Kullback-Leibler divergence, for two types
of single-variable estimations when the statistical regularity is satisfied.
Our results indicate that the accuracies of the Bayes and maximum-likelihood
methods are asymptotically equivalent and clarify that the Bayes method is only
advantageous for multivariable estimations.
|
Improved Distributed Principal Component Analysis | We study the distributed computing setting in which there are multiple
servers, each holding a set of points, who wish to compute functions on the
union of their point sets. A key task in this setting is Principal Component
Analysis (PCA), in which the servers would like to compute a low dimensional
subspace capturing as much of the variance of the union of their point sets as
possible. Given a procedure for approximate PCA, one can use it to
approximately solve $\ell_2$-error fitting problems such as $k$-means
clustering and subspace clustering. The essential properties of an approximate
distributed PCA algorithm are its communication cost and computational
efficiency for a given desired accuracy in downstream applications. We give new
algorithms and analyses for distributed PCA which lead to improved
communication and computational costs for $k$-means clustering and related
problems. Our empirical study on real world data shows a speedup of orders of
magnitude, preserving communication with only a negligible degradation in
solution quality. Some of these techniques we develop, such as a general
transformation from a constant success probability subspace embedding to a high
success probability subspace embedding with a dimension and sparsity
independent of the success probability, may be of independent interest.
|
Analysis of a Reduced-Communication Diffusion LMS Algorithm | In diffusion-based algorithms for adaptive distributed estimation, each node
of an adaptive network estimates a target parameter vector by creating an
intermediate estimate and then combining the intermediate estimates available
within its closed neighborhood. We analyze the performance of a
reduced-communication diffusion least mean-square (RC-DLMS) algorithm, which
allows each node to receive the intermediate estimates of only a subset of its
neighbors at each iteration. This algorithm eases the usage of network
communication resources and delivers a trade-off between estimation performance
and communication cost. We show analytically that the RC-DLMS algorithm is
stable and convergent in both mean and mean-square senses. We also calculate
its theoretical steady-state mean-square deviation. Simulation results
demonstrate a good match between theory and experiment.
|
Label Distribution Learning | Although multi-label learning can deal with many problems with label
ambiguity, it does not fit some real applications well where the overall
distribution of the importance of the labels matters. This paper proposes a
novel learning paradigm named \emph{label distribution learning} (LDL) for such
kind of applications. The label distribution covers a certain number of labels,
representing the degree to which each label describes the instance. LDL is a
more general learning framework which includes both single-label and
multi-label learning as its special cases. This paper proposes six working LDL
algorithms in three ways: problem transformation, algorithm adaptation, and
specialized algorithm design. In order to compare the performance of the LDL
algorithms, six representative and diverse evaluation measures are selected via
a clustering analysis, and the first batch of label distribution datasets are
collected and made publicly available. Experimental results on one artificial
and fifteen real-world datasets show clear advantages of the specialized
algorithms, which indicates the importance of special design for the
characteristics of the LDL problem.
|
PMCE: efficient inference of expressive models of cancer evolution with
high prognostic power | Motivation: Driver (epi)genomic alterations underlie the positive selection
of cancer subpopulations, which promotes drug resistance and relapse. Even
though substantial heterogeneity is witnessed in most cancer types, mutation
accumulation patterns can be regularly found and can be exploited to
reconstruct predictive models of cancer evolution. Yet, available methods
cannot infer logical formulas connecting events to represent alternative
evolutionary routes or convergent evolution. Results: We introduce PMCE, an
expressive framework that leverages mutational profiles from cross-sectional
sequencing data to infer probabilistic graphical models of cancer evolution
including arbitrary logical formulas, and which outperforms the
state-of-the-art in terms of accuracy and robustness to noise, on simulations.
The application of PMCE to 7866 samples from the TCGA database allows us to
identify a highly significant correlation between the predicted evolutionary
paths and the overall survival in 7 tumor types, proving that our approach can
effectively stratify cancer patients in reliable risk groups. Availability:
PMCE is freely available at https://github.com/BIMIB-DISCo/PMCE, in addition to
the code to replicate all the analyses presented in the manuscript. Contacts:
[email protected], [email protected].
|
Recursive Total Least-Squares Algorithm Based on Inverse Power Method
and Dichotomous Coordinate-Descent Iterations | We develop a recursive total least-squares (RTLS) algorithm for
errors-in-variables system identification utilizing the inverse power method
and the dichotomous coordinate-descent (DCD) iterations. The proposed
algorithm, called DCD-RTLS, outperforms the previously-proposed RTLS
algorithms, which are based on the line-search method, with reduced
computational complexity. We perform a comprehensive analysis of the DCD-RTLS
algorithm and show that it is asymptotically unbiased as well as being stable
in the mean. We also find a lower bound for the forgetting factor that ensures
mean-square stability of the algorithm and calculate the theoretical
steady-state mean-square deviation (MSD). We verify the effectiveness of the
proposed algorithm and the accuracy of the predicted steady-state MSD via
simulations.
|
A Methodology for the Diagnostic of Aircraft Engine Based on Indicators
Aggregation | Aircraft engine manufacturers collect large amount of engine related data
during flights. These data are used to detect anomalies in the engines in order
to help companies optimize their maintenance costs. This article introduces and
studies a generic methodology that allows one to build automatic early signs of
anomaly detection in a way that is understandable by human operators who make
the final maintenance decision. The main idea of the method is to generate a
very large number of binary indicators based on parametric anomaly scores
designed by experts, complemented by simple aggregations of those scores. The
best indicators are selected via a classical forward scheme, leading to a much
reduced number of indicators that are tuned to a data set. We illustrate the
interest of the method on simulated data which contain realistic early signs of
anomalies.
|
Large Scale Purchase Prediction with Historical User Actions on B2C
Online Retail Platform | This paper describes the solution of Bazinga Team for Tmall Recommendation
Prize 2014. With real-world user action data provided by Tmall, one of the
largest B2C online retail platforms in China, this competition requires to
predict future user purchases on Tmall website. Predictions are judged on
F1Score, which considers both precision and recall for fair evaluation. The
data set provided by Tmall contains more than half billion action records from
over ten million distinct users. Such massive data volume poses a big
challenge, and drives competitors to write every single program in MapReduce
fashion and run it on distributed cluster. We model the purchase prediction
problem as standard machine learning problem, and mainly employ regression and
classification methods as single models. Individual models are then aggregated
in a two-stage approach, using linear regression for blending, and finally a
linear ensemble of blended models. The competition is approaching the end but
still in running during writing this paper. In the end, our team achieves
F1Score 6.11 and ranks 7th (out of 7,276 teams in total).
|
Task-group Relatedness and Generalization Bounds for Regularized
Multi-task Learning | In this paper, we study the generalization performance of regularized
multi-task learning (RMTL) in a vector-valued framework, where MTL is
considered as a learning process for vector-valued functions. We are mainly
concerned with two theoretical questions: 1) under what conditions does RMTL
perform better with a smaller task sample size than STL? 2) under what
conditions is RMTL generalizable and can guarantee the consistency of each task
during simultaneous learning?
In particular, we investigate two types of task-group relatedness: the
observed discrepancy-dependence measure (ODDM) and the empirical
discrepancy-dependence measure (EDDM), both of which detect the dependence
between two groups of multiple related tasks (MRTs). We then introduce the
Cartesian product-based uniform entropy number (CPUEN) to measure the
complexities of vector-valued function classes. By applying the specific
deviation and the symmetrization inequalities to the vector-valued framework,
we obtain the generalization bound for RMTL, which is the upper bound of the
joint probability of the event that there is at least one task with a large
empirical discrepancy between the expected and empirical risks. Finally, we
present a sufficient condition to guarantee the consistency of each task in the
simultaneous learning process, and we discuss how task relatedness affects the
generalization performance of RMTL. Our theoretical findings answer the
aforementioned two questions.
|
Falsifiable implies Learnable | The paper demonstrates that falsifiability is fundamental to learning. We
prove the following theorem for statistical learning and sequential prediction:
If a theory is falsifiable then it is learnable -- i.e. admits a strategy that
predicts optimally. An analogous result is shown for universal induction.
|
Sparse Generalized Eigenvalue Problem via Smooth Optimization | In this paper, we consider an $\ell_{0}$-norm penalized formulation of the
generalized eigenvalue problem (GEP), aimed at extracting the leading sparse
generalized eigenvector of a matrix pair. The formulation involves maximization
of a discontinuous nonconcave objective function over a nonconvex constraint
set, and is therefore computationally intractable. To tackle the problem, we
first approximate the $\ell_{0}$-norm by a continuous surrogate function. Then
an algorithm is developed via iteratively majorizing the surrogate function by
a quadratic separable function, which at each iteration reduces to a regular
generalized eigenvalue problem. A preconditioned steepest ascent algorithm for
finding the leading generalized eigenvector is provided. A systematic way based
on smoothing is proposed to deal with the "singularity issue" that arises when
a quadratic function is used to majorize the nondifferentiable surrogate
function. For sparse GEPs with special structure, algorithms that admit a
closed-form solution at every iteration are derived. Numerical experiments show
that the proposed algorithms match or outperform existing algorithms in terms
of computational complexity and support recovery.
|
Non-Standard Words as Features for Text Categorization | This paper presents categorization of Croatian texts using Non-Standard Words
(NSW) as features. Non-Standard Words are: numbers, dates, acronyms,
abbreviations, currency, etc. NSWs in Croatian language are determined
according to Croatian NSW taxonomy. For the purpose of this research, 390 text
documents were collected and formed the SKIPEZ collection with 6 classes:
official, literary, informative, popular, educational and scientific. Text
categorization experiment was conducted on three different representations of
the SKIPEZ collection: in the first representation, the frequencies of NSWs are
used as features; in the second representation, the statistic measures of NSWs
(variance, coefficient of variation, standard deviation, etc.) are used as
features; while the third representation combines the first two feature sets.
Naive Bayes, CN2, C4.5, kNN, Classification Trees and Random Forest algorithms
were used in text categorization experiments. The best categorization results
are achieved using the first feature set (NSW frequencies) with the
categorization accuracy of 87%. This suggests that the NSWs should be
considered as features in highly inflectional languages, such as Croatian. NSW
based features reduce the dimensionality of the feature space without standard
lemmatization procedures, and therefore the bag-of-NSWs should be considered
for further Croatian texts categorization experiments.
|
A Multi-Plane Block-Coordinate Frank-Wolfe Algorithm for Training
Structural SVMs with a Costly max-Oracle | Structural support vector machines (SSVMs) are amongst the best performing
models for structured computer vision tasks, such as semantic image
segmentation or human pose estimation. Training SSVMs, however, is
computationally costly, because it requires repeated calls to a structured
prediction subroutine (called \emph{max-oracle}), which has to solve an
optimization problem itself, e.g. a graph cut.
In this work, we introduce a new algorithm for SSVM training that is more
efficient than earlier techniques when the max-oracle is computationally
expensive, as it is frequently the case in computer vision tasks. The main idea
is to (i) combine the recent stochastic Block-Coordinate Frank-Wolfe algorithm
with efficient hyperplane caching, and (ii) use an automatic selection rule for
deciding whether to call the exact max-oracle or to rely on an approximate one
based on the cached hyperplanes.
We show experimentally that this strategy leads to faster convergence to the
optimum with respect to the number of requires oracle calls, and that this
translates into faster convergence with respect to the total runtime when the
max-oracle is slow compared to the other steps of the algorithm.
A publicly available C++ implementation is provided at
http://pub.ist.ac.at/~vnk/papers/SVM.html .
|
A Plug&Play P300 BCI Using Information Geometry | This paper presents a new classification methods for Event Related Potentials
(ERP) based on an Information geometry framework. Through a new estimation of
covariance matrices, this work extend the use of Riemannian geometry, which was
previously limited to SMR-based BCI, to the problem of classification of ERPs.
As compared to the state-of-the-art, this new method increases performance,
reduces the number of data needed for the calibration and features good
generalisation across sessions and subjects. This method is illustrated on data
recorded with the P300-based game brain invaders. Finally, an online and
adaptive implementation is described, where the BCI is initialized with generic
parameters derived from a database and continuously adapt to the individual,
allowing the user to play the game without any calibration while keeping a high
accuracy.
|
Ad Hoc Microphone Array Calibration: Euclidean Distance Matrix
Completion Algorithm and Theoretical Guarantees | This paper addresses the problem of ad hoc microphone array calibration where
only partial information about the distances between microphones is available.
We construct a matrix consisting of the pairwise distances and propose to
estimate the missing entries based on a novel Euclidean distance matrix
completion algorithm by alternative low-rank matrix completion and projection
onto the Euclidean distance space. This approach confines the recovered matrix
to the EDM cone at each iteration of the matrix completion algorithm. The
theoretical guarantees of the calibration performance are obtained considering
the random and locally structured missing entries as well as the measurement
noise on the known distances. This study elucidates the links between the
calibration error and the number of microphones along with the noise level and
the ratio of missing distances. Thorough experiments on real data recordings
and simulated setups are conducted to demonstrate these theoretical insights. A
significant improvement is achieved by the proposed Euclidean distance matrix
completion algorithm over the state-of-the-art techniques for ad hoc microphone
array calibration.
|
Multi-task Sparse Structure Learning | Multi-task learning (MTL) aims to improve generalization performance by
learning multiple related tasks simultaneously. While sometimes the underlying
task relationship structure is known, often the structure needs to be estimated
from data at hand. In this paper, we present a novel family of models for MTL,
applicable to regression and classification problems, capable of learning the
structure of task relationships. In particular, we consider a joint estimation
problem of the task relationship structure and the individual task parameters,
which is solved using alternating minimization. The task relationship structure
learning component builds on recent advances in structure learning of Gaussian
graphical models based on sparse estimators of the precision (inverse
covariance) matrix. We illustrate the effectiveness of the proposed model on a
variety of synthetic and benchmark datasets for regression and classification.
We also consider the problem of combining climate model outputs for better
projections of future climate, with focus on temperature in South America, and
show that the proposed model outperforms several existing methods for the
problem.
|
Neural Machine Translation by Jointly Learning to Align and Translate | Neural machine translation is a recently proposed approach to machine
translation. Unlike the traditional statistical machine translation, the neural
machine translation aims at building a single neural network that can be
jointly tuned to maximize the translation performance. The models proposed
recently for neural machine translation often belong to a family of
encoder-decoders and consists of an encoder that encodes a source sentence into
a fixed-length vector from which a decoder generates a translation. In this
paper, we conjecture that the use of a fixed-length vector is a bottleneck in
improving the performance of this basic encoder-decoder architecture, and
propose to extend this by allowing a model to automatically (soft-)search for
parts of a source sentence that are relevant to predicting a target word,
without having to form these parts as a hard segment explicitly. With this new
approach, we achieve a translation performance comparable to the existing
state-of-the-art phrase-based system on the task of English-to-French
translation. Furthermore, qualitative analysis reveals that the
(soft-)alignments found by the model agree well with our intuition.
|
Sampling-based Approximations with Quantitative Performance for the
Probabilistic Reach-Avoid Problem over General Markov Processes | This article deals with stochastic processes endowed with the Markov
(memoryless) property and evolving over general (uncountable) state spaces. The
models further depend on a non-deterministic quantity in the form of a control
input, which can be selected to affect the probabilistic dynamics. We address
the computation of maximal reach-avoid specifications, together with the
synthesis of the corresponding optimal controllers. The reach-avoid
specification deals with assessing the likelihood that any finite-horizon
trajectory of the model enters a given goal set, while avoiding a given set of
undesired states. This article newly provides an approximate computational
scheme for the reach-avoid specification based on the Fitted Value Iteration
algorithm, which hinges on random sample extractions, and gives a-priori
computable formal probabilistic bounds on the error made by the approximation
algorithm: as such, the output of the numerical scheme is quantitatively
assessed and thus meaningful for safety-critical applications. Furthermore, we
provide tighter probabilistic error bounds that are sample-based. The overall
computational scheme is put in relationship with alternative approximation
algorithms in the literature, and finally its performance is practically
assessed over a benchmark case study.
|
On the Equivalence Between Deep NADE and Generative Stochastic Networks | Neural Autoregressive Distribution Estimators (NADEs) have recently been
shown as successful alternatives for modeling high dimensional multimodal
distributions. One issue associated with NADEs is that they rely on a
particular order of factorization for $P(\mathbf{x})$. This issue has been
recently addressed by a variant of NADE called Orderless NADEs and its deeper
version, Deep Orderless NADE. Orderless NADEs are trained based on a criterion
that stochastically maximizes $P(\mathbf{x})$ with all possible orders of
factorizations. Unfortunately, ancestral sampling from deep NADE is very
expensive, corresponding to running through a neural net separately predicting
each of the visible variables given some others. This work makes a connection
between this criterion and the training criterion for Generative Stochastic
Networks (GSNs). It shows that training NADEs in this way also trains a GSN,
which defines a Markov chain associated with the NADE model. Based on this
connection, we show an alternative way to sample from a trained Orderless NADE
that allows to trade-off computing time and quality of the samples: a 3 to
10-fold speedup (taking into account the waste due to correlations between
consecutive samples of the chain) can be obtained without noticeably reducing
the quality of the samples. This is achieved using a novel sampling procedure
for GSNs called annealed GSN sampling, similar to tempering methods that
combines fast mixing (obtained thanks to steps at high noise levels) with
accurate samples (obtained thanks to steps at low noise levels).
|
Multi-rank Sparse Hierarchical Clustering | There has been a surge in the number of large and flat data sets - data sets
containing a large number of features and a relatively small number of
observations - due to the growing ability to collect and store information in
medical research and other fields. Hierarchical clustering is a widely used
clustering tool. In hierarchical clustering, large and flat data sets may allow
for a better coverage of clustering features (features that help explain the
true underlying clusters) but, such data sets usually include a large fraction
of noise features (non-clustering features) that may hide the underlying
clusters. Witten and Tibshirani (2010) proposed a sparse hierarchical
clustering framework to cluster the observations using an adaptively chosen
subset of the features, however, we show that this framework has some
limitations when the data sets contain clustering features with complex
structure. In this paper, we propose the Multi-rank sparse hierarchical
clustering (MrSHC). We show that, using simulation studies and real data
examples, MrSHC produces superior feature selection and clustering performance
comparing to the classical (of-the-shelf) hierarchical clustering and the
existing sparse hierarchical clustering framework.
|
Comparison of algorithms that detect drug side effects using electronic
healthcare databases | The electronic healthcare databases are starting to become more readily
available and are thought to have excellent potential for generating adverse
drug reaction signals. The Health Improvement Network (THIN) database is an
electronic healthcare database containing medical information on over 11
million patients that has excellent potential for detecting ADRs. In this paper
we apply four existing electronic healthcare database signal detecting
algorithms (MUTARA, HUNT, Temporal Pattern Discovery and modified ROR) on the
THIN database for a selection of drugs from six chosen drug families. This is
the first comparison of ADR signalling algorithms that includes MUTARA and HUNT
and enabled us to set a benchmark for the adverse drug reaction signalling
ability of the THIN database. The drugs were selectively chosen to enable a
comparison with previous work and for variety. It was found that no algorithm
was generally superior and the algorithms' natural thresholds act at variable
stringencies. Furthermore, none of the algorithms perform well at detecting
rare ADRs.
|
Data classification using the Dempster-Shafer method | In this paper, the Dempster-Shafer method is employed as the theoretical
basis for creating data classification systems. Testing is carried out using
three popular (multiple attribute) benchmark datasets that have two, three and
four classes. In each case, a subset of the available data is used for training
to establish thresholds, limits or likelihoods of class membership for each
attribute, and hence create mass functions that establish probability of class
membership for each attribute of the test data. Classification of each data
item is achieved by combination of these probabilities via Dempster's Rule of
Combination. Results for the first two datasets show extremely high
classification accuracy that is competitive with other popular methods. The
third dataset is non-numerical and difficult to classify, but good results can
be achieved provided the system and mass functions are designed carefully and
the right attributes are chosen for combination. In all cases the
Dempster-Shafer method provides comparable performance to other more popular
algorithms, but the overhead of generating accurate mass functions increases
the complexity with the addition of new attributes. Overall, the results
suggest that the D-S approach provides a suitable framework for the design of
classification systems and that automating the mass function design and
calculation would increase the viability of the algorithm for complex
classification problems.
|
A Novel Semi-Supervised Algorithm for Rare Prescription Side Effect
Discovery | Drugs are frequently prescribed to patients with the aim of improving each
patient's medical state, but an unfortunate consequence of most prescription
drugs is the occurrence of undesirable side effects. Side effects that occur in
more than one in a thousand patients are likely to be signalled efficiently by
current drug surveillance methods, however, these same methods may take decades
before generating signals for rarer side effects, risking medical morbidity or
mortality in patients prescribed the drug while the rare side effect is
undiscovered. In this paper we propose a novel computational meta-analysis
framework for signalling rare side effects that integrates existing methods,
knowledge from the web, metric learning and semi-supervised clustering. The
novel framework was able to signal many known rare and serious side effects for
the selection of drugs investigated, such as tendon rupture when prescribed
Ciprofloxacin or Levofloxacin, renal failure with Naproxen and depression
associated with Rimonabant. Furthermore, for the majority of the drug
investigated it generated signals for rare side effects at a more stringent
signalling threshold than existing methods and shows the potential to become a
fundamental part of post marketing surveillance to detect rare side effects.
|
Signalling Paediatric Side Effects using an Ensemble of Simple Study
Designs | Background: Children are frequently prescribed medication off-label, meaning
there has not been sufficient testing of the medication to determine its safety
or effectiveness. The main reason this safety knowledge is lacking is due to
ethical restrictions that prevent children from being included in the majority
of clinical trials. Objective: The objective of this paper is to investigate
whether an ensemble of simple study designs can be implemented to signal
acutely occurring side effects effectively within the paediatric population by
using historical longitudinal data. The majority of pharmacovigilance
techniques are unsupervised, but this research presents a supervised framework.
Methods: Multiple measures of association are calculated for each drug and
medical event pair and these are used as features that are fed into a
classiffier to determine the likelihood of the drug and medical event pair
corresponding to an adverse drug reaction. The classiffier is trained using
known adverse drug reactions or known non-adverse drug reaction relationships.
Results: The novel ensemble framework obtained a false positive rate of 0:149,
a sensitivity of 0:547 and a specificity of 0:851 when implemented on a
reference set of drug and medical event pairs. The novel framework consistently
outperformed each individual simple study design. Conclusion: This research
shows that it is possible to exploit the mechanism of causality and presents a
framework for signalling adverse drug reactions effectively.
|
Feature selection in detection of adverse drug reactions from the Health
Improvement Network (THIN) database | Adverse drug reaction (ADR) is widely concerned for public health issue. ADRs
are one of most common causes to withdraw some drugs from market. Prescription
event monitoring (PEM) is an important approach to detect the adverse drug
reactions. The main problem to deal with this method is how to automatically
extract the medical events or side effects from high-throughput medical events,
which are collected from day to day clinical practice. In this study we propose
a novel concept of feature matrix to detect the ADRs. Feature matrix, which is
extracted from big medical data from The Health Improvement Network (THIN)
database, is created to characterize the medical events for the patients who
take drugs. Feature matrix builds the foundation for the irregular and big
medical data. Then feature selection methods are performed on feature matrix to
detect the significant features. Finally the ADRs can be located based on the
significant features. The experiments are carried out on three drugs:
Atorvastatin, Alendronate, and Metoclopramide. Major side effects for each drug
are detected and better performance is achieved compared to other computerized
methods. The detected ADRs are based on computerized methods, further
investigation is needed.
|
Ensemble Learning of Colorectal Cancer Survival Rates | In this paper, we describe a dataset relating to cellular and physical
conditions of patients who are operated upon to remove colorectal tumours. This
data provides a unique insight into immunological status at the point of tumour
removal, tumour classification and post-operative survival. We build on
existing research on clustering and machine learning facets of this data to
demonstrate a role for an ensemble approach to highlighting patients with
clearer prognosis parameters. Results for survival prediction using 3 different
approaches are shown for a subset of the data which is most difficult to model.
The performance of each model individually is compared with subsets of the data
where some agreement is reached for multiple models. Significant improvements
in model accuracy on an unseen test set can be achieved for patients where
agreement between models is achieved.
|
Feature Selection in Conditional Random Fields for Map Matching of GPS
Trajectories | Map matching of the GPS trajectory serves the purpose of recovering the
original route on a road network from a sequence of noisy GPS observations. It
is a fundamental technique to many Location Based Services. However, map
matching of a low sampling rate on urban road network is still a challenging
task. In this paper, the characteristics of Conditional Random Fields with
regard to inducing many contextual features and feature selection are explored
for the map matching of the GPS trajectories at a low sampling rate.
Experiments on a taxi trajectory dataset show that our method may achieve
competitive results along with the success of reducing model complexity for
computation-limited applications.
|
Feature Engineering for Map Matching of Low-Sampling-Rate GPS
Trajectories in Road Network | Map matching of GPS trajectories from a sequence of noisy observations serves
the purpose of recovering the original routes in a road network. In this work
in progress, we attempt to share our experience of feature construction in a
spatial database by reporting our ongoing experiment of feature extrac-tion in
Conditional Random Fields (CRFs) for map matching. Our preliminary results are
obtained from real-world taxi GPS trajectories.
|
Solving the Problem of the K Parameter in the KNN Classifier Using an
Ensemble Learning Approach | This paper presents a new solution for choosing the K parameter in the
k-nearest neighbor (KNN) algorithm, the solution depending on the idea of
ensemble learning, in which a weak KNN classifier is used each time with a
different K, starting from one to the square root of the size of the training
set. The results of the weak classifiers are combined using the weighted sum
rule. The proposed solution was tested and compared to other solutions using a
group of experiments in real life problems. The experimental results show that
the proposed classifier outperforms the traditional KNN classifier that uses a
different number of neighbors, is competitive with other classifiers, and is a
promising classifier with strong potential for a wide range of applications.
|
Dimensionality Invariant Similarity Measure | This paper presents a new similarity measure to be used for general tasks
including supervised learning, which is represented by the K-nearest neighbor
classifier (KNN). The proposed similarity measure is invariant to large
differences in some dimensions in the feature space. The proposed metric is
proved mathematically to be a metric. To test its viability for different
applications, the KNN used the proposed metric for classifying test examples
chosen from a number of real datasets. Compared to some other well known
metrics, the experimental results show that the proposed metric is a promising
distance measure for the KNN classifier with strong potential for a wide range
of applications.
|
Breakdown Point of Robust Support Vector Machine | The support vector machine (SVM) is one of the most successful learning
methods for solving classification problems. Despite its popularity, SVM has a
serious drawback, that is sensitivity to outliers in training samples. The
penalty on misclassification is defined by a convex loss called the hinge loss,
and the unboundedness of the convex loss causes the sensitivity to outliers. To
deal with outliers, robust variants of SVM have been proposed, such as the
robust outlier detection algorithm and an SVM with a bounded loss called the
ramp loss. In this paper, we propose a robust variant of SVM and investigate
its robustness in terms of the breakdown point. The breakdown point is a
robustness measure that is the largest amount of contamination such that the
estimated classifier still gives information about the non-contaminated data.
The main contribution of this paper is to show an exact evaluation of the
breakdown point for the robust SVM. For learning parameters such as the
regularization parameter in our algorithm, we derive a simple formula that
guarantees the robustness of the classifier. When the learning parameters are
determined with a grid search using cross validation, our formula works to
reduce the number of candidate search points. The robustness of the proposed
method is confirmed in numerical experiments. We show that the statistical
properties of the robust SVM are well explained by a theoretical analysis of
the breakdown point.
|
High-performance Kernel Machines with Implicit Distributed Optimization
and Randomization | In order to fully utilize "big data", it is often required to use "big
models". Such models tend to grow with the complexity and size of the training
data, and do not make strong parametric assumptions upfront on the nature of
the underlying statistical dependencies. Kernel methods fit this need well, as
they constitute a versatile and principled statistical methodology for solving
a wide range of non-parametric modelling problems. However, their high
computational costs (in storage and time) pose a significant barrier to their
widespread adoption in big data applications.
We propose an algorithmic framework and high-performance implementation for
massive-scale training of kernel-based statistical models, based on combining
two key technical ingredients: (i) distributed general purpose convex
optimization, and (ii) the use of randomization to improve the scalability of
kernel methods. Our approach is based on a block-splitting variant of the
Alternating Directions Method of Multipliers, carefully reconfigured to handle
very large random feature matrices, while exploiting hybrid parallelism
typically found in modern clusters of multicore machines. Our implementation
supports a variety of statistical learning tasks by enabling several loss
functions, regularization schemes, kernels, and layers of randomized
approximations for both dense and sparse datasets, in a highly extensible
framework. We evaluate the ability of our framework to learn models on data
from applications, and provide a comparison against existing sequential and
parallel libraries.
|
Constructing a Non-Negative Low Rank and Sparse Graph with Data-Adaptive
Features | This paper aims at constructing a good graph for discovering intrinsic data
structures in a semi-supervised learning setting. Firstly, we propose to build
a non-negative low-rank and sparse (referred to as NNLRS) graph for the given
data representation. Specifically, the weights of edges in the graph are
obtained by seeking a nonnegative low-rank and sparse matrix that represents
each data sample as a linear combination of others. The so-obtained NNLRS-graph
can capture both the global mixture of subspaces structure (by the low
rankness) and the locally linear structure (by the sparseness) of the data,
hence is both generative and discriminative. Secondly, as good features are
extremely important for constructing a good graph, we propose to learn the data
embedding matrix and construct the graph jointly within one framework, which is
termed as NNLRS with embedded features (referred to as NNLRS-EF). Extensive
experiments on three publicly available datasets demonstrate that the proposed
method outperforms the state-of-the-art graph construction method by a large
margin for both semi-supervised classification and discriminative analysis,
which verifies the effectiveness of our proposed method.
|
Variability of Behaviour in Electricity Load Profile Clustering; Who
Does Things at the Same Time Each Day? | UK electricity market changes provide opportunities to alter households'
electricity usage patterns for the benefit of the overall electricity network.
Work on clustering similar households has concentrated on daily load profiles
and the variability in regular household behaviours has not been considered.
Those households with most variability in regular activities may be the most
receptive to incentives to change timing.
Whether using the variability of regular behaviour allows the creation of
more consistent groupings of households is investigated and compared with daily
load profile clustering. 204 UK households are analysed to find repeating
patterns (motifs). Variability in the time of the motif is used as the basis
for clustering households. Different clustering algorithms are assessed by the
consistency of the results.
Findings show that variability of behaviour, using motifs, provides more
consistent groupings of households across different clustering algorithms and
allows for more efficient targeting of behaviour change interventions.
|
Tuning a Multiple Classifier System for Side Effect Discovery using
Genetic Algorithms | In previous work, a novel supervised framework implementing a binary
classifier was presented that obtained excellent results for side effect
discovery. Interestingly, unique side effects were identified when different
binary classifiers were used within the framework, prompting the investigation
of applying a multiple classifier system. In this paper we investigate tuning a
side effect multiple classifying system using genetic algorithms. The results
of this research show that the novel framework implementing a multiple
classifying system trained using genetic algorithms can obtain a higher partial
area under the receiver operating characteristic curve than implementing a
single classifier. Furthermore, the framework is able to detect side effects
efficiently and obtains a low false positive rate.
|
Augmented Neural Networks for Modelling Consumer Indebtness | Consumer Debt has risen to be an important problem of modern societies,
generating a lot of research in order to understand the nature of consumer
indebtness, which so far its modelling has been carried out by statistical
models. In this work we show that Computational Intelligence can offer a more
holistic approach that is more suitable for the complex relationships an
indebtness dataset has and Linear Regression cannot uncover. In particular, as
our results show, Neural Networks achieve the best performance in modelling
consumer indebtness, especially when they manage to incorporate the significant
and experimentally verified results of the Data Mining process in the model,
exploiting the flexibility Neural Networks offer in designing their topology.
This novel method forms an elaborate framework to model Consumer indebtness
that can be extended to any other real world application.
|
Structured Low-Rank Matrix Factorization with Missing and Grossly
Corrupted Observations | Recovering low-rank and sparse matrices from incomplete or corrupted
observations is an important problem in machine learning, statistics,
bioinformatics, computer vision, as well as signal and image processing. In
theory, this problem can be solved by the natural convex joint/mixed
relaxations (i.e., l_{1}-norm and trace norm) under certain conditions.
However, all current provable algorithms suffer from superlinear per-iteration
cost, which severely limits their applicability to large-scale problems. In
this paper, we propose a scalable, provable structured low-rank matrix
factorization method to recover low-rank and sparse matrices from missing and
grossly corrupted data, i.e., robust matrix completion (RMC) problems, or
incomplete and grossly corrupted measurements, i.e., compressive principal
component pursuit (CPCP) problems. Specifically, we first present two
small-scale matrix trace norm regularized bilinear structured factorization
models for RMC and CPCP problems, in which repetitively calculating SVD of a
large-scale matrix is replaced by updating two much smaller factor matrices.
Then, we apply the alternating direction method of multipliers (ADMM) to
efficiently solve the RMC problems. Finally, we provide the convergence
analysis of our algorithm, and extend it to address general CPCP problems.
Experimental results verified both the efficiency and effectiveness of our
method compared with the state-of-the-art methods.
|
Domain Transfer Structured Output Learning | In this paper, we propose the problem of domain transfer structured output
learn- ing and the first solution to solve it. The problem is defined on two
different data domains sharing the same input and output spaces, named as
source domain and target domain. The outputs are structured, and for the data
samples of the source domain, the corresponding outputs are available, while
for most data samples of the target domain, the corresponding outputs are
missing. The input distributions of the two domains are significantly
different. The problem is to learn a predictor for the target domain to predict
the structured outputs from the input. Due to the limited number of outputs
available for the samples form the target domain, it is difficult to directly
learn the predictor from the target domain, thus it is necessary to use the
output information available in source domain. We propose to learn the target
domain predictor by adapting a auxiliary predictor trained by using source
domain data to the target domain. The adaptation is implemented by adding a
delta function on the basis of the auxiliary predictor. An algorithm is
developed to learn the parameter of the delta function to minimize loss
functions associat- ed with the predicted outputs against the true outputs of
the data samples with available outputs of the target domain.
|
Overcoming the Curse of Sentence Length for Neural Machine Translation
using Automatic Segmentation | The authors of (Cho et al., 2014a) have shown that the recently introduced
neural network translation systems suffer from a significant drop in
translation quality when translating long sentences, unlike existing
phrase-based translation systems. In this paper, we propose a way to address
this issue by automatically segmenting an input sentence into phrases that can
be easily translated by the neural network translation model. Once each segment
has been independently translated by the neural machine translation model, the
translated clauses are concatenated to form a final translation. Empirical
results show a significant improvement in translation quality for long
sentences.
|
Marginal Structured SVM with Hidden Variables | In this work, we propose the marginal structured SVM (MSSVM) for structured
prediction with hidden variables. MSSVM properly accounts for the uncertainty
of hidden variables, and can significantly outperform the previously proposed
latent structured SVM (LSSVM; Yu & Joachims (2009)) and other state-of-art
methods, especially when that uncertainty is large. Our method also results in
a smoother objective function, making gradient-based optimization of MSSVMs
converge significantly faster than for LSSVMs. We also show that our method
consistently outperforms hidden conditional random fields (HCRFs; Quattoni et
al. (2007)) on both simulated and real-world datasets. Furthermore, we propose
a unified framework that includes both our and several other existing methods
as special cases, and provides insights into the comparison of different models
in practice.
|
Communication-Efficient Distributed Dual Coordinate Ascent | Communication remains the most significant bottleneck in the performance of
distributed optimization algorithms for large-scale machine learning. In this
paper, we propose a communication-efficient framework, CoCoA, that uses local
computation in a primal-dual setting to dramatically reduce the amount of
necessary communication. We provide a strong convergence rate analysis for this
class of algorithms, as well as experiments on real-world distributed datasets
with implementations in Spark. In our experiments, we find that as compared to
state-of-the-art mini-batch versions of SGD and SDCA algorithms, CoCoA
converges to the same .001-accurate solution quality on average 25x as quickly.
|
Machine Learning Etudes in Astrophysics: Selection Functions for Mock
Cluster Catalogs | Making mock simulated catalogs is an important component of astrophysical
data analysis. Selection criteria for observed astronomical objects are often
too complicated to be derived from first principles. However the existence of
an observed group of objects is a well-suited problem for machine learning
classification. In this paper we use one-class classifiers to learn the
properties of an observed catalog of clusters of galaxies from ROSAT and to
pick clusters from mock simulations that resemble the observed ROSAT catalog.
We show how this method can be used to study the cross-correlations of thermal
Sunya'ev-Zeldovich signals with number density maps of X-ray selected cluster
catalogs. The method reduces the bias due to hand-tuning the selection function
and is readily scalable to large catalogs with a high-dimensional space of
astrophysical features.
|
Novel Methods for Activity Classification and Occupany Prediction
Enabling Fine-grained HVAC Control | Much of the energy consumption in buildings is due to HVAC systems, which has
motivated several recent studies on making these systems more energy-
efficient. Occupancy and activity are two important aspects, which need to be
correctly estimated for optimal HVAC control. However, state-of-the-art methods
to estimate occupancy and classify activity require infrastructure and/or
wearable sensors which suffers from lower acceptability due to higher cost.
Encouragingly, with the advancement of the smartphones, these are becoming more
achievable. Most of the existing occupancy estimation tech- niques have the
underlying assumption that the phone is always carried by its user. However,
phones are often left at desk while attending meeting or other events, which
generates estimation error for the existing phone based occupancy algorithms.
Similarly, in the recent days the emerging theory of Sparse Random Classifier
(SRC) has been applied for activity classification on smartphone, however,
there are rooms to improve the on-phone process- ing. We propose a novel sensor
fusion method which offers almost 100% accuracy for occupancy estimation. We
also propose an activity classifica- tion algorithm, which offers similar
accuracy as of the state-of-the-art SRC algorithms while offering 50% reduction
in processing.
|
A Reduction of the Elastic Net to Support Vector Machines with an
Application to GPU Computing | The past years have witnessed many dedicated open-source projects that built
and maintain implementations of Support Vector Machines (SVM), parallelized for
GPU, multi-core CPUs and distributed systems. Up to this point, no comparable
effort has been made to parallelize the Elastic Net, despite its popularity in
many high impact applications, including genetics, neuroscience and systems
biology. The first contribution in this paper is of theoretical nature. We
establish a tight link between two seemingly different algorithms and prove
that Elastic Net regression can be reduced to SVM with squared hinge loss
classification. Our second contribution is to derive a practical algorithm
based on this reduction. The reduction enables us to utilize prior efforts in
speeding up and parallelizing SVMs to obtain a highly optimized and parallel
solver for the Elastic Net and Lasso. With a simple wrapper, consisting of only
11 lines of MATLAB code, we obtain an Elastic Net implementation that naturally
utilizes GPU and multi-core CPUs. We demonstrate on twelve real world data
sets, that our algorithm yields identical results as the popular (and highly
optimized) glmnet implementation but is one or several orders of magnitude
faster.
|
Global Convergence of Online Limited Memory BFGS | Global convergence of an online (stochastic) limited memory version of the
Broyden-Fletcher- Goldfarb-Shanno (BFGS) quasi-Newton method for solving
optimization problems with stochastic objectives that arise in large scale
machine learning is established. Lower and upper bounds on the Hessian
eigenvalues of the sample functions are shown to suffice to guarantee that the
curvature approximation matrices have bounded determinants and traces, which,
in turn, permits establishing convergence to optimal arguments with probability
1. Numerical experiments on support vector machines with synthetic data
showcase reductions in convergence time relative to stochastic gradient descent
algorithms as well as reductions in storage and computation relative to other
online quasi-Newton methods. Experimental evaluation on a search engine
advertising problem corroborates that these advantages also manifest in
practical applications.
|
The Large Margin Mechanism for Differentially Private Maximization | A basic problem in the design of privacy-preserving algorithms is the private
maximization problem: the goal is to pick an item from a universe that
(approximately) maximizes a data-dependent function, all under the constraint
of differential privacy. This problem has been used as a sub-routine in many
privacy-preserving algorithms for statistics and machine-learning.
Previous algorithms for this problem are either range-dependent---i.e., their
utility diminishes with the size of the universe---or only apply to very
restricted function classes. This work provides the first general-purpose,
range-independent algorithm for private maximization that guarantees
approximate differential privacy. Its applicability is demonstrated on two
fundamental tasks in data mining and machine learning.
|
When coding meets ranking: A joint framework based on local learning | Sparse coding, which represents a data point as a sparse reconstruction code
with regard to a dictionary, has been a popular data representation method.
Meanwhile, in database retrieval problems, learning the ranking scores from
data points plays an important role. Up to now, these two problems have always
been considered separately, assuming that data coding and ranking are two
independent and irrelevant problems. However, is there any internal
relationship between sparse coding and ranking score learning? If yes, how to
explore and make use of this internal relationship? In this paper, we try to
answer these questions by developing the first joint sparse coding and ranking
score learning algorithm. To explore the local distribution in the sparse code
space, and also to bridge coding and ranking problems, we assume that in the
neighborhood of each data point, the ranking scores can be approximated from
the corresponding sparse codes by a local linear function. By considering the
local approximation error of ranking scores, the reconstruction error and
sparsity of sparse coding, and the query information provided by the user, we
construct a unified objective function for learning of sparse codes, the
dictionary and ranking scores. We further develop an iterative algorithm to
solve this optimization problem.
|
Variational Inference for Uncertainty on the Inputs of Gaussian Process
Models | The Gaussian process latent variable model (GP-LVM) provides a flexible
approach for non-linear dimensionality reduction that has been widely applied.
However, the current approach for training GP-LVMs is based on maximum
likelihood, where the latent projection variables are maximized over rather
than integrated out. In this paper we present a Bayesian method for training
GP-LVMs by introducing a non-standard variational inference framework that
allows to approximately integrate out the latent variables and subsequently
train a GP-LVM by maximizing an analytic lower bound on the exact marginal
likelihood. We apply this method for learning a GP-LVM from iid observations
and for learning non-linear dynamical systems where the observations are
temporally correlated. We show that a benefit of the variational Bayesian
procedure is its robustness to overfitting and its ability to automatically
select the dimensionality of the nonlinear latent space. The resulting
framework is generic, flexible and easy to extend for other purposes, such as
Gaussian process regression with uncertain inputs and semi-supervised Gaussian
processes. We demonstrate our method on synthetic data and standard machine
learning benchmarks, as well as challenging real world datasets, including high
resolution video data.
|
Sparse Additive Model using Symmetric Nonnegative Definite Smoothers | We introduce a new algorithm, called adaptive sparse backfitting algorithm,
for solving high dimensional Sparse Additive Model (SpAM) utilizing symmetric,
non-negative definite smoothers. Unlike the previous sparse backfitting
algorithm, our method is essentially a block coordinate descent algorithm that
guarantees to converge to the optimal solution. It bridges the gap between the
population backfitting algorithm and that of the data version. We also prove
variable selection consistency under suitable conditions. Numerical studies on
both synthesis and real data are conducted to show that adaptive sparse
backfitting algorithm outperforms previous sparse backfitting algorithm in
fitting and predicting high dimensional nonparametric models.
|
Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures | Model-based methods and deep neural networks have both been tremendously
successful paradigms in machine learning. In model-based methods, problem
domain knowledge can be built into the constraints of the model, typically at
the expense of difficulties during inference. In contrast, deterministic deep
neural networks are constructed in such a way that inference is
straightforward, but their architectures are generic and it is unclear how to
incorporate knowledge. This work aims to obtain the advantages of both
approaches. To do so, we start with a model-based approach and an associated
inference algorithm, and \emph{unfold} the inference iterations as layers in a
deep network. Rather than optimizing the original model, we \emph{untie} the
model parameters across layers, in order to create a more powerful network. The
resulting architecture can be trained discriminatively to perform accurate
inference within a fixed network size. We show how this framework allows us to
interpret conventional networks as mean-field inference in Markov random
fields, and to obtain new architectures by instead using belief propagation as
the inference algorithm. We then show its application to a non-negative matrix
factorization model that incorporates the problem-domain knowledge that sound
sources are additive. Deep unfolding of this model yields a new kind of
non-negative deep neural network, that can be trained using a multiplicative
backpropagation-style update algorithm. We present speech enhancement
experiments showing that our approach is competitive with conventional neural
networks despite using far fewer parameters.
|
A theoretical contribution to the fast implementation of null linear
discriminant analysis method using random matrix multiplication with scatter
matrices | The null linear discriminant analysis method is a competitive approach for
dimensionality reduction. The implementation of this method, however, is
computationally expensive. Recently, a fast implementation of null linear
discriminant analysis method using random matrix multiplication with scatter
matrices was proposed. However, if the random matrix is chosen arbitrarily, the
orientation matrix may be rank deficient, and some useful discriminant
information will be lost. In this paper, we investigate how to choose the
random matrix properly, such that the two criteria of the null LDA method are
satisfied theoretically. We give a necessary and sufficient condition to
guarantee full column rank of the orientation matrix. Moreover, the geometric
characterization of the condition is also described.
|
Learning Machines Implemented on Non-Deterministic Hardware | This paper highlights new opportunities for designing large-scale machine
learning systems as a consequence of blurring traditional boundaries that have
allowed algorithm designers and application-level practitioners to stay -- for
the most part -- oblivious to the details of the underlying hardware-level
implementations. The hardware/software co-design methodology advocated here
hinges on the deployment of compute-intensive machine learning kernels onto
compute platforms that trade-off determinism in the computation for improvement
in speed and/or energy efficiency. To achieve this, we revisit digital
stochastic circuits for approximating matrix computations that are ubiquitous
in machine learning algorithms. Theoretical and empirical evaluation is
undertaken to assess the impact of the hardware-induced computational noise on
algorithm performance. As a proof-of-concept, a stochastic hardware simulator
is employed for training deep neural networks for image recognition problems.
|
Weighted Classification Cascades for Optimizing Discovery Significance
in the HiggsML Challenge | We introduce a minorization-maximization approach to optimizing common
measures of discovery significance in high energy physics. The approach
alternates between solving a weighted binary classification problem and
updating class weights in a simple, closed-form manner. Moreover, an argument
based on convex duality shows that an improvement in weighted classification
error on any round yields a commensurate improvement in discovery significance.
We complement our derivation with experimental results from the 2014 Higgs
boson machine learning challenge.
|
Winner-Take-All Autoencoders | In this paper, we propose a winner-take-all method for learning hierarchical
sparse representations in an unsupervised fashion. We first introduce
fully-connected winner-take-all autoencoders which use mini-batch statistics to
directly enforce a lifetime sparsity in the activations of the hidden units. We
then propose the convolutional winner-take-all autoencoder which combines the
benefits of convolutional architectures and autoencoders for learning
shift-invariant sparse representations. We describe a way to train
convolutional autoencoders layer by layer, where in addition to lifetime
sparsity, a spatial sparsity within each feature map is achieved using
winner-take-all activation functions. We will show that winner-take-all
autoencoders can be used to to learn deep sparse representations from the
MNIST, CIFAR-10, ImageNet, Street View House Numbers and Toronto Face datasets,
and achieve competitive classification performance.
|
Far-Field Compression for Fast Kernel Summation Methods in High
Dimensions | We consider fast kernel summations in high dimensions: given a large set of
points in $d$ dimensions (with $d \gg 3$) and a pair-potential function (the
{\em kernel} function), we compute a weighted sum of all pairwise kernel
interactions for each point in the set. Direct summation is equivalent to a
(dense) matrix-vector multiplication and scales quadratically with the number
of points. Fast kernel summation algorithms reduce this cost to log-linear or
linear complexity.
Treecodes and Fast Multipole Methods (FMMs) deliver tremendous speedups by
constructing approximate representations of interactions of points that are far
from each other. In algebraic terms, these representations correspond to
low-rank approximations of blocks of the overall interaction matrix. Existing
approaches require an excessive number of kernel evaluations with increasing
$d$ and number of points in the dataset.
To address this issue, we use a randomized algebraic approach in which we
first sample the rows of a block and then construct its approximate, low-rank
interpolative decomposition. We examine the feasibility of this approach
theoretically and experimentally. We provide a new theoretical result showing a
tighter bound on the reconstruction error from uniformly sampling rows than the
existing state-of-the-art. We demonstrate that our sampling approach is
competitive with existing (but prohibitively expensive) methods from the
literature. We also construct kernel matrices for the Laplacian, Gaussian, and
polynomial kernels -- all commonly used in physics and data analysis. We
explore the numerical properties of blocks of these matrices, and show that
they are amenable to our approach. Depending on the data set, our randomized
algorithm can successfully compute low rank approximations in high dimensions.
We report results for data sets with ambient dimensions from four to 1,000.
|
A Stochastic PCA and SVD Algorithm with an Exponential Convergence Rate | We describe and analyze a simple algorithm for principal component analysis
and singular value decomposition, VR-PCA, which uses computationally cheap
stochastic iterations, yet converges exponentially fast to the optimal
solution. In contrast, existing algorithms suffer either from slow convergence,
or computationally intensive iterations whose runtime scales with the data
size. The algorithm builds on a recent variance-reduced stochastic gradient
technique, which was previously analyzed for strongly convex optimization,
whereas here we apply it to an inherently non-convex problem, using a very
different analysis.
|
Non-Convex Boosting Overcomes Random Label Noise | The sensitivity of Adaboost to random label noise is a well-studied problem.
LogitBoost, BrownBoost and RobustBoost are boosting algorithms claimed to be
less sensitive to noise than AdaBoost. We present the results of experiments
evaluating these algorithms on both synthetic and real datasets. We compare the
performance on each of datasets when the labels are corrupted by different
levels of independent label noise. In presence of random label noise, we found
that BrownBoost and RobustBoost perform significantly better than AdaBoost and
LogitBoost, while the difference between each pair of algorithms is
insignificant. We provide an explanation for the difference based on the margin
distributions of the algorithms.
|
Collaborative Deep Learning for Recommender Systems | Collaborative filtering (CF) is a successful approach commonly used by many
recommender systems. Conventional CF-based methods use the ratings given to
items by users as the sole source of information for learning to make
recommendation. However, the ratings are often very sparse in many
applications, causing CF-based methods to degrade significantly in their
recommendation performance. To address this sparsity problem, auxiliary
information such as item content information may be utilized. Collaborative
topic regression (CTR) is an appealing recent method taking this approach which
tightly couples the two components that learn from two different sources of
information. Nevertheless, the latent representation learned by CTR may not be
very effective when the auxiliary information is very sparse. To address this
problem, we generalize recent advances in deep learning from i.i.d. input to
non-i.i.d. (CF-based) input and propose in this paper a hierarchical Bayesian
model called collaborative deep learning (CDL), which jointly performs deep
representation learning for the content information and collaborative filtering
for the ratings (feedback) matrix. Extensive experiments on three real-world
datasets from different domains show that CDL can significantly advance the
state of the art.
|
"Look Ma, No Hands!" A Parameter-Free Topic Model | It has always been a burden to the users of statistical topic models to
predetermine the right number of topics, which is a key parameter of most topic
models. Conventionally, automatic selection of this parameter is done through
either statistical model selection (e.g., cross-validation, AIC, or BIC) or
Bayesian nonparametric models (e.g., hierarchical Dirichlet process). These
methods either rely on repeated runs of the inference algorithm to search
through a large range of parameter values which does not suit the mining of big
data, or replace this parameter with alternative parameters that are less
intuitive and still hard to be determined. In this paper, we explore to
"eliminate" this parameter from a new perspective. We first present a
nonparametric treatment of the PLSA model named nonparametric probabilistic
latent semantic analysis (nPLSA). The inference procedure of nPLSA allows for
the exploration and comparison of different numbers of topics within a single
execution, yet remains as simple as that of PLSA. This is achieved by
substituting the parameter of the number of topics with an alternative
parameter that is the minimal goodness of fit of a document. We show that the
new parameter can be further eliminated by two parameter-free treatments:
either by monitoring the diversity among the discovered topics or by a weak
supervision from users in the form of an exemplar topic. The parameter-free
topic model finds the appropriate number of topics when the diversity among the
discovered topics is maximized, or when the granularity of the discovered
topics matches the exemplar topic. Experiments on both synthetic and real data
prove that the parameter-free topic model extracts topics with a comparable
quality comparing to classical topic models with "manual transmission". The
quality of the topics outperforms those extracted through classical Bayesian
nonparametric models.
|
Towards Optimal Algorithms for Prediction with Expert Advice | We study the classical problem of prediction with expert advice in the
adversarial setting with a geometric stopping time. In 1965, Cover gave the
optimal algorithm for the case of 2 experts. In this paper, we design the
optimal algorithm, adversary and regret for the case of 3 experts. Further, we
show that the optimal algorithm for $2$ and $3$ experts is a probability
matching algorithm (analogous to Thompson sampling) against a particular
randomized adversary. Remarkably, our proof shows that the probability matching
algorithm is not only optimal against this particular randomized adversary, but
also minimax optimal.
Our analysis develops upper and lower bounds simultaneously, analogous to the
primal-dual method. Our analysis of the optimal adversary goes through delicate
asymptotics of the random walk of a particle between multiple walls. We use the
connection we develop to random walks to derive an improved algorithm and
regret bound for the case of $4$ experts, and, provide a general framework for
designing the optimal algorithm and adversary for an arbitrary number of
experts.
|
Metric Learning for Temporal Sequence Alignment | In this paper, we propose to learn a Mahalanobis distance to perform
alignment of multivariate time series. The learning examples for this task are
time series for which the true alignment is known. We cast the alignment
problem as a structured prediction task, and propose realistic losses between
alignments for which the optimization is tractable. We provide experiments on
real data in the audio to audio context, where we show that the learning of a
similarity measure leads to improvements in the performance of the alignment
task. We also propose to use this metric learning framework to perform feature
selection and, from basic audio features, build a combination of these with
better performance for the alignment.
|
Sequence to Sequence Learning with Neural Networks | Deep Neural Networks (DNNs) are powerful models that have achieved excellent
performance on difficult learning tasks. Although DNNs work well whenever large
labeled training sets are available, they cannot be used to map sequences to
sequences. In this paper, we present a general end-to-end approach to sequence
learning that makes minimal assumptions on the sequence structure. Our method
uses a multilayered Long Short-Term Memory (LSTM) to map the input sequence to
a vector of a fixed dimensionality, and then another deep LSTM to decode the
target sequence from the vector. Our main result is that on an English to
French translation task from the WMT'14 dataset, the translations produced by
the LSTM achieve a BLEU score of 34.8 on the entire test set, where the LSTM's
BLEU score was penalized on out-of-vocabulary words. Additionally, the LSTM did
not have difficulty on long sentences. For comparison, a phrase-based SMT
system achieves a BLEU score of 33.3 on the same dataset. When we used the LSTM
to rerank the 1000 hypotheses produced by the aforementioned SMT system, its
BLEU score increases to 36.5, which is close to the previous best result on
this task. The LSTM also learned sensible phrase and sentence representations
that are sensitive to word order and are relatively invariant to the active and
the passive voice. Finally, we found that reversing the order of the words in
all source sentences (but not target sentences) improved the LSTM's performance
markedly, because doing so introduced many short term dependencies between the
source and the target sentence which made the optimization problem easier.
|
Building Program Vector Representations for Deep Learning | Deep learning has made significant breakthroughs in various fields of
artificial intelligence. Advantages of deep learning include the ability to
capture highly complicated features, weak involvement of human engineering,
etc. However, it is still virtually impossible to use deep learning to analyze
programs since deep architectures cannot be trained effectively with pure back
propagation. In this pioneering paper, we propose the "coding criterion" to
build program vector representations, which are the premise of deep learning
for program analysis. Our representation learning approach directly makes deep
learning a reality in this new field. We evaluate the learned vector
representations both qualitatively and quantitatively. We conclude, based on
the experiments, the coding criterion is successful in building program
representations. To evaluate whether deep learning is beneficial for program
analysis, we feed the representations to deep neural networks, and achieve
higher accuracy in the program classification task than "shallow" methods, such
as logistic regression and the support vector machine. This result confirms the
feasibility of deep learning to analyze programs. It also gives primary
evidence of its success in this new field. We believe deep learning will become
an outstanding technique for program analysis in the near future.
|
Consensus-Based Modelling using Distributed Feature Construction | A particularly successful role for Inductive Logic Programming (ILP) is as a
tool for discovering useful relational features for subsequent use in a
predictive model. Conceptually, the case for using ILP to construct relational
features rests on treating these features as functions, the automated discovery
of which necessarily requires some form of first-order learning. Practically,
there are now several reports in the literature that suggest that augmenting
any existing features with ILP-discovered relational features can substantially
improve the predictive power of a model. While the approach is straightforward
enough, much still needs to be done to scale it up to explore more fully the
space of possible features that can be constructed by an ILP system. This is in
principle, infinite and in practice, extremely large. Applications have been
confined to heuristic or random selections from this space. In this paper, we
address this computational difficulty by allowing features to be constructed in
a distributed manner. That is, there is a network of computational units, each
of which employs an ILP engine to construct some small number of features and
then builds a (local) model. We then employ a consensus-based algorithm, in
which neighboring nodes share information to update local models. For a
category of models (those with convex loss functions), it can be shown that the
algorithm will result in all nodes converging to a consensus model. In
practice, it may be slow to achieve this convergence. Nevertheless, our results
on synthetic and real datasets that suggests that in relatively short time the
"best" node in the network reaches a model whose predictive accuracy is
comparable to that obtained using more computational effort in a
non-distributed setting (the best node is identified as the one whose weights
converge first).
|
Topic Modeling of Hierarchical Corpora | We study the problem of topic modeling in corpora whose documents are
organized in a multi-level hierarchy. We explore a parametric approach to this
problem, assuming that the number of topics is known or can be estimated by
cross-validation. The models we consider can be viewed as special
(finite-dimensional) instances of hierarchical Dirichlet processes (HDPs). For
these models we show that there exists a simple variational approximation for
probabilistic inference. The approximation relies on a previously unexploited
inequality that handles the conditional dependence between Dirichlet latent
variables in adjacent levels of the model's hierarchy. We compare our approach
to existing implementations of nonparametric HDPs. On several benchmarks we
find that our approach is faster than Gibbs sampling and able to learn more
predictive models than existing variational methods. Finally, we demonstrate
the large-scale viability of our approach on two newly available corpora from
researchers in computer security---one with 350,000 documents and over 6,000
internal subcategories, the other with a five-level deep hierarchy.
|
10,000+ Times Accelerated Robust Subset Selection (ARSS) | Subset selection from massive data with noised information is increasingly
popular for various applications. This problem is still highly challenging as
current methods are generally slow in speed and sensitive to outliers. To
address the above two issues, we propose an accelerated robust subset selection
(ARSS) method. Specifically in the subset selection area, this is the first
attempt to employ the $\ell_{p}(0<p\leq1)$-norm based measure for the
representation loss, preventing large errors from dominating our objective. As
a result, the robustness against outlier elements is greatly enhanced.
Actually, data size is generally much larger than feature length, i.e. $N\gg
L$. Based on this observation, we propose a speedup solver (via ALM and
equivalent derivations) to highly reduce the computational cost, theoretically
from $O(N^{4})$ to $O(N{}^{2}L)$. Extensive experiments on ten benchmark
datasets verify that our method not only outperforms state of the art methods,
but also runs 10,000+ times faster than the most related method.
|
Optimization Methods for Sparse Pseudo-Likelihood Graphical Model
Selection | Sparse high dimensional graphical model selection is a popular topic in
contemporary machine learning. To this end, various useful approaches have been
proposed in the context of $\ell_1$-penalized estimation in the Gaussian
framework. Though many of these inverse covariance estimation approaches are
demonstrably scalable and have leveraged recent advances in convex
optimization, they still depend on the Gaussian functional form. To address
this gap, a convex pseudo-likelihood based partial correlation graph estimation
method (CONCORD) has been recently proposed. This method uses coordinate-wise
minimization of a regression based pseudo-likelihood, and has been shown to
have robust model selection properties in comparison with the Gaussian
approach. In direct contrast to the parallel work in the Gaussian setting
however, this new convex pseudo-likelihood framework has not leveraged the
extensive array of methods that have been proposed in the machine learning
literature for convex optimization. In this paper, we address this crucial gap
by proposing two proximal gradient methods (CONCORD-ISTA and CONCORD-FISTA) for
performing $\ell_1$-regularized inverse covariance matrix estimation in the
pseudo-likelihood framework. We present timing comparisons with coordinate-wise
minimization and demonstrate that our approach yields tremendous payoffs for
$\ell_1$-penalized partial correlation graph estimation outside the Gaussian
setting, thus yielding the fastest and most scalable approach for such
problems. We undertake a theoretical analysis of our approach and rigorously
demonstrate convergence, and also derive rates thereof.
|
Computational Implications of Reducing Data to Sufficient Statistics | Given a large dataset and an estimation task, it is common to pre-process the
data by reducing them to a set of sufficient statistics. This step is often
regarded as straightforward and advantageous (in that it simplifies statistical
analysis). I show that -on the contrary- reducing data to sufficient statistics
can change a computationally tractable estimation problem into an intractable
one. I discuss connections with recent work in theoretical computer science,
and implications for some techniques to estimate graphical models.
|
Linear, Deterministic, and Order-Invariant Initialization Methods for
the K-Means Clustering Algorithm | Over the past five decades, k-means has become the clustering algorithm of
choice in many application domains primarily due to its simplicity, time/space
efficiency, and invariance to the ordering of the data points. Unfortunately,
the algorithm's sensitivity to the initial selection of the cluster centers
remains to be its most serious drawback. Numerous initialization methods have
been proposed to address this drawback. Many of these methods, however, have
time complexity superlinear in the number of data points, which makes them
impractical for large data sets. On the other hand, linear methods are often
random and/or sensitive to the ordering of the data points. These methods are
generally unreliable in that the quality of their results is unpredictable.
Therefore, it is common practice to perform multiple runs of such methods and
take the output of the run that produces the best results. Such a practice,
however, greatly increases the computational requirements of the otherwise
highly efficient k-means algorithm. In this chapter, we investigate the
empirical performance of six linear, deterministic (non-random), and
order-invariant k-means initialization methods on a large and diverse
collection of data sets from the UCI Machine Learning Repository. The results
demonstrate that two relatively unknown hierarchical initialization methods due
to Su and Dy outperform the remaining four methods with respect to two
objective effectiveness criteria. In addition, a recent method due to Erisoglu
et al. performs surprisingly poorly.
|
Unsupervised learning of clutter-resistant visual representations from
natural videos | Populations of neurons in inferotemporal cortex (IT) maintain an explicit
code for object identity that also tolerates transformations of object
appearance e.g., position, scale, viewing angle [1, 2, 3]. Though the learning
rules are not known, recent results [4, 5, 6] suggest the operation of an
unsupervised temporal-association-based method e.g., Foldiak's trace rule [7].
Such methods exploit the temporal continuity of the visual world by assuming
that visual experience over short timescales will tend to have invariant
identity content. Thus, by associating representations of frames from nearby
times, a representation that tolerates whatever transformations occurred in the
video may be achieved. Many previous studies verified that such rules can work
in simple situations without background clutter, but the presence of visual
clutter has remained problematic for this approach. Here we show that temporal
association based on large class-specific filters (templates) avoids the
problem of clutter. Our system learns in an unsupervised way from natural
videos gathered from the internet, and is able to perform a difficult
unconstrained face recognition task on natural images: Labeled Faces in the
Wild [8].
|
An Approach to Reducing Annotation Costs for BioNLP | There is a broad range of BioNLP tasks for which active learning (AL) can
significantly reduce annotation costs and a specific AL algorithm we have
developed is particularly effective in reducing annotation costs for these
tasks. We have previously developed an AL algorithm called ClosestInitPA that
works best with tasks that have the following characteristics: redundancy in
training material, burdensome annotation costs, Support Vector Machines (SVMs)
work well for the task, and imbalanced datasets (i.e. when set up as a binary
classification problem, one class is substantially rarer than the other). Many
BioNLP tasks have these characteristics and thus our AL algorithm is a natural
approach to apply to BioNLP tasks.
|
Parallel Distributed Block Coordinate Descent Methods based on Pairwise
Comparison Oracle | This paper provides a block coordinate descent algorithm to solve
unconstrained optimization problems. In our algorithm, computation of function
values or gradients is not required. Instead, pairwise comparison of function
values is used. Our algorithm consists of two steps; one is the direction
estimate step and the other is the search step. Both steps require only
pairwise comparison of function values, which tells us only the order of
function values over two points. In the direction estimate step, a Newton type
search direction is estimated. A computation method like block coordinate
descent methods is used with the pairwise comparison. In the search step, a
numerical solution is updated along the estimated direction. The computation in
the direction estimate step can be easily parallelized, and thus, the algorithm
works efficiently to find the minimizer of the objective function. Also, we
show an upper bound of the convergence rate. In numerical experiments, we show
that our method efficiently finds the optimal solution compared to some
existing methods based on the pairwise comparison.
|
A study on effectiveness of extreme learning machine | Extreme learning machine (ELM), proposed by Huang et al., has been shown a
promising learning algorithm for single-hidden layer feedforward neural
networks (SLFNs). Nevertheless, because of the random choice of input weights
and biases, the ELM algorithm sometimes makes the hidden layer output matrix H
of SLFN not full column rank, which lowers the effectiveness of ELM. This paper
discusses the effectiveness of ELM and proposes an improved algorithm called
EELM that makes a proper selection of the input weights and bias before
calculating the output weights, which ensures the full column rank of H in
theory. This improves to some extend the learning rate (testing accuracy,
prediction accuracy, learning time) and the robustness property of the
networks. The experimental results based on both the benchmark function
approximation and real-world problems including classification and regression
applications show the good performances of EELM.
|
A Deep and Autoregressive Approach for Topic Modeling of Multimodal Data | Topic modeling based on latent Dirichlet allocation (LDA) has been a
framework of choice to deal with multimodal data, such as in image annotation
tasks. Another popular approach to model the multimodal data is through deep
neural networks, such as the deep Boltzmann machine (DBM). Recently, a new type
of topic model called the Document Neural Autoregressive Distribution Estimator
(DocNADE) was proposed and demonstrated state-of-the-art performance for text
document modeling. In this work, we show how to successfully apply and extend
this model to multimodal data, such as simultaneous image classification and
annotation. First, we propose SupDocNADE, a supervised extension of DocNADE,
that increases the discriminative power of the learned hidden topic features
and show how to employ it to learn a joint representation from image visual
words, annotation words and class label information. We test our model on the
LabelMe and UIUC-Sports data sets and show that it compares favorably to other
topic models. Second, we propose a deep extension of our model and provide an
efficient way of training the deep model. Experimental results show that our
deep model outperforms its shallow version and reaches state-of-the-art
performance on the Multimedia Information Retrieval (MIR) Flickr data set.
|
EquiNMF: Graph Regularized Multiview Nonnegative Matrix Factorization | Nonnegative matrix factorization (NMF) methods have proved to be powerful
across a wide range of real-world clustering applications. Integrating multiple
types of measurements for the same objects/subjects allows us to gain a deeper
understanding of the data and refine the clustering. We have developed a novel
Graph-reguarized multiview NMF-based method for data integration called
EquiNMF. The parameters for our method are set in a completely automated
data-specific unsupervised fashion, a highly desirable property in real-world
applications. We performed extensive and comprehensive experiments on multiview
imaging data. We show that EquiNMF consistently outperforms other single-view
NMF methods used on concatenated data and multi-view NMF methods with different
types of regularizations.
|
A new approach in machine learning | In this technical report we presented a novel approach to machine learning.
Once the new framework is presented, we will provide a simple and yet very
powerful learning algorithm which will be benchmark on various dataset.
The framework we proposed is based on booleen circuits; more specifically the
classifier produced by our algorithm have that form. Using bits and boolean
gates instead of real numbers and multiplication enable the the learning
algorithm and classifier to use very efficient boolean vector operations. This
enable both the learning algorithm and classifier to be extremely efficient.
The accuracy of the classifier we obtain with our framework compares very
favorably those produced by conventional techniques, both in terms of
efficiency and accuracy.
|
Transfer Learning for Video Recognition with Scarce Training Data for
Deep Convolutional Neural Network | Unconstrained video recognition and Deep Convolution Network (DCN) are two
active topics in computer vision recently. In this work, we apply DCNs as
frame-based recognizers for video recognition. Our preliminary studies,
however, show that video corpora with complete ground truth are usually not
large and diverse enough to learn a robust model. The networks trained directly
on the video data set suffer from significant overfitting and have poor
recognition rate on the test set. The same lack-of-training-sample problem
limits the usage of deep models on a wide range of computer vision problems
where obtaining training data are difficult. To overcome the problem, we
perform transfer learning from images to videos to utilize the knowledge in the
weakly labeled image corpus for video recognition. The image corpus help to
learn important visual patterns for natural images, while these patterns are
ignored by models trained only on the video corpus. Therefore, the resultant
networks have better generalizability and better recognition rate. We show that
by means of transfer learning from image to video, we can learn a frame-based
recognizer with only 4k videos. Because the image corpus is weakly labeled, the
entire learning process requires only 4k annotated instances, which is far less
than the million scale image data sets required by previous works. The same
approach may be applied to other visual recognition tasks where only scarce
training data is available, and it improves the applicability of DCNs in
various computer vision problems. Our experiments also reveal the correlation
between meta-parameters and the performance of DCNs, given the properties of
the target problem and data. These results lead to a heuristic for
meta-parameter selection for future researches, which does not rely on the time
consuming meta-parameter search.
|
Active Metric Learning from Relative Comparisons | This work focuses on active learning of distance metrics from relative
comparison information. A relative comparison specifies, for a data point
triplet $(x_i,x_j,x_k)$, that instance $x_i$ is more similar to $x_j$ than to
$x_k$. Such constraints, when available, have been shown to be useful toward
defining appropriate distance metrics. In real-world applications, acquiring
constraints often require considerable human effort. This motivates us to study
how to select and query the most useful relative comparisons to achieve
effective metric learning with minimum user effort. Given an underlying class
concept that is employed by the user to provide such constraints, we present an
information-theoretic criterion that selects the triplet whose answer leads to
the highest expected gain in information about the classes of a set of
examples. Directly applying the proposed criterion requires examining $O(n^3)$
triplets with $n$ instances, which is prohibitive even for datasets of moderate
size. We show that a randomized selection strategy can be used to reduce the
selection pool from $O(n^3)$ to $O(n)$, allowing us to scale up to larger-size
problems. Experiments show that the proposed method consistently outperforms
two baseline policies.
|
Machine learning for ultrafast X-ray diffraction patterns on large-scale
GPU clusters | The classical method of determining the atomic structure of complex molecules
by analyzing diffraction patterns is currently undergoing drastic developments.
Modern techniques for producing extremely bright and coherent X-ray lasers
allow a beam of streaming particles to be intercepted and hit by an ultrashort
high energy X-ray beam. Through machine learning methods the data thus
collected can be transformed into a three-dimensional volumetric intensity map
of the particle itself. The computational complexity associated with this
problem is very high such that clusters of data parallel accelerators are
required.
We have implemented a distributed and highly efficient algorithm for
inversion of large collections of diffraction patterns targeting clusters of
hundreds of GPUs. With the expected enormous amount of diffraction data to be
produced in the foreseeable future, this is the required scale to approach real
time processing of data at the beam site. Using both real and synthetic data we
look at the scaling properties of the application and discuss the overall
computational viability of this exciting and novel imaging technique.
|
The Ordered Weighted $\ell_1$ Norm: Atomic Formulation, Projections, and
Algorithms | The ordered weighted $\ell_1$ norm (OWL) was recently proposed, with two
different motivations: its good statistical properties as a sparsity promoting
regularizer; the fact that it generalizes the so-called {\it octagonal
shrinkage and clustering algorithm for regression} (OSCAR), which has the
ability to cluster/group regression variables that are highly correlated. This
paper contains several contributions to the study and application of OWL
regularization: the derivation of the atomic formulation of the OWL norm; the
derivation of the dual of the OWL norm, based on its atomic formulation; a new
and simpler derivation of the proximity operator of the OWL norm; an efficient
scheme to compute the Euclidean projection onto an OWL ball; the instantiation
of the conditional gradient (CG, also known as Frank-Wolfe) algorithm for
linear regression problems under OWL regularization; the instantiation of
accelerated projected gradient algorithms for the same class of problems.
Finally, a set of experiments give evidence that accelerated projected gradient
algorithms are considerably faster than CG, for the class of problems
considered.
|
A Fast Quartet Tree Heuristic for Hierarchical Clustering | The Minimum Quartet Tree Cost problem is to construct an optimal weight tree
from the $3{n \choose 4}$ weighted quartet topologies on $n$ objects, where
optimality means that the summed weight of the embedded quartet topologies is
optimal (so it can be the case that the optimal tree embeds all quartets as
nonoptimal topologies). We present a Monte Carlo heuristic, based on randomized
hill climbing, for approximating the optimal weight tree, given the quartet
topology weights. The method repeatedly transforms a dendrogram, with all
objects involved as leaves, achieving a monotonic approximation to the exact
single globally optimal tree. The problem and the solution heuristic has been
extensively used for general hierarchical clustering of nontree-like
(non-phylogeny) data in various domains and across domains with heterogeneous
data. We also present a greatly improved heuristic, reducing the running time
by a factor of order a thousand to ten thousand. All this is implemented and
available, as part of the CompLearn package. We compare performance and running
time of the original and improved versions with those of UPGMA, BioNJ, and NJ,
as implemented in the SplitsTree package on genomic data for which the latter
are optimized.
Keywords: Data and knowledge visualization, Pattern
matching--Clustering--Algorithms/Similarity measures, Hierarchical clustering,
Global optimization, Quartet tree, Randomized hill-climbing,
|
Computing the Stereo Matching Cost with a Convolutional Neural Network | We present a method for extracting depth information from a rectified image
pair. We train a convolutional neural network to predict how well two image
patches match and use it to compute the stereo matching cost. The cost is
refined by cross-based cost aggregation and semiglobal matching, followed by a
left-right consistency check to eliminate errors in the occluded regions. Our
stereo method achieves an error rate of 2.61 % on the KITTI stereo dataset and
is currently (August 2014) the top performing method on this dataset.
|
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.