title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
BISTRO: An Efficient Relaxation-Based Method for Contextual Bandits | We present efficient algorithms for the problem of contextual bandits with
i.i.d. covariates, an arbitrary sequence of rewards, and an arbitrary class of
policies. Our algorithm BISTRO requires d calls to the empirical risk
minimization (ERM) oracle per round, where d is the number of actions. The
method uses unlabeled data to make the problem computationally simple. When the
ERM problem itself is computationally hard, we extend the approach by employing
multiplicative approximation algorithms for the ERM. The integrality gap of the
relaxation only enters in the regret bound rather than the benchmark. Finally,
we show that the adversarial version of the contextual bandit problem is
learnable (and efficient) whenever the full-information supervised online
learning problem has a non-trivial regret guarantee (and efficient).
|
Efficient Second Order Online Learning by Sketching | We propose Sketched Online Newton (SON), an online second order learning
algorithm that enjoys substantially improved regret guarantees for
ill-conditioned data. SON is an enhanced version of the Online Newton Step,
which, via sketching techniques enjoys a running time linear in the dimension
and sketch size. We further develop sparse forms of the sketching methods (such
as Oja's rule), making the computation linear in the sparsity of features.
Together, the algorithm eliminates all computational obstacles in previous
second order online learning approaches.
|
Classification accuracy as a proxy for two sample testing | When data analysts train a classifier and check if its accuracy is
significantly different from chance, they are implicitly performing a
two-sample test. We investigate the statistical properties of this flexible
approach in the high-dimensional setting. We prove two results that hold for
all classifiers in any dimensions: if its true error remains $\epsilon$-better
than chance for some $\epsilon>0$ as $d,n \to \infty$, then (a) the
permutation-based test is consistent (has power approaching to one), (b) a
computationally efficient test based on a Gaussian approximation of the null
distribution is also consistent. To get a finer understanding of the rates of
consistency, we study a specialized setting of distinguishing Gaussians with
mean-difference $\delta$ and common (known or unknown) covariance $\Sigma$,
when $d/n \to c \in (0,\infty)$. We study variants of Fisher's linear
discriminant analysis (LDA) such as "naive Bayes" in a nontrivial regime when
$\epsilon \to 0$ (the Bayes classifier has true accuracy approaching 1/2), and
contrast their power with corresponding variants of Hotelling's test.
Surprisingly, the expressions for their power match exactly in terms of
$n,d,\delta,\Sigma$, and the LDA approach is only worse by a constant factor,
achieving an asymptotic relative efficiency (ARE) of $1/\sqrt{\pi}$ for
balanced samples. We also extend our results to high-dimensional elliptical
distributions with finite kurtosis. Other results of independent interest
include minimax lower bounds, and the optimality of Hotelling's test when
$d=o(n)$. Simulation results validate our theory, and we present practical
takeaway messages along with natural open problems.
|
Strongly-Typed Recurrent Neural Networks | Recurrent neural networks are increasing popular models for sequential
learning. Unfortunately, although the most effective RNN architectures are
perhaps excessively complicated, extensive searches have not found simpler
alternatives. This paper imports ideas from physics and functional programming
into RNN design to provide guiding principles. From physics, we introduce type
constraints, analogous to the constraints that forbids adding meters to
seconds. From functional programming, we require that strongly-typed
architectures factorize into stateless learnware and state-dependent firmware,
reducing the impact of side-effects. The features learned by strongly-typed
nets have a simple semantic interpretation via dynamic average-pooling on
one-dimensional convolutions. We also show that strongly-typed gradients are
better behaved than in classical architectures, and characterize the
representational power of strongly-typed nets. Finally, experiments show that,
despite being more constrained, strongly-typed architectures achieve lower
training and comparable generalization error to classical architectures.
|
Improved Dropout for Shallow and Deep Learning | Dropout has been witnessed with great success in training deep neural
networks by independently zeroing out the outputs of neurons at random. It has
also received a surge of interest for shallow learning, e.g., logistic
regression. However, the independent sampling for dropout could be suboptimal
for the sake of convergence. In this paper, we propose to use multinomial
sampling for dropout, i.e., sampling features or neurons according to a
multinomial distribution with different probabilities for different
features/neurons. To exhibit the optimal dropout probabilities, we analyze the
shallow learning with multinomial dropout and establish the risk bound for
stochastic optimization. By minimizing a sampling dependent factor in the risk
bound, we obtain a distribution-dependent dropout with sampling probabilities
dependent on the second order statistics of the data distribution. To tackle
the issue of evolving distribution of neurons in deep learning, we propose an
efficient adaptive dropout (named \textbf{evolutional dropout}) that computes
the sampling probabilities on-the-fly from a mini-batch of examples. Empirical
studies on several benchmark datasets demonstrate that the proposed dropouts
achieve not only much faster convergence and but also a smaller testing error
than the standard dropout. For example, on the CIFAR-100 data, the evolutional
dropout achieves relative improvements over 10\% on the prediction performance
and over 50\% on the convergence speed compared to the standard dropout.
|
A Tractable Fully Bayesian Method for the Stochastic Block Model | The stochastic block model (SBM) is a generative model revealing macroscopic
structures in graphs. Bayesian methods are used for (i) cluster assignment
inference and (ii) model selection for the number of clusters. In this paper,
we study the behavior of Bayesian inference in the SBM in the large sample
limit. Combining variational approximation and Laplace's method, a consistent
criterion of the fully marginalized log-likelihood is established. Based on
that, we derive a tractable algorithm that solves tasks (i) and (ii)
concurrently, obviating the need for an outer loop to check all model
candidates. Our empirical and theoretical results demonstrate that our method
is scalable in computation, accurate in approximation, and concise in model
selection.
|
Recovery guarantee of weighted low-rank approximation via alternating
minimization | Many applications require recovering a ground truth low-rank matrix from
noisy observations of the entries, which in practice is typically formulated as
a weighted low-rank approximation problem and solved by non-convex optimization
heuristics such as alternating minimization. In this paper, we provide provable
recovery guarantee of weighted low-rank via a simple alternating minimization
algorithm. In particular, for a natural class of matrices and weights and
without any assumption on the noise, we bound the spectral norm of the
difference between the recovered matrix and the ground truth, by the spectral
norm of the weighted noise plus an additive error that decreases exponentially
with the number of rounds of alternating minimization, from either
initialization by SVD or, more importantly, random initialization. These
provide the first theoretical results for weighted low-rank via alternating
minimization with non-binary deterministic weights, significantly generalizing
those for matrix completion, the special case with binary weights, since our
assumptions are similar or weaker than those made in existing works.
Furthermore, this is achieved by a very simple algorithm that improves the
vanilla alternating minimization with a simple clipping step.
The key technical challenge is that under non-binary deterministic weights,
na\"ive alternating steps will destroy the incoherence and spectral properties
of the intermediate solutions, which are needed for making progress towards the
ground truth. We show that the properties only need to hold in an average sense
and can be achieved by the clipping step.
We further provide an alternating algorithm that uses a whitening step that
keeps the properties via SDP and Rademacher rounding and thus requires weaker
assumptions. This technique can potentially be applied in some other
applications and is of independent interest.
|
DOLPHIn - Dictionary Learning for Phase Retrieval | We propose a new algorithm to learn a dictionary for reconstructing and
sparsely encoding signals from measurements without phase. Specifically, we
consider the task of estimating a two-dimensional image from squared-magnitude
measurements of a complex-valued linear transformation of the original image.
Several recent phase retrieval algorithms exploit underlying sparsity of the
unknown signal in order to improve recovery performance. In this work, we
consider such a sparse signal prior in the context of phase retrieval, when the
sparsifying dictionary is not known in advance. Our algorithm jointly
reconstructs the unknown signal - possibly corrupted by noise - and learns a
dictionary such that each patch of the estimated image can be sparsely
represented. Numerical experiments demonstrate that our approach can obtain
significantly better reconstructions for phase retrieval problems with noise
than methods that cannot exploit such "hidden" sparsity. Moreover, on the
theoretical side, we provide a convergence result for our method.
|
Ladder Variational Autoencoders | Variational Autoencoders are powerful models for unsupervised learning.
However deep models with several layers of dependent stochastic variables are
difficult to train which limits the improvements obtained using these highly
expressive models. We propose a new inference model, the Ladder Variational
Autoencoder, that recursively corrects the generative distribution by a data
dependent approximate likelihood in a process resembling the recently proposed
Ladder Network. We show that this model provides state of the art predictive
log-likelihood and tighter log-likelihood lower bound compared to the purely
bottom-up inference in layered Variational Autoencoders and other generative
models. We provide a detailed analysis of the learned hierarchical latent
representation and show that our new inference model is qualitatively different
and utilizes a deeper more distributed hierarchy of latent variables. Finally,
we observe that batch normalization and deterministic warm-up (gradually
turning on the KL-term) are crucial for training variational models with many
stochastic layers.
|
Importance Sampling for Minibatches | Minibatching is a very well studied and highly popular technique in
supervised learning, used by practitioners due to its ability to accelerate
training through better utilization of parallel processing power and reduction
of stochastic variance. Another popular technique is importance sampling -- a
strategy for preferential sampling of more important examples also capable of
accelerating the training process. However, despite considerable effort by the
community in these areas, and due to the inherent technical difficulty of the
problem, there is no existing work combining the power of importance sampling
with the strength of minibatching. In this paper we propose the first {\em
importance sampling for minibatches} and give simple and rigorous complexity
analysis of its performance. We illustrate on synthetic problems that for
training data of certain properties, our sampling can lead to several orders of
magnitude improvement in training time. We then test the new sampling on
several popular datasets, and show that the improvement can reach an order of
magnitude.
|
A Deep Learning Approach to Unsupervised Ensemble Learning | We show how deep learning methods can be applied in the context of
crowdsourcing and unsupervised ensemble learning. First, we prove that the
popular model of Dawid and Skene, which assumes that all classifiers are
conditionally independent, is {\em equivalent} to a Restricted Boltzmann
Machine (RBM) with a single hidden node. Hence, under this model, the posterior
probabilities of the true labels can be instead estimated via a trained RBM.
Next, to address the more general case, where classifiers may strongly violate
the conditional independence assumption, we propose to apply RBM-based Deep
Neural Net (DNN). Experimental results on various simulated and real-world
datasets demonstrate that our proposed DNN approach outperforms other
state-of-the-art methods, in particular when the data violates the conditional
independence assumption.
|
R\'enyi Divergence Variational Inference | This paper introduces the variational R\'enyi bound (VR) that extends
traditional variational inference to R\'enyi's alpha-divergences. This new
family of variational methods unifies a number of existing approaches, and
enables a smooth interpolation from the evidence lower-bound to the log
(marginal) likelihood that is controlled by the value of alpha that
parametrises the divergence. The reparameterization trick, Monte Carlo
approximation and stochastic optimisation methods are deployed to obtain a
tractable and unified framework for optimisation. We further consider negative
alpha values and propose a novel variational inference method as a new special
case in the proposed framework. Experiments on Bayesian neural networks and
variational auto-encoders demonstrate the wide applicability of the VR bound.
|
ERBlox: Combining Matching Dependencies with Machine Learning for Entity
Resolution | Entity resolution (ER), an important and common data cleaning problem, is
about detecting data duplicate representations for the same external entities,
and merging them into single representations. Relatively recently, declarative
rules called "matching dependencies" (MDs) have been proposed for specifying
similarity conditions under which attribute values in database records are
merged. In this work we show the process and the benefits of integrating four
components of ER: (a) Building a classifier for duplicate/non-duplicate record
pairs built using machine learning (ML) techniques; (b) Use of MDs for
supporting the blocking phase of ML; (c) Record merging on the basis of the
classifier results; and (d) The use of the declarative language "LogiQL" -an
extended form of Datalog supported by the "LogicBlox" platform- for all
activities related to data processing, and the specification and enforcement of
MDs.
|
Stratified Bayesian Optimization | We consider derivative-free black-box global optimization of expensive noisy
functions, when most of the randomness in the objective is produced by a few
influential scalar random inputs. We present a new Bayesian global optimization
algorithm, called Stratified Bayesian Optimization (SBO), which uses this
strong dependence to improve performance. Our algorithm is similar in spirit to
stratification, a technique from simulation, which uses strong dependence on a
categorical representation of the random input to reduce variance. We
demonstrate in numerical experiments that SBO outperforms state-of-the-art
Bayesian optimization benchmarks that do not leverage this dependence.
|
Solving Ridge Regression using Sketched Preconditioned SVRG | We develop a novel preconditioning method for ridge regression, based on
recent linear sketching methods. By equipping Stochastic Variance Reduced
Gradient (SVRG) with this preconditioning process, we obtain a significant
speed-up relative to fast stochastic methods such as SVRG, SDCA and SAG.
|
Hyperparameter optimization with approximate gradient | Most models in machine learning contain at least one hyperparameter to
control for model complexity. Choosing an appropriate set of hyperparameters is
both crucial in terms of model accuracy and computationally challenging. In
this work we propose an algorithm for the optimization of continuous
hyperparameters using inexact gradient information. An advantage of this method
is that hyperparameters can be updated before model parameters have fully
converged. We also give sufficient conditions for the global convergence of
this method, based on regularity conditions of the involved functions and
summability of errors. Finally, we validate the empirical performance of this
method on the estimation of regularization constants of L2-regularized logistic
regression and kernel Ridge regression. Empirical benchmarks indicate that our
approach is highly competitive with respect to state of the art methods.
|
NED: An Inter-Graph Node Metric Based On Edit Distance | Node similarity is a fundamental problem in graph analytics. However, node
similarity between nodes in different graphs (inter-graph nodes) has not
received a lot of attention yet. The inter-graph node similarity is important
in learning a new graph based on the knowledge of an existing graph (transfer
learning on graphs) and has applications in biological, communication, and
social networks. In this paper, we propose a novel distance function for
measuring inter-graph node similarity with edit distance, called NED. In NED,
two nodes are compared according to their local neighborhood structures which
are represented as unordered k-adjacent trees, without relying on labels or
other assumptions. Since the computation problem of tree edit distance on
unordered trees is NP-Complete, we propose a modified tree edit distance,
called TED*, for comparing neighborhood trees. TED* is a metric distance, as
the original tree edit distance, but more importantly, TED* is polynomially
computable. As a metric distance, NED admits efficient indexing, provides
interpretable results, and shows to perform better than existing approaches on
a number of data analysis tasks, including graph de-anonymization. Finally, the
efficiency and effectiveness of NED are empirically demonstrated using
real-world graphs.
|
Supervised and Semi-Supervised Text Categorization using LSTM for Region
Embeddings | One-hot CNN (convolutional neural network) has been shown to be effective for
text categorization (Johnson & Zhang, 2015). We view it as a special case of a
general framework which jointly trains a linear model with a non-linear feature
generator consisting of `text region embedding + pooling'. Under this
framework, we explore a more sophisticated region embedding method using Long
Short-Term Memory (LSTM). LSTM can embed text regions of variable (and possibly
large) sizes, whereas the region size needs to be fixed in a CNN. We seek
effective and efficient use of LSTM for this purpose in the supervised and
semi-supervised settings. The best results were obtained by combining region
embeddings in the form of LSTM and convolution layers trained on unlabeled
data. The results indicate that on this task, embeddings of text regions, which
can convey complex concepts, are more useful than embeddings of single words in
isolation. We report performances exceeding the previous best results on four
benchmark datasets.
|
Disentangled Representations in Neural Models | Representation learning is the foundation for the recent success of neural
network models. However, the distributed representations generated by neural
networks are far from ideal. Due to their highly entangled nature, they are di
cult to reuse and interpret, and they do a poor job of capturing the sparsity
which is present in real- world transformations. In this paper, I describe
methods for learning disentangled representations in the two domains of
graphics and computation. These methods allow neural methods to learn
representations which are easy to interpret and reuse, yet they incur little or
no penalty to performance. In the Graphics section, I demonstrate the ability
of these methods to infer the generating parameters of images and rerender
those images under novel conditions. In the Computation section, I describe a
model which is able to factorize a multitask learning problem into subtasks and
which experiences no catastrophic forgetting. Together these techniques provide
the tools to design a wide range of models that learn disentangled
representations and better model the factors of variation in the real world.
|
Network Inference by Learned Node-Specific Degree Prior | We propose a novel method for network inference from partially observed edges
using a node-specific degree prior. The degree prior is derived from observed
edges in the network to be inferred, and its hyper-parameters are determined by
cross validation. Then we formulate network inference as a matrix completion
problem regularized by our degree prior. Our theoretical analysis indicates
that this prior favors a network following the learned degree distribution, and
may lead to improved network recovery error bound than previous work.
Experimental results on both simulated and real biological networks demonstrate
the superior performance of our method in various settings.
|
Ensemble Robustness and Generalization of Stochastic Deep Learning
Algorithms | The question why deep learning algorithms generalize so well has attracted
increasing research interest. However, most of the well-established approaches,
such as hypothesis capacity, stability or sparseness, have not provided
complete explanations (Zhang et al., 2016; Kawaguchi et al., 2017). In this
work, we focus on the robustness approach (Xu & Mannor, 2012), i.e., if the
error of a hypothesis will not change much due to perturbations of its training
examples, then it will also generalize well. As most deep learning algorithms
are stochastic (e.g., Stochastic Gradient Descent, Dropout, and
Bayes-by-backprop), we revisit the robustness arguments of Xu & Mannor, and
introduce a new approach, ensemble robustness, that concerns the robustness of
a population of hypotheses. Through the lens of ensemble robustness, we reveal
that a stochastic learning algorithm can generalize well as long as its
sensitiveness to adversarial perturbations is bounded in average over training
examples. Moreover, an algorithm may be sensitive to some adversarial examples
(Goodfellow et al., 2015) but still generalize well. To support our claims, we
provide extensive simulations for different deep learning algorithms and
different network architectures exhibiting a strong correlation between
ensemble robustness and the ability to generalize.
|
A Simple Practical Accelerated Method for Finite Sums | We describe a novel optimization method for finite sums (such as empirical
risk minimization problems) building on the recently introduced SAGA method.
Our method achieves an accelerated convergence rate on strongly convex smooth
problems. Our method has only one parameter (a step size), and is radically
simpler than other accelerated methods for finite sums. Additionally it can be
applied when the terms are non-smooth, yielding a method applicable in many
areas where operator splitting methods would traditionally be applied.
|
Loss factorization, weakly supervised learning and label noise
robustness | We prove that the empirical risk of most well-known loss functions factors
into a linear term aggregating all labels with a term that is label free, and
can further be expressed by sums of the loss. This holds true even for
non-smooth, non-convex losses and in any RKHS. The first term is a (kernel)
mean operator --the focal quantity of this work-- which we characterize as the
sufficient statistic for the labels. The result tightens known generalization
bounds and sheds new light on their interpretation.
Factorization has a direct application on weakly supervised learning. In
particular, we demonstrate that algorithms like SGD and proximal methods can be
adapted with minimal effort to handle weak supervision, once the mean operator
has been estimated. We apply this idea to learning with asymmetric noisy
labels, connecting and extending prior work. Furthermore, we show that most
losses enjoy a data-dependent (by the mean operator) form of noise robustness,
in contrast with known negative results.
|
Efficient Algorithms for Adversarial Contextual Learning | We provide the first oracle efficient sublinear regret algorithms for
adversarial versions of the contextual bandit problem. In this problem, the
learner repeatedly makes an action on the basis of a context and receives
reward for the chosen action, with the goal of achieving reward competitive
with a large class of policies. We analyze two settings: i) in the transductive
setting the learner knows the set of contexts a priori, ii) in the small
separator setting, there exists a small set of contexts such that any two
policies behave differently in one of the contexts in the set. Our algorithms
fall into the follow the perturbed leader family \cite{Kalai2005} and achieve
regret $O(T^{3/4}\sqrt{K\log(N)})$ in the transductive setting and $O(T^{2/3}
d^{3/4} K\sqrt{\log(N)})$ in the separator setting, where $K$ is the number of
actions, $N$ is the number of baseline policies, and $d$ is the size of the
separator. We actually solve the more general adversarial contextual
semi-bandit linear optimization problem, whilst in the full information setting
we address the even more general contextual combinatorial optimization. We
provide several extensions and implications of our algorithms, such as
switching regret and efficient learning with predictable sequences.
|
Binarized Neural Networks | We introduce a method to train Binarized Neural Networks (BNNs) - neural
networks with binary weights and activations at run-time and when computing the
parameters' gradient at train-time. We conduct two sets of experiments, each
based on a different framework, namely Torch7 and Theano, where we train BNNs
on MNIST, CIFAR-10 and SVHN, and achieve nearly state-of-the-art results.
During the forward pass, BNNs drastically reduce memory size and accesses, and
replace most arithmetic operations with bit-wise operations, which might lead
to a great increase in power-efficiency. Last but not least, we wrote a binary
matrix multiplication GPU kernel with which it is possible to run our MNIST BNN
7 times faster than with an unoptimized GPU kernel, without suffering any loss
in classification accuracy. The code for training and running our BNNs is
available.
|
Fast K-Means with Accurate Bounds | We propose a novel accelerated exact k-means algorithm, which performs better
than the current state-of-the-art low-dimensional algorithm in 18 of 22
experiments, running up to 3 times faster. We also propose a general
improvement of existing state-of-the-art accelerated exact k-means algorithms
through better estimates of the distance bounds used to reduce the number of
distance calculations, and get a speedup in 36 of 44 experiments, up to 1.8
times faster.
We have conducted experiments with our own implementations of existing
methods to ensure homogeneous evaluation of performance, and we show that our
implementations perform as well or better than existing available
implementations. Finally, we propose simplified variants of standard approaches
and show that they are faster than their fully-fledged counterparts in 59 of 62
experiments.
|
Multi-view Kernel Completion | In this paper, we introduce the first method that (1) can complete kernel
matrices with completely missing rows and columns as opposed to individual
missing kernel values, (2) does not require any of the kernels to be complete a
priori, and (3) can tackle non-linear kernels. These aspects are necessary in
practical applications such as integrating legacy data sets, learning under
sensor failures and learning when measurements are costly for some of the
views. The proposed approach predicts missing rows by modelling both
within-view and between-view relationships among kernel values. We show, both
on simulated data and real world data, that the proposed method outperforms
existing techniques in the restricted settings where they are available, and
extends applicability to new settings.
|
Data-Efficient Reinforcement Learning in Continuous-State POMDPs | We present a data-efficient reinforcement learning algorithm resistant to
observation noise. Our method extends the highly data-efficient PILCO algorithm
(Deisenroth & Rasmussen, 2011) into partially observed Markov decision
processes (POMDPs) by considering the filtering process during policy
evaluation. PILCO conducts policy search, evaluating each policy by first
predicting an analytic distribution of possible system trajectories. We
additionally predict trajectories w.r.t. a filtering process, achieving
significantly higher performance than combining a filter with a policy
optimised by the original (unfiltered) framework. Our test setup is the
cartpole swing-up task with sensor noise, which involves nonlinear dynamics and
requires nonlinear control.
|
Homogeneity of Cluster Ensembles | The expectation and the mean of partitions generated by a cluster ensemble
are not unique in general. This issue poses challenges in statistical inference
and cluster stability. In this contribution, we state sufficient conditions for
uniqueness of expectation and mean. The proposed conditions show that a unique
mean is neither exceptional nor generic. To cope with this issue, we introduce
homogeneity as a measure of how likely is a unique mean for a sample of
partitions. We show that homogeneity is related to cluster stability. This
result points to a possible conflict between cluster stability and diversity in
consensus clustering. To assess homogeneity in a practical setting, we propose
an efficient way to compute a lower bound of homogeneity. Empirical results
using the k-means algorithm suggest that uniqueness of the mean partition is
not exceptional for real-world data. Moreover, for samples of high homogeneity,
uniqueness can be enforced by increasing the number of data points or by
removing outlier partitions. In a broader context, this contribution can be
placed as a further step towards a statistical theory of partitions.
|
Generating Images with Perceptual Similarity Metrics based on Deep
Networks | Image-generating machine learning models are typically trained with loss
functions based on distance in the image space. This often leads to
over-smoothed results. We propose a class of loss functions, which we call deep
perceptual similarity metrics (DeePSiM), that mitigate this problem. Instead of
computing distances in the image space, we compute distances between image
features extracted by deep neural networks. This metric better reflects
perceptually similarity of images and thus leads to better results. We show
three applications: autoencoder training, a modification of a variational
autoencoder, and inversion of deep convolutional networks. In all cases, the
generated images look sharp and resemble natural images.
|
Graying the black box: Understanding DQNs | In recent years there is a growing interest in using deep representations for
reinforcement learning. In this paper, we present a methodology and tools to
analyze Deep Q-networks (DQNs) in a non-blind matter. Moreover, we propose a
new model, the Semi Aggregated Markov Decision Process (SAMDP), and an
algorithm that learns it automatically. The SAMDP model allows us to identify
spatio-temporal abstractions directly from features and may be used as a
sub-goal detector in future work. Using our tools we reveal that the features
learned by DQNs aggregate the state space in a hierarchical fashion, explaining
its success. Moreover, we are able to understand and describe the policies
learned by DQNs for three different Atari2600 games and suggest ways to
interpret, debug and optimize deep neural networks in reinforcement learning.
|
Exploiting Cyclic Symmetry in Convolutional Neural Networks | Many classes of images exhibit rotational symmetry. Convolutional neural
networks are sometimes trained using data augmentation to exploit this, but
they are still required to learn the rotation equivariance properties from the
data. Encoding these properties into the network architecture, as we are
already used to doing for translation equivariance by using convolutional
layers, could result in a more efficient use of the parameter budget by
relieving the model from learning them. We introduce four operations which can
be inserted into neural network models as layers, and which can be combined to
make these models partially equivariant to rotations. They also enable
parameter sharing across different orientations. We evaluate the effect of
these architectural modifications on three datasets which exhibit rotational
symmetry and demonstrate improved performance with smaller models.
|
A Variational Analysis of Stochastic Gradient Algorithms | Stochastic Gradient Descent (SGD) is an important algorithm in machine
learning. With constant learning rates, it is a stochastic process that, after
an initial phase of convergence, generates samples from a stationary
distribution. We show that SGD with constant rates can be effectively used as
an approximate posterior inference algorithm for probabilistic modeling.
Specifically, we show how to adjust the tuning parameters of SGD such as to
match the resulting stationary distribution to the posterior. This analysis
rests on interpreting SGD as a continuous-time stochastic process and then
minimizing the Kullback-Leibler divergence between its stationary distribution
and the target posterior. (This is in the spirit of variational inference.) In
more detail, we model SGD as a multivariate Ornstein-Uhlenbeck process and then
use properties of this process to derive the optimal parameters. This
theoretical framework also connects SGD to modern scalable inference
algorithms; we analyze the recently proposed stochastic gradient Fisher scoring
under this perspective. We demonstrate that SGD with properly chosen constant
rates gives a new way to optimize hyperparameters in probabilistic models.
|
Learning to Communicate to Solve Riddles with Deep Distributed Recurrent
Q-Networks | We propose deep distributed recurrent Q-networks (DDRQN), which enable teams
of agents to learn to solve communication-based coordination tasks. In these
tasks, the agents are not given any pre-designed communication protocol.
Therefore, in order to successfully communicate, they must first automatically
develop and agree upon their own communication protocol. We present empirical
results on two multi-agent learning problems based on well-known riddles,
demonstrating that DDRQN can successfully solve such tasks and discover elegant
communication protocols to do so. To our knowledge, this is the first time deep
reinforcement learning has succeeded in learning communication protocols. In
addition, we present ablation experiments that confirm that each of the main
components of the DDRQN architecture are critical to its success.
|
Predicting Clinical Events by Combining Static and Dynamic Information
Using Recurrent Neural Networks | In clinical data sets we often find static information (e.g. patient gender,
blood type, etc.) combined with sequences of data that are recorded during
multiple hospital visits (e.g. medications prescribed, tests performed, etc.).
Recurrent Neural Networks (RNNs) have proven to be very successful for
modelling sequences of data in many areas of Machine Learning. In this work we
present an approach based on RNNs, specifically designed for the clinical
domain, that combines static and dynamic information in order to predict future
events. We work with a database collected in the Charit\'{e} Hospital in Berlin
that contains complete information concerning patients that underwent a kidney
transplantation. After the transplantation three main endpoints can occur:
rejection of the kidney, loss of the kidney and death of the patient. Our goal
is to predict, based on information recorded in the Electronic Health Record of
each patient, whether any of those endpoints will occur within the next six or
twelve months after each visit to the clinic. We compared different types of
RNNs that we developed for this work, with a model based on a Feedforward
Neural Network and a Logistic Regression model. We found that the RNN that we
developed based on Gated Recurrent Units provides the best performance for this
task. We also used the same models for a second task, i.e., next event
prediction, and found that here the model based on a Feedforward Neural Network
outperformed the other models. Our hypothesis is that long-term dependencies
are not as relevant in this task.
|
Practical Black-Box Attacks against Machine Learning | Machine learning (ML) models, e.g., deep neural networks (DNNs), are
vulnerable to adversarial examples: malicious inputs modified to yield
erroneous model outputs, while appearing unmodified to human observers.
Potential attacks include having malicious content like malware identified as
legitimate or controlling vehicle behavior. Yet, all existing adversarial
example attacks require knowledge of either the model internals or its training
data. We introduce the first practical demonstration of an attacker controlling
a remotely hosted DNN with no such knowledge. Indeed, the only capability of
our black-box adversary is to observe labels given by the DNN to chosen inputs.
Our attack strategy consists in training a local model to substitute for the
target DNN, using inputs synthetically generated by an adversary and labeled by
the target DNN. We use the local substitute to craft adversarial examples, and
find that they are misclassified by the targeted DNN. To perform a real-world
and properly-blinded evaluation, we attack a DNN hosted by MetaMind, an online
deep learning API. We find that their DNN misclassifies 84.24% of the
adversarial examples crafted with our substitute. We demonstrate the general
applicability of our strategy to many ML techniques by conducting the same
attack against models hosted by Amazon and Google, using logistic regression
substitutes. They yield adversarial examples misclassified by Amazon and Google
at rates of 96.19% and 88.94%. We also find that this black-box attack strategy
is capable of evading defense strategies previously found to make adversarial
example crafting harder.
|
Compressed Online Dictionary Learning for Fast fMRI Decomposition | We present a method for fast resting-state fMRI spatial decomposi-tions of
very large datasets, based on the reduction of the temporal dimension before
applying dictionary learning on concatenated individual records from groups of
subjects. Introducing a measure of correspondence between spatial
decompositions of rest fMRI, we demonstrates that time-reduced dictionary
learning produces result as reliable as non-reduced decompositions. We also
show that this reduction significantly improves computational scalability.
|
Decoy Bandits Dueling on a Poset | We adress the problem of dueling bandits defined on partially ordered sets,
or posets. In this setting, arms may not be comparable, and there may be
several (incomparable) optimal arms. We propose an algorithm, UnchainedBandits,
that efficiently finds the set of optimal arms of any poset even when pairs of
comparable arms cannot be distinguished from pairs of incomparable arms, with a
set of minimal assumptions. This algorithm relies on the concept of decoys,
which stems from social psychology. For the easier case where the
incomparability information may be accessible, we propose a second algorithm,
SlicingBandits, which takes advantage of this information and achieves a very
significant gain of performance compared to UnchainedBandits. We provide
theoretical guarantees and experimental evaluation for both algorithms.
|
PAC Reinforcement Learning with Rich Observations | We propose and study a new model for reinforcement learning with rich
observations, generalizing contextual bandits to sequential decision making.
These models require an agent to take actions based on observations (features)
with the goal of achieving long-term performance competitive with a large set
of policies. To avoid barriers to sample-efficient learning associated with
large observation spaces and general POMDPs, we focus on problems that can be
summarized by a small number of hidden states and have long-term rewards that
are predictable by a reactive function class. In this setting, we design and
analyze a new reinforcement learning algorithm, Least Squares Value Elimination
by Exploration. We prove that the algorithm learns near optimal behavior after
a number of episodes that is polynomial in all relevant parameters, logarithmic
in the number of policies, and independent of the size of the observation
space. Our result provides theoretical justification for reinforcement learning
with function approximation.
|
Local and Global Convergence of a General Inertial Proximal Splitting
Scheme | This paper is concerned with convex composite minimization problems in a
Hilbert space. In these problems, the objective is the sum of two closed,
proper, and convex functions where one is smooth and the other admits a
computationally inexpensive proximal operator. We analyze a general family of
inertial proximal splitting algorithms (GIPSA) for solving such problems. We
establish finiteness of the sum of squared increments of the iterates and
optimality of the accumulation points. Weak convergence of the entire sequence
then follows if the minimum is attained. Our analysis unifies and extends
several previous results.
We then focus on $\ell_1$-regularized optimization, which is the ubiquitous
special case where the nonsmooth term is the $\ell_1$-norm. For certain
parameter choices, GIPSA is amenable to a local analysis for this problem. For
these choices we show that GIPSA achieves finite "active manifold
identification", i.e. convergence in a finite number of iterations to the
optimal support and sign, after which GIPSA reduces to minimizing a local
smooth function. Local linear convergence then holds under certain conditions.
We determine the rate in terms of the inertia, stepsize, and local curvature.
Our local analysis is applicable to certain recent variants of the Fast
Iterative Shrinkage-Thresholding Algorithm (FISTA), for which we establish
active manifold identification and local linear convergence. Our analysis
motivates the use of a momentum restart scheme in these FISTA variants to
obtain the optimal local linear convergence rate.
|
Poor starting points in machine learning | Poor (even random) starting points for learning/training/optimization are
common in machine learning. In many settings, the method of Robbins and Monro
(online stochastic gradient descent) is known to be optimal for good starting
points, but may not be optimal for poor starting points -- indeed, for poor
starting points Nesterov acceleration can help during the initial iterations,
even though Nesterov methods not designed for stochastic approximation could
hurt during later iterations. The common practice of training with nontrivial
minibatches enhances the advantage of Nesterov acceleration.
|
Binarized Neural Networks: Training Deep Neural Networks with Weights
and Activations Constrained to +1 or -1 | We introduce a method to train Binarized Neural Networks (BNNs) - neural
networks with binary weights and activations at run-time. At training-time the
binary weights and activations are used for computing the parameters gradients.
During the forward pass, BNNs drastically reduce memory size and accesses, and
replace most arithmetic operations with bit-wise operations, which is expected
to substantially improve power-efficiency. To validate the effectiveness of
BNNs we conduct two sets of experiments on the Torch7 and Theano frameworks. On
both, BNNs achieved nearly state-of-the-art results over the MNIST, CIFAR-10
and SVHN datasets. Last but not least, we wrote a binary matrix multiplication
GPU kernel with which it is possible to run our MNIST BNN 7 times faster than
with an unoptimized GPU kernel, without suffering any loss in classification
accuracy. The code for training and running our BNNs is available on-line.
|
Collaborative filtering via sparse Markov random fields | Recommender systems play a central role in providing individualized access to
information and services. This paper focuses on collaborative filtering, an
approach that exploits the shared structure among mind-liked users and similar
items. In particular, we focus on a formal probabilistic framework known as
Markov random fields (MRF). We address the open problem of structure learning
and introduce a sparsity-inducing algorithm to automatically estimate the
interaction structures between users and between items. Item-item and user-user
correlation networks are obtained as a by-product. Large-scale experiments on
movie recommendation and date matching datasets demonstrate the power of the
proposed method.
|
Online Active Linear Regression via Thresholding | We consider the problem of online active learning to collect data for
regression modeling. Specifically, we consider a decision maker with a limited
experimentation budget who must efficiently learn an underlying linear
population model. Our main contribution is a novel threshold-based algorithm
for selection of most informative observations; we characterize its performance
and fundamental lower bounds. We extend the algorithm and its guarantees to
sparse linear regression in high-dimensional settings. Simulations suggest the
algorithm is remarkably robust: it provides significant benefits over passive
random sampling in real-world datasets that exhibit high nonlinearity and high
dimensionality --- significantly reducing both the mean and variance of the
squared error.
|
Toward Optimal Feature Selection in Naive Bayes for Text Categorization | Automated feature selection is important for text categorization to reduce
the feature size and to speed up the learning process of classifiers. In this
paper, we present a novel and efficient feature selection framework based on
the Information Theory, which aims to rank the features with their
discriminative capacity for classification. We first revisit two information
measures: Kullback-Leibler divergence and Jeffreys divergence for binary
hypothesis testing, and analyze their asymptotic properties relating to type I
and type II errors of a Bayesian classifier. We then introduce a new divergence
measure, called Jeffreys-Multi-Hypothesis (JMH) divergence, to measure
multi-distribution divergence for multi-class classification. Based on the
JMH-divergence, we develop two efficient feature selection methods, termed
maximum discrimination ($MD$) and $MD-\chi^2$ methods, for text categorization.
The promising results of extensive experiments demonstrate the effectiveness of
the proposed approaches.
|
Compliance-Aware Bandits | Motivated by clinical trials, we study bandits with observable
non-compliance. At each step, the learner chooses an arm, after, instead of
observing only the reward, it also observes the action that took place. We show
that such noncompliance can be helpful or hurtful to the learner in general.
Unfortunately, naively incorporating compliance information into bandit
algorithms loses guarantees on sublinear regret. We present hybrid algorithms
that maintain regret bounds up to a multiplicative factor and can incorporate
compliance information. Simulations based on real data from the International
Stoke Trial show the practical potential of these algorithms.
|
The Role of Typicality in Object Classification: Improving The
Generalization Capacity of Convolutional Neural Networks | Deep artificial neural networks have made remarkable progress in different
tasks in the field of computer vision. However, the empirical analysis of these
models and investigation of their failure cases has received attention
recently. In this work, we show that deep learning models cannot generalize to
atypical images that are substantially different from training images. This is
in contrast to the superior generalization ability of the visual system in the
human brain. We focus on Convolutional Neural Networks (CNN) as the
state-of-the-art models in object recognition and classification; investigate
this problem in more detail, and hypothesize that training CNN models suffer
from unstructured loss minimization. We propose computational models to improve
the generalization capacity of CNNs by considering how typical a training image
looks like. By conducting an extensive set of experiments we show that
involving a typicality measure can improve the classification results on a new
set of images by a large margin. More importantly, this significant improvement
is achieved without fine-tuning the CNN model on the target image set.
|
Value Iteration Networks | We introduce the value iteration network (VIN): a fully differentiable neural
network with a `planning module' embedded within. VINs can learn to plan, and
are suitable for predicting outcomes that involve planning-based reasoning,
such as policies for reinforcement learning. Key to our approach is a novel
differentiable approximation of the value-iteration algorithm, which can be
represented as a convolutional neural network, and trained end-to-end using
standard backpropagation. We evaluate VIN based policies on discrete and
continuous path-planning domains, and on a natural-language based search task.
We show that by learning an explicit planning computation, VIN policies
generalize better to new, unseen domains.
|
Classification with Boosting of Extreme Learning Machine Over
Arbitrarily Partitioned Data | Machine learning based computational intelligence methods are widely used to
analyze large scale data sets in this age of big data. Extracting useful
predictive modeling from these types of data sets is a challenging problem due
to their high complexity. Analyzing large amount of streaming data that can be
leveraged to derive business value is another complex problem to solve. With
high levels of data availability (\textit{i.e. Big Data}) automatic
classification of them has become an important and complex task. Hence, we
explore the power of applying MapReduce based Distributed AdaBoosting of
Extreme Learning Machine (ELM) to build a predictive bag of classification
models. Accordingly, (i) data set ensembles are created; (ii) ELM algorithm is
used to build weak learners (classifier functions); and (iii) builds a strong
learner from a set of weak learners. We applied this training model to the
benchmark knowledge discovery and data mining data sets.
|
Robust Ensemble Classifier Combination Based on Noise Removal with
One-Class SVM | In machine learning area, as the number of labeled input samples becomes very
large, it is very difficult to build a classification model because of input
data set is not fit in a memory in training phase of the algorithm, therefore,
it is necessary to utilize data partitioning to handle overall data set.
Bagging and boosting based data partitioning methods have been broadly used in
data mining and pattern recognition area. Both of these methods have shown a
great possibility for improving classification model performance. This study is
concerned with the analysis of data set partitioning with noise removal and its
impact on the performance of multiple classifier models. In this study, we
propose noise filtering preprocessing at each data set partition to increment
classifier model performance. We applied Gini impurity approach to find the
best split percentage of noise filter ratio. The filtered sub data set is then
used to train individual ensemble models.
|
Secure Multi-Party Computation Based Privacy Preserving Extreme Learning
Machine Algorithm Over Vertically Distributed Data | Especially in the Big Data era, the usage of different classification methods
is increasing day by day. The success of these classification methods depends
on the effectiveness of learning methods. Extreme learning machine (ELM)
classification algorithm is a relatively new learning method built on
feed-forward neural-network. ELM classification algorithm is a simple and fast
method that can create a model from high-dimensional data sets. Traditional ELM
learning algorithm implicitly assumes complete access to whole data set. This
is a major privacy concern in most of cases. Sharing of private data (i.e.
medical records) is prevented because of security concerns. In this research,
we propose an efficient and secure privacy-preserving learning algorithm for
ELM classification over data that is vertically partitioned among several
parties. The new learning method preserves the privacy on numerical attributes,
builds a classification model without sharing private data without disclosing
the data of each party to others.
|
Nested Mini-Batch K-Means | A new algorithm is proposed which accelerates the mini-batch k-means
algorithm of Sculley (2010) by using the distance bounding approach of Elkan
(2003). We argue that, when incorporating distance bounds into a mini-batch
algorithm, already used data should preferentially be reused. To this end we
propose using nested mini-batches, whereby data in a mini-batch at iteration t
is automatically reused at iteration t+1.
Using nested mini-batches presents two difficulties. The first is that
unbalanced use of data can bias estimates, which we resolve by ensuring that
each data sample contributes exactly once to centroids. The second is in
choosing mini-batch sizes, which we address by balancing premature fine-tuning
of centroids with redundancy induced slow-down. Experiments show that the
resulting nmbatch algorithm is very effective, often arriving within 1% of the
empirical minimum 100 times earlier than the standard mini-batch algorithm.
|
Spoofing detection under noisy conditions: a preliminary investigation
and an initial database | Spoofing detection for automatic speaker verification (ASV), which is to
discriminate between live speech and attacks, has received increasing
attentions recently. However, all the previous studies have been done on the
clean data without significant additive noise. To simulate the real-life
scenarios, we perform a preliminary investigation of spoofing detection under
additive noisy conditions, and also describe an initial database for this task.
The noisy database is based on the ASVspoof challenge 2015 database and
generated by artificially adding background noises at different signal-to-noise
ratios (SNRs). Five different additive noises are included. Our preliminary
results show that using the model trained from clean data, the system
performance degrades significantly in noisy conditions. Phase-based feature is
more noise robust than magnitude-based features. And the systems perform
significantly differ under different noise scenarios.
|
Self-organized control for musculoskeletal robots | With the accelerated development of robot technologies, optimal control
becomes one of the central themes of research. In traditional approaches, the
controller, by its internal functionality, finds appropriate actions on the
basis of the history of sensor values, guided by the goals, intentions,
objectives, learning schemes, and so on planted into it. The idea is that the
controller controls the world---the body plus its environment---as reliably as
possible. However, in elastically actuated robots this approach faces severe
difficulties. This paper advocates for a new paradigm of self-organized
control. The paper presents a solution with a controller that is devoid of any
functionalities of its own, given by a fixed, explicit and context-free
function of the recent history of the sensor values. When applying this
controller to a muscle-tendon driven arm-shoulder system from the Myorobotics
toolkit, we observe a vast variety of self-organized behavior patterns: when
left alone, the arm realizes pseudo-random sequences of different poses but one
can also manipulate the system into definite motion patterns. But most
interestingly, after attaching an object, the controller gets in a functional
resonance with the object's internal dynamics: when given a half-filled bottle,
the system spontaneously starts shaking the bottle so that maximum response
from the dynamics of the water is being generated. After attaching a pendulum
to the arm, the controller drives the pendulum into a circular mode. In this
way, the robot discovers dynamical affordances of objects its body is
interacting with. We also discuss perspectives for using this controller
paradigm for intention driven behavior generation.
|
A Convolutional Attention Network for Extreme Summarization of Source
Code | Attention mechanisms in neural networks have proved useful for problems in
which the input and output do not have fixed dimension. Often there exist
features that are locally translation invariant and would be valuable for
directing the model's attention, but previous attentional architectures are not
constructed to learn such features specifically. We introduce an attentional
neural network that employs convolution on the input tokens to detect local
time-invariant and long-range topical attention features in a context-dependent
way. We apply this architecture to the problem of extreme summarization of
source code snippets into short, descriptive function name-like summaries.
Using those features, the model sequentially generates a summary by
marginalizing over two attention mechanisms: one that predicts the next summary
token based on the attention weights of the input tokens and another that is
able to copy a code token as-is directly into the summary. We demonstrate our
convolutional attention neural network's performance on 10 popular Java
projects showing that it achieves better performance compared to previous
attentional mechanisms.
|
Herding as a Learning System with Edge-of-Chaos Dynamics | Herding defines a deterministic dynamical system at the edge of chaos. It
generates a sequence of model states and parameters by alternating parameter
perturbations with state maximizations, where the sequence of states can be
interpreted as "samples" from an associated MRF model. Herding differs from
maximum likelihood estimation in that the sequence of parameters does not
converge to a fixed point and differs from an MCMC posterior sampling approach
in that the sequence of states is generated deterministically. Herding may be
interpreted as a"perturb and map" method where the parameter perturbations are
generated using a deterministic nonlinear dynamical system rather than randomly
from a Gumbel distribution. This chapter studies the distinct statistical
characteristics of the herding algorithm and shows that the fast convergence
rate of the controlled moments may be attributed to edge of chaos dynamics. The
herding algorithm can also be generalized to models with latent variables and
to a discriminative learning setting. The perceptron cycling theorem ensures
that the fast moment matching property is preserved in the more general
framework.
|
Minimax Lower Bounds for Realizable Transductive Classification | Transductive learning considers a training set of $m$ labeled samples and a
test set of $u$ unlabeled samples, with the goal of best labeling that
particular test set. Conversely, inductive learning considers a training set of
$m$ labeled samples drawn iid from $P(X,Y)$, with the goal of best labeling any
future samples drawn iid from $P(X)$. This comparison suggests that
transduction is a much easier type of inference than induction, but is this
really the case? This paper provides a negative answer to this question, by
proving the first known minimax lower bounds for transductive, realizable,
binary classification. Our lower bounds show that $m$ should be at least
$\Omega(d/\epsilon + \log(1/\delta)/\epsilon)$ when $\epsilon$-learning a
concept class $\mathcal{H}$ of finite VC-dimension $d<\infty$ with confidence
$1-\delta$, for all $m \leq u$. This result draws three important conclusions.
First, general transduction is as hard as general induction, since both
problems have $\Omega(d/m)$ minimax values. Second, the use of unlabeled data
does not help general transduction, since supervised learning algorithms such
as ERM and (Hanneke, 2015) match our transductive lower bounds while ignoring
the unlabeled test set. Third, our transductive lower bounds imply lower bounds
for semi-supervised learning, which add to the important discussion about the
role of unlabeled data in machine learning.
|
The Structured Weighted Violations Perceptron Algorithm | We present the Structured Weighted Violations Perceptron (SWVP) algorithm, a
new structured prediction algorithm that generalizes the Collins Structured
Perceptron (CSP). Unlike CSP, the update rule of SWVP explicitly exploits the
internal structure of the predicted labels. We prove the convergence of SWVP
for linearly separable training sets, provide mistake and generalization
bounds, and show that in the general case these bounds are tighter than those
of the CSP special case. In synthetic data experiments with data drawn from an
HMM, various variants of SWVP substantially outperform its CSP special case.
SWVP also provides encouraging initial dependency parsing results.
|
Minimum Conditional Description Length Estimation for Markov Random
Fields | In this paper we discuss a method, which we call Minimum Conditional
Description Length (MCDL), for estimating the parameters of a subset of sites
within a Markov random field. We assume that the edges are known for the entire
graph $G=(V,E)$. Then, for a subset $U\subset V$, we estimate the parameters
for nodes and edges in $U$ as well as for edges incident to a node in $U$, by
finding the exponential parameter for that subset that yields the best
compression conditioned on the values on the boundary $\partial U$. Our
estimate is derived from a temporally stationary sequence of observations on
the set $U$. We discuss how this method can also be applied to estimate a
spatially invariant parameter from a single configuration, and in so doing,
derive the Maximum Pseudo-Likelihood (MPL) estimate.
|
Graphical Model Sketch | Structured high-cardinality data arises in many domains, and poses a major
challenge for both modeling and inference. Graphical models are a popular
approach to modeling structured data but they are unsuitable for
high-cardinality variables. The count-min (CM) sketch is a popular approach to
estimating probabilities in high-cardinality data but it does not scale well
beyond a few variables. In this work, we bring together the ideas of graphical
models and count sketches; and propose and analyze several approaches to
estimating probabilities in structured high-cardinality streams of data. The
key idea of our approximations is to use the structure of a graphical model and
approximately estimate its factors by "sketches", which hash high-cardinality
variables using random projections. Our approximations are computationally
efficient and their space complexity is independent of the cardinality of
variables. Our error bounds are multiplicative and significantly improve upon
those of the CM sketch, a state-of-the-art approach to estimating probabilities
in streams. We evaluate our approximations on synthetic and real-world
problems, and report an order of magnitude improvements over the CM sketch.
|
DCM Bandits: Learning to Rank with Multiple Clicks | A search engine recommends to the user a list of web pages. The user examines
this list, from the first page to the last, and clicks on all attractive pages
until the user is satisfied. This behavior of the user can be described by the
dependent click model (DCM). We propose DCM bandits, an online learning variant
of the DCM where the goal is to maximize the probability of recommending
satisfactory items, such as web pages. The main challenge of our learning
problem is that we do not observe which attractive item is satisfactory. We
propose a computationally-efficient learning algorithm for solving our problem,
dcmKL-UCB; derive gap-dependent upper bounds on its regret under reasonable
assumptions; and also prove a matching lower bound up to logarithmic factors.
We evaluate our algorithm on synthetic and real-world problems, and show that
it performs well even when our model is misspecified. This work presents the
first practical and regret-optimal online algorithm for learning to rank with
multiple clicks in a cascade-like click model.
|
Learning Efficient Algorithms with Hierarchical Attentive Memory | In this paper, we propose and investigate a novel memory architecture for
neural networks called Hierarchical Attentive Memory (HAM). It is based on a
binary tree with leaves corresponding to memory cells. This allows HAM to
perform memory access in O(log n) complexity, which is a significant
improvement over the standard attention mechanism that requires O(n)
operations, where n is the size of the memory.
We show that an LSTM network augmented with HAM can learn algorithms for
problems like merging, sorting or binary searching from pure input-output
examples. In particular, it learns to sort n numbers in time O(n log n) and
generalizes well to input sequences much longer than the ones seen during the
training. We also show that HAM can be trained to act like classic data
structures: a stack, a FIFO queue and a priority queue.
|
Discriminative Regularization for Generative Models | We explore the question of whether the representations learned by classifiers
can be used to enhance the quality of generative models. Our conjecture is that
labels correspond to characteristics of natural data which are most salient to
humans: identity in faces, objects in images, and utterances in speech. We
propose to take advantage of this by using the representations from
discriminative classifiers to augment the objective function corresponding to a
generative model. In particular we enhance the objective function of the
variational autoencoder, a popular generative model, with a discriminative
regularization term. We show that enhancing the objective function in this way
leads to samples that are clearer and have higher visual quality than the
samples from the standard variational autoencoders.
|
Interactive Bayesian Hierarchical Clustering | Clustering is a powerful tool in data analysis, but it is often difficult to
find a grouping that aligns with a user's needs. To address this, several
methods incorporate constraints obtained from users into clustering algorithms,
but unfortunately do not apply to hierarchical clustering. We design an
interactive Bayesian algorithm that incorporates user interaction into
hierarchical clustering while still utilizing the geometry of the data by
sampling a constrained posterior distribution over hierarchies. We also suggest
several ways to intelligently query a user. The algorithm, along with the
querying schemes, shows promising results on real data.
|
A Theory of Generative ConvNet | We show that a generative random field model, which we call generative
ConvNet, can be derived from the commonly used discriminative ConvNet, by
assuming a ConvNet for multi-category classification and assuming one of the
categories is a base category generated by a reference distribution. If we
further assume that the non-linearity in the ConvNet is Rectified Linear Unit
(ReLU) and the reference distribution is Gaussian white noise, then we obtain a
generative ConvNet model that is unique among energy-based models: The model is
piecewise Gaussian, and the means of the Gaussian pieces are defined by an
auto-encoder, where the filters in the bottom-up encoding become the basis
functions in the top-down decoding, and the binary activation variables
detected by the filters in the bottom-up convolution process become the
coefficients of the basis functions in the top-down deconvolution process. The
Langevin dynamics for sampling the generative ConvNet is driven by the
reconstruction error of this auto-encoder. The contrastive divergence learning
of the generative ConvNet reconstructs the training images by the auto-encoder.
The maximum likelihood learning algorithm can synthesize realistic natural
image patterns.
|
Iterative Hierarchical Optimization for Misspecified Problems (IHOMP) | For complex, high-dimensional Markov Decision Processes (MDPs), it may be
necessary to represent the policy with function approximation. A problem is
misspecified whenever, the representation cannot express any policy with
acceptable performance. We introduce IHOMP : an approach for solving
misspecified problems. IHOMP iteratively learns a set of context specialized
options and combines these options to solve an otherwise misspecified problem.
Our main contribution is proving that IHOMP enjoys theoretical convergence
guarantees. In addition, we extend IHOMP to exploit Option Interruption (OI)
enabling it to decide where the learned options can be reused. Our experiments
demonstrate that IHOMP can find near-optimal solutions to otherwise
misspecified problems and that OI can further improve the solutions.
|
Adaptive Skills, Adaptive Partitions (ASAP) | We introduce the Adaptive Skills, Adaptive Partitions (ASAP) framework that
(1) learns skills (i.e., temporally extended actions or options) as well as (2)
where to apply them. We believe that both (1) and (2) are necessary for a truly
general skill learning framework, which is a key building block needed to scale
up to lifelong learning agents. The ASAP framework can also solve related new
tasks simply by adapting where it applies its existing learned skills. We prove
that ASAP converges to a local optimum under natural conditions. Finally, our
experimental results, which include a RoboCup domain, demonstrate the ability
of ASAP to learn where to reuse skills as well as solve multiple tasks with
considerably less experience than solving each task from scratch.
|
Fast model selection by limiting SVM training times | Kernelized Support Vector Machines (SVMs) are among the best performing
supervised learning methods. But for optimal predictive performance,
time-consuming parameter tuning is crucial, which impedes application. To
tackle this problem, the classic model selection procedure based on grid-search
and cross-validation was refined, e.g. by data subsampling and direct search
heuristics. Here we focus on a different aspect, the stopping criterion for SVM
training. We show that by limiting the training time given to the SVM solver
during parameter tuning we can reduce model selection times by an order of
magnitude.
|
Conditional Dependence via Shannon Capacity: Axioms, Estimators and
Applications | We conduct an axiomatic study of the problem of estimating the strength of a
known causal relationship between a pair of variables. We propose that an
estimate of causal strength should be based on the conditional distribution of
the effect given the cause (and not on the driving distribution of the cause),
and study dependence measures on conditional distributions. Shannon capacity,
appropriately regularized, emerges as a natural measure under these axioms. We
examine the problem of calculating Shannon capacity from the observed samples
and propose a novel fixed-$k$ nearest neighbor estimator, and demonstrate its
consistency. Finally, we demonstrate an application to single-cell
flow-cytometry, where the proposed estimators significantly reduce sample
complexity.
|
Achieving Budget-optimality with Adaptive Schemes in Crowdsourcing | Crowdsourcing platforms provide marketplaces where task requesters can pay to
get labels on their data. Such markets have emerged recently as popular venues
for collecting annotations that are crucial in training machine learning models
in various applications. However, as jobs are tedious and payments are low,
errors are common in such crowdsourced labels. A common strategy to overcome
such noise in the answers is to add redundancy by getting multiple answers for
each task and aggregating them using some methods such as majority voting. For
such a system, there is a fundamental question of interest: how can we maximize
the accuracy given a fixed budget on how many responses we can collect on the
crowdsourcing system. We characterize this fundamental trade-off between the
budget (how many answers the requester can collect in total) and the accuracy
in the estimated labels. In particular, we ask whether adaptive task assignment
schemes lead to a more efficient trade-off between the accuracy and the budget.
Adaptive schemes, where tasks are assigned adaptively based on the data
collected thus far, are widely used in practical crowdsourcing systems to
efficiently use a given fixed budget. However, existing theoretical analyses of
crowdsourcing systems suggest that the gain of adaptive task assignments is
minimal. To bridge this gap, we investigate this question under a strictly more
general probabilistic model, which has been recently introduced to model
practical crowdsourced annotations. Under this generalized Dawid-Skene model,
we characterize the fundamental trade-off between budget and accuracy. We
introduce a novel adaptive scheme that matches this fundamental limit. We
further quantify the fundamental gap between adaptive and non-adaptive schemes,
by comparing the trade-off with the one for non-adaptive schemes. Our analyses
confirm that the gap is significant.
|
Learning Distributed Representations of Sentences from Unlabelled Data | Unsupervised methods for learning distributed representations of words are
ubiquitous in today's NLP research, but far less is known about the best ways
to learn distributed phrase or sentence representations from unlabelled data.
This paper is a systematic comparison of models that learn such
representations. We find that the optimal approach depends critically on the
intended application. Deeper, more complex models are preferable for
representations to be used in supervised systems, but shallow log-linear models
work best for building representation spaces that can be decoded with simple
spatial distance metrics. We also propose two new unsupervised
representation-learning objectives designed to optimise the trade-off between
training time, domain portability and performance.
|
Unsupervised Transductive Domain Adaptation | Supervised learning with large scale labeled datasets and deep layered models
has made a paradigm shift in diverse areas in learning and recognition.
However, this approach still suffers generalization issues under the presence
of a domain shift between the training and the test data distribution. In this
regard, unsupervised domain adaptation algorithms have been proposed to
directly address the domain shift problem. In this paper, we approach the
problem from a transductive perspective. We incorporate the domain shift and
the transductive target inference into our framework by jointly solving for an
asymmetric similarity metric and the optimal transductive target label
assignment. We also show that our model can easily be extended for deep feature
learning in order to learn features which are discriminative in the target
domain. Our experiments show that the proposed method significantly outperforms
state-of-the-art algorithms in both object recognition and digit classification
experiments by a large margin.
|
Learning Privately from Multiparty Data | Learning a classifier from private data collected by multiple parties is an
important problem that has many potential applications. How can we build an
accurate and differentially private global classifier by combining
locally-trained classifiers from different parties, without access to any
party's private data? We propose to transfer the `knowledge' of the local
classifier ensemble by first creating labeled data from auxiliary unlabeled
data, and then train a global $\epsilon$-differentially private classifier. We
show that majority voting is too sensitive and therefore propose a new risk
weighted by class probabilities estimated from the ensemble. Relative to a
non-private solution, our private solution has a generalization error bounded
by $O(\epsilon^{-2}M^{-2})$ where $M$ is the number of parties. This allows
strong privacy without performance loss when $M$ is large, such as in
crowdsensing applications. We demonstrate the performance of our method with
realistic tasks of activity recognition, network intrusion detection, and
malicious URL detection.
|
High Dimensional Inference with Random Maximum A-Posteriori
Perturbations | This paper presents a new approach, called perturb-max, for high-dimensional
statistical inference that is based on applying random perturbations followed
by optimization. This framework injects randomness to maximum a-posteriori
(MAP) predictors by randomly perturbing the potential function for the input. A
classic result from extreme value statistics asserts that perturb-max
operations generate unbiased samples from the Gibbs distribution using
high-dimensional perturbations. Unfortunately, the computational cost of
generating so many high-dimensional random variables can be prohibitive.
However, when the perturbations are of low dimension, sampling the perturb-max
prediction is as efficient as MAP optimization. This paper shows that the
expected value of perturb-max inference with low dimensional perturbations can
be used sequentially to generate unbiased samples from the Gibbs distribution.
Furthermore the expected value of the maximal perturbations is a natural bound
on the entropy of such perturb-max models. A measure concentration result for
perturb-max values shows that the deviation of their sampled average from its
expectation decays exponentially in the number of samples, allowing effective
approximation of the expectation.
|
Data-Driven Online Decision Making with Costly Information Acquisition | In most real-world settings such as recommender systems, finance, and
healthcare, collecting useful information is costly and requires an active
choice on the part of the decision maker. The decision-maker needs to learn
simultaneously what observations to make and what actions to take. This paper
incorporates the information acquisition decision into an online learning
framework. We propose two different algorithms for this dual learning problem:
Sim-OOS and Seq-OOS where observations are made simultaneously and
sequentially, respectively. We prove that both algorithms achieve a regret that
is sublinear in time. The developed framework and algorithms can be used in
many applications including medical informatics, recommender systems and
actionable intelligence in transportation, finance, cyber-security etc., in
which collecting information prior to making decisions is costly. We validate
our algorithms in a breast cancer example setting in which we show substantial
performance gains for our proposed algorithms.
|
Attentive Pooling Networks | In this work, we propose Attentive Pooling (AP), a two-way attention
mechanism for discriminative model training. In the context of pair-wise
ranking or classification with neural networks, AP enables the pooling layer to
be aware of the current input pair, in a way that information from the two
input items can directly influence the computation of each other's
representations. Along with such representations of the paired inputs, AP
jointly learns a similarity measure over projected segments (e.g. trigrams) of
the pair, and subsequently, derives the corresponding attention vector for each
input to guide the pooling. Our two-way attention mechanism is a general
framework independent of the underlying representation learning, and it has
been applied to both convolutional neural networks (CNNs) and recurrent neural
networks (RNNs) in our studies. The empirical results, from three very
different benchmark tasks of question answering/answer selection, demonstrate
that our proposed models outperform a variety of strong baselines and achieve
state-of-the-art performance in all the benchmarks.
|
Optimal Inference in Crowdsourced Classification via Belief Propagation | Crowdsourcing systems are popular for solving large-scale labelling tasks
with low-paid workers. We study the problem of recovering the true labels from
the possibly erroneous crowdsourced labels under the popular Dawid-Skene model.
To address this inference problem, several algorithms have recently been
proposed, but the best known guarantee is still significantly larger than the
fundamental limit. We close this gap by introducing a tighter lower bound on
the fundamental limit and proving that Belief Propagation (BP) exactly matches
this lower bound. The guaranteed optimality of BP is the strongest in the sense
that it is information-theoretically impossible for any other algorithm to
correctly label a larger fraction of the tasks. Experimental results suggest
that BP is close to optimal for all regimes considered and improves upon
competing state-of-the-art algorithms.
|
On the Difficulty of Selecting Ising Models with Approximate Recovery | In this paper, we consider the problem of estimating the underlying graph
associated with an Ising model given a number of independent and identically
distributed samples. We adopt an \emph{approximate recovery} criterion that
allows for a number of missed edges or incorrectly-included edges, in contrast
with the widely-studied exact recovery problem. Our main results provide
information-theoretic lower bounds on the sample complexity for graph classes
imposing constraints on the number of edges, maximal degree, and other
properties. We identify a broad range of scenarios where, either up to constant
factors or logarithmic factors, our lower bounds match the best known lower
bounds for the exact recovery criterion, several of which are known to be tight
or near-tight. Hence, in these cases, approximate recovery has a similar
difficulty to exact recovery in the minimax sense.
Our bounds are obtained via a modification of Fano's inequality for handling
the approximate recovery criterion, along with suitably-designed ensembles of
graphs that can broadly be classed into two categories: (i) Those containing
graphs that contain several isolated edges or cliques and are thus difficult to
distinguish from the empty graph; (ii) Those containing graphs for which
certain groups of nodes are highly correlated, thus making it difficult to
determine precisely which edges connect them. We support our theoretical
results on these ensembles with numerical experiments.
|
Medical Concept Representation Learning from Electronic Health Records
and its Application on Heart Failure Prediction | Objective: To transform heterogeneous clinical data from electronic health
records into clinically meaningful constructed features using data driven
method that rely, in part, on temporal relations among data. Materials and
Methods: The clinically meaningful representations of medical concepts and
patients are the key for health analytic applications. Most of existing
approaches directly construct features mapped to raw data (e.g., ICD or CPT
codes), or utilize some ontology mapping such as SNOMED codes. However, none of
the existing approaches leverage EHR data directly for learning such concept
representation. We propose a new way to represent heterogeneous medical
concepts (e.g., diagnoses, medications and procedures) based on co-occurrence
patterns in longitudinal electronic health records. The intuition behind the
method is to map medical concepts that are co-occuring closely in time to
similar concept vectors so that their distance will be small. We also derive a
simple method to construct patient vectors from the related medical concept
vectors. Results: For qualitative evaluation, we study similar medical concepts
across diagnosis, medication and procedure. In quantitative evaluation, our
proposed representation significantly improves the predictive modeling
performance for onset of heart failure (HF), where classification methods (e.g.
logistic regression, neural network, support vector machine and K-nearest
neighbors) achieve up to 23% improvement in area under the ROC curve (AUC)
using this proposed representation. Conclusion: We proposed an effective method
for patient and medical concept representation learning. The resulting
representation can map relevant concepts together and also improves predictive
modeling performance.
|
Network of Bandits insure Privacy of end-users | In order to distribute the best arm identification task as close as possible
to the user's devices, on the edge of the Radio Access Network, we propose a
new problem setting, where distributed players collaborate to find the best
arm. This architecture guarantees privacy to end-users since no events are
stored. The only thing that can be observed by an adversary through the core
network is aggregated information across users. We provide a first algorithm,
Distributed Median Elimination, which is optimal in term of number of
transmitted bits and near optimal in term of speed-up factor with respect to an
optimal algorithm run independently on each player. In practice, this first
algorithm cannot handle the trade-off between the communication cost and the
speed-up factor, and requires some knowledge about the distribution of players.
Extended Distributed Median Elimination overcomes these limitations, by playing
in parallel different instances of Distributed Median Elimination and selecting
the best one. Experiments illustrate and complete the analysis. According to
the analysis, in comparison to Median Elimination performed on each player, the
proposed algorithm shows significant practical improvements.
|
Semi-supervised Learning with Explicit Relationship Regularization | In many learning tasks, the structure of the target space of a function holds
rich information about the relationships between evaluations of functions on
different data points. Existing approaches attempt to exploit this relationship
information implicitly by enforcing smoothness on function evaluations only.
However, what happens if we explicitly regularize the relationships between
function evaluations? Inspired by homophily, we regularize based on a smooth
relationship function, either defined from the data or with labels. In
experiments, we demonstrate that this significantly improves the performance of
state-of-the-art algorithms in semi-supervised classification and in spectral
data embedding for constrained clustering and dimensionality reduction.
|
A Critical Connectivity Radius for Segmenting Randomly-Generated, High
Dimensional Data Points | Motivated by a $2$-dimensional (unsupervised) image segmentation task whereby
local regions of pixels are clustered via edge detection methods, a more
general probabilistic mathematical framework is devised. Critical thresholds
are calculated that indicate strong correlation between randomly-generated,
high dimensional data points that have been projected into structures in a
partition of a bounded, $2$-dimensional area, of which, an image is a special
case. A neighbor concept for structures in the partition is defined and a
critical radius is uncovered. Measured from a central structure in localized
regions of the partition, the radius indicates strong, long and short range
correlation in the count of occupied structures. The size of a short interval
of radii is estimated upon which the transition from short-to-long range
correlation is virtually assured, which defines a demarcation of when an image
ceases to be "interesting".
|
Community Recovery in Graphs with Locality | Motivated by applications in domains such as social networks and
computational biology, we study the problem of community recovery in graphs
with locality. In this problem, pairwise noisy measurements of whether two
nodes are in the same community or different communities come mainly or
exclusively from nearby nodes rather than uniformly sampled between all nodes
pairs, as in most existing models. We present an algorithm that runs nearly
linearly in the number of measurements and which achieves the information
theoretic limit for exact recovery.
|
Wavelet-Based Semantic Features for Hyperspectral Signature
Discrimination | Hyperspectral signature classification is a quantitative analysis approach
for hyperspectral imagery which performs detection and classification of the
constituent materials at the pixel level in the scene. The classification
procedure can be operated directly on hyperspectral data or performed by using
some features extracted from the corresponding hyperspectral signatures
containing information like the signature's energy or shape. In this paper, we
describe a technique that applies non-homogeneous hidden Markov chain (NHMC)
models to hyperspectral signature classification. The basic idea is to use
statistical models (such as NHMC) to characterize wavelet coefficients which
capture the spectrum semantics (i.e., structural information) at multiple
levels. Experimental results show that the approach based on NHMC models can
outperform existing approaches relevant in classification tasks.
|
Second-Order Stochastic Optimization for Machine Learning in Linear Time | First-order stochastic methods are the state-of-the-art in large-scale
machine learning optimization owing to efficient per-iteration complexity.
Second-order methods, while able to provide faster convergence, have been much
less explored due to the high cost of computing the second-order information.
In this paper we develop second-order stochastic methods for optimization
problems in machine learning that match the per-iteration cost of gradient
based methods, and in certain settings improve upon the overall running time
over popular first-order methods. Furthermore, our algorithm has the desirable
property of being implementable in time linear in the sparsity of the input
data.
|
General Vector Machine | The support vector machine (SVM) is an important class of learning machines
for function approach, pattern recognition, and time-serious prediction, etc.
It maps samples into the feature space by so-called support vectors of selected
samples, and then feature vectors are separated by maximum margin hyperplane.
The present paper presents the general vector machine (GVM) to replace the SVM.
The support vectors are replaced by general project vectors selected from the
usual vector space, and a Monte Carlo (MC) algorithm is developed to find the
general vectors. The general project vectors improves the feature-extraction
ability, and the MC algorithm can control the width of the separation margin of
the hyperplane. By controlling the separation margin, we show that the maximum
margin hyperplane can usually induce the overlearning, and the best learning
machine is achieved with a proper separation margin. Applications in function
approach, pattern recognition, and classification indicate that the developed
method is very successful, particularly for small-set training problems.
Additionally, our algorithm may induce some particular applications, such as
for the transductive inference.
|
Orthogonal Sparse PCA and Covariance Estimation via Procrustes
Reformulation | The problem of estimating sparse eigenvectors of a symmetric matrix attracts
a lot of attention in many applications, especially those with high dimensional
data set. While classical eigenvectors can be obtained as the solution of a
maximization problem, existing approaches formulate this problem by adding a
penalty term into the objective function that encourages a sparse solution.
However, the resulting methods achieve sparsity at the expense of sacrificing
the orthogonality property. In this paper, we develop a new method to estimate
dominant sparse eigenvectors without trading off their orthogonality. The
problem is highly non-convex and hard to handle. We apply the MM framework
where we iteratively maximize a tight lower bound (surrogate function) of the
objective function over the Stiefel manifold. The inner maximization problem
turns out to be a rectangular Procrustes problem, which has a closed form
solution. In addition, we propose a method to improve the covariance estimation
problem when its underlying eigenvectors are known to be sparse. We use the
eigenvalue decomposition of the covariance matrix to formulate an optimization
problem where we impose sparsity on the corresponding eigenvectors. Numerical
experiments show that the proposed eigenvector extraction algorithm matches or
outperforms existing algorithms in terms of support recovery and explained
variance, while the covariance estimation algorithms improve significantly the
sample covariance estimator.
|
Using Deep Q-Learning to Control Optimization Hyperparameters | We present a novel definition of the reinforcement learning state, actions
and reward function that allows a deep Q-network (DQN) to learn to control an
optimization hyperparameter. Using Q-learning with experience replay, we train
two DQNs to accept a state representation of an objective function as input and
output the expected discounted return of rewards, or q-values, connected to the
actions of either adjusting the learning rate or leaving it unchanged. The two
DQNs learn a policy similar to a line search, but differ in the number of
allowed actions. The trained DQNs in combination with a gradient-based update
routine form the basis of the Q-gradient descent algorithms. To demonstrate the
viability of this framework, we show that the DQN's q-values associated with
optimal action converge and that the Q-gradient descent algorithms outperform
gradient descent with an Armijo or nonmonotone line search. Unlike traditional
optimization methods, Q-gradient descent can incorporate any objective
statistic and by varying the actions we gain insight into the type of learning
rate adjustment strategies that are successful for neural network optimization.
|
Convolutional Radio Modulation Recognition Networks | We study the adaptation of convolutional neural networks to the complex
temporal radio signal domain. We compare the efficacy of radio modulation
classification using naively learned features against using expert features
which are widely used in the field today and we show significant performance
improvements. We show that blind temporal learning on large and densely encoded
time series using deep convolutional neural networks is viable and a strong
candidate approach for this task especially at low signal to noise ratio.
|
Coin Betting and Parameter-Free Online Learning | In the recent years, a number of parameter-free algorithms have been
developed for online linear optimization over Hilbert spaces and for learning
with expert advice. These algorithms achieve optimal regret bounds that depend
on the unknown competitors, without having to tune the learning rates with
oracle choices.
We present a new intuitive framework to design parameter-free algorithms for
\emph{both} online linear optimization over Hilbert spaces and for learning
with expert advice, based on reductions to betting on outcomes of adversarial
coins. We instantiate it using a betting algorithm based on the
Krichevsky-Trofimov estimator. The resulting algorithms are simple, with no
parameters to be tuned, and they improve or match previous results in terms of
regret guarantee and per-round complexity.
|
Deep Gaussian Processes for Regression using Approximate Expectation
Propagation | Deep Gaussian processes (DGPs) are multi-layer hierarchical generalisations
of Gaussian processes (GPs) and are formally equivalent to neural networks with
multiple, infinitely wide hidden layers. DGPs are nonparametric probabilistic
models and as such are arguably more flexible, have a greater capacity to
generalise, and provide better calibrated uncertainty estimates than
alternative deep models. This paper develops a new approximate Bayesian
learning scheme that enables DGPs to be applied to a range of medium to large
scale regression problems for the first time. The new method uses an
approximate Expectation Propagation procedure and a novel and efficient
extension of the probabilistic backpropagation algorithm for learning. We
evaluate the new method for non-linear regression on eleven real-world
datasets, showing that it always outperforms GP regression and is almost always
better than state-of-the-art deterministic and sampling-based approximate
inference methods for Bayesian neural networks. As a by-product, this work
provides a comprehensive analysis of six approximate Bayesian methods for
training neural networks.
|
Pursuits in Structured Non-Convex Matrix Factorizations | Efficiently representing real world data in a succinct and parsimonious
manner is of central importance in many fields. We present a generalized greedy
pursuit framework, allowing us to efficiently solve structured matrix
factorization problems, where the factors are allowed to be from arbitrary sets
of structured vectors. Such structure may include sparsity, non-negativeness,
order, or a combination thereof. The algorithm approximates a given matrix by a
linear combination of few rank-1 matrices, each factorized into an outer
product of two vector atoms of the desired structure. For the non-convex
subproblems of obtaining good rank-1 structured matrix atoms, we employ and
analyze a general atomic power method. In addition to the above applications,
we prove linear convergence for generalized pursuit variants in Hilbert spaces
- for the task of approximation over the linear span of arbitrary dictionaries
- which generalizes OMP and is useful beyond matrix problems. Our experiments
on real datasets confirm both the efficiency and also the broad applicability
of our framework in practice.
|
A Minimalistic Approach to Sum-Product Network Learning for Real
Applications | Sum-Product Networks (SPNs) are a class of expressive yet tractable
hierarchical graphical models. LearnSPN is a structure learning algorithm for
SPNs that uses hierarchical co-clustering to simultaneously identifying similar
entities and similar features. The original LearnSPN algorithm assumes that all
the variables are discrete and there is no missing data. We introduce a
practical, simplified version of LearnSPN, MiniSPN, that runs faster and can
handle missing data and heterogeneous features common in real applications. We
demonstrate the performance of MiniSPN on standard benchmark datasets and on
two datasets from Google's Knowledge Graph exhibiting high missingness rates
and a mix of discrete and continuous features.
|
Lasso Guarantees for Time Series Estimation Under Subgaussian Tails and
$ \beta $-Mixing | Many theoretical results on estimation of high dimensional time series
require specifying an underlying data generating model (DGM). Instead, along
the footsteps of~\cite{wong2017lasso}, this paper relies only on (strict)
stationarity and $ \beta $-mixing condition to establish consistency of lasso
when data comes from a $\beta$-mixing process with marginals having subgaussian
tails. Because of the general assumptions, the data can come from DGMs
different than standard time series models such as VAR or ARCH. When the true
DGM is not VAR, the lasso estimates correspond to those of the best linear
predictors using the past observations. We establish non-asymptotic
inequalities for estimation and prediction errors of the lasso estimates.
Together with~\cite{wong2017lasso}, we provide lasso guarantees that cover full
spectrum of the parameters in specifications of $ \beta $-mixing subgaussian
time series. Applications of these results potentially extend to non-Gaussian,
non-Markovian and non-linear times series models as the examples we provide
demonstrate. In order to prove our results, we derive a novel Hanson-Wright
type concentration inequality for $\beta$-mixing subgaussian random vectors
that may be of independent interest.
|
Evaluation of Protein Structural Models Using Random Forests | Protein structure prediction has been a grand challenge problem in the
structure biology over the last few decades. Protein quality assessment plays a
very important role in protein structure prediction. In the paper, we propose a
new protein quality assessment method which can predict both local and global
quality of the protein 3D structural models. Our method uses both multi and
single model quality assessment method for global quality assessment, and uses
chemical, physical, geo-metrical features, and global quality score for local
quality assessment. CASP9 targets are used to generate the features for local
quality assessment. We evaluate the performance of our local quality assessment
method on CASP10, which is comparable with two stage-of-art QA methods based on
the average absolute distance between the real and predicted distance. In
addition, we blindly tested our method on CASP11, and the good performance
shows that combining single and multiple model quality assessment method could
be a good way to improve the accuracy of model quality assessment, and the
random forest technique could be used to train a good local quality assessment
model.
|
Conservative Bandits | We study a novel multi-armed bandit problem that models the challenge faced
by a company wishing to explore new strategies to maximize revenue whilst
simultaneously maintaining their revenue above a fixed baseline, uniformly over
time. While previous work addressed the problem under the weaker requirement of
maintaining the revenue constraint only at a given fixed time in the future,
the algorithms previously proposed are unsuitable due to their design under the
more stringent constraints. We consider both the stochastic and the adversarial
settings, where we propose, natural, yet novel strategies and analyze the price
for maintaining the constraints. Amongst other things, we prove both high
probability and expectation bounds on the regret, while we also consider both
the problem of maintaining the constraints with high probability or
expectation. For the adversarial setting the price of maintaining the
constraint appears to be higher, at least for the algorithm considered. A lower
bound is given showing that the algorithm for the stochastic setting is almost
optimal. Empirical results obtained in synthetic environments complement our
theoretical findings.
|
Deep Learning on FPGAs: Past, Present, and Future | The rapid growth of data size and accessibility in recent years has
instigated a shift of philosophy in algorithm design for artificial
intelligence. Instead of engineering algorithms by hand, the ability to learn
composable systems automatically from massive amounts of data has led to
ground-breaking performance in important domains such as computer vision,
speech recognition, and natural language processing. The most popular class of
techniques used in these domains is called deep learning, and is seeing
significant attention from industry. However, these models require incredible
amounts of data and compute power to train, and are limited by the need for
better hardware acceleration to accommodate scaling beyond current data and
model sizes. While the current solution has been to use clusters of graphics
processing units (GPU) as general purpose processors (GPGPU), the use of field
programmable gate arrays (FPGA) provide an interesting alternative. Current
trends in design tools for FPGAs have made them more compatible with the
high-level software practices typically practiced in the deep learning
community, making FPGAs more accessible to those who build and deploy models.
Since FPGA architectures are flexible, this could also allow researchers the
ability to explore model-level optimizations beyond what is possible on fixed
architectures such as GPUs. As well, FPGAs tend to provide high performance per
watt of power consumption, which is of particular importance for application
scientists interested in large scale server-based deployment or
resource-limited embedded applications. This review takes a look at deep
learning and FPGAs from a hardware acceleration perspective, identifying trends
and innovations that make these technologies a natural fit, and motivates a
discussion on how FPGAs may best serve the needs of the deep learning community
moving forward.
|
A Minimax Theory for Adaptive Data Analysis | In adaptive data analysis, the user makes a sequence of queries on the data,
where at each step the choice of query may depend on the results in previous
steps. The releases are often randomized in order to reduce overfitting for
such adaptively chosen queries. In this paper, we propose a minimax framework
for adaptive data analysis. Assuming Gaussianity of queries, we establish the
first sharp minimax lower bound on the squared error in the order of
$O(\frac{\sqrt{k}\sigma^2}{n})$, where $k$ is the number of queries asked, and
$\sigma^2/n$ is the ordinary signal-to-noise ratio for a single query. Our
lower bound is based on the construction of an approximately least favorable
adversary who picks a sequence of queries that are most likely to be affected
by overfitting. This approximately least favorable adversary uses only one
level of adaptivity, suggesting that the minimax risk for 1-step adaptivity
with k-1 initial releases and that for $k$-step adaptivity are on the same
order. The key technical component of the lower bound proof is a reduction to
finding the convoluting distribution that optimally obfuscates the sign of a
Gaussian signal. Our lower bound construction also reveals a transparent and
elementary proof of the matching upper bound as an alternative approach to
Russo and Zou (2015), who used information-theoretic tools to provide the same
upper bound. We believe that the proposed framework opens up opportunities to
obtain theoretical insights for many other settings of adaptive data analysis,
which would extend the idea to more practical realms.
|
Convex Optimization for Linear Query Processing under Approximate
Differential Privacy | Differential privacy enables organizations to collect accurate aggregates
over sensitive data with strong, rigorous guarantees on individuals' privacy.
Previous work has found that under differential privacy, computing multiple
correlated aggregates as a batch, using an appropriate \emph{strategy}, may
yield higher accuracy than computing each of them independently. However,
finding the best strategy that maximizes result accuracy is non-trivial, as it
involves solving a complex constrained optimization program that appears to be
non-linear and non-convex. Hence, in the past much effort has been devoted in
solving this non-convex optimization program. Existing approaches include
various sophisticated heuristics and expensive numerical solutions. None of
them, however, guarantees to find the optimal solution of this optimization
problem.
This paper points out that under ($\epsilon$, $\delta$)-differential privacy,
the optimal solution of the above constrained optimization problem in search of
a suitable strategy can be found, rather surprisingly, by solving a simple and
elegant convex optimization program. Then, we propose an efficient algorithm
based on Newton's method, which we prove to always converge to the optimal
solution with linear global convergence rate and quadratic local convergence
rate. Empirical evaluations demonstrate the accuracy and efficiency of the
proposed solution.
|
Look, Listen and Learn - A Multimodal LSTM for Speaker Identification | Speaker identification refers to the task of localizing the face of a person
who has the same identity as the ongoing voice in a video. This task not only
requires collective perception over both visual and auditory signals, the
robustness to handle severe quality degradations and unconstrained content
variations are also indispensable. In this paper, we describe a novel
multimodal Long Short-Term Memory (LSTM) architecture which seamlessly unifies
both visual and auditory modalities from the beginning of each sequence input.
The key idea is to extend the conventional LSTM by not only sharing weights
across time steps, but also sharing weights across modalities. We show that
modeling the temporal dependency across face and voice can significantly
improve the robustness to content quality degradations and variations. We also
found that our multimodal LSTM is robustness to distractors, namely the
non-speaking identities. We applied our multimodal LSTM to The Big Bang Theory
dataset and showed that our system outperforms the state-of-the-art systems in
speaker identification with lower false alarm rate and higher recognition
accuracy.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.