title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
A Unifying View of Explicit and Implicit Feature Maps of Graph Kernels | Non-linear kernel methods can be approximated by fast linear ones using
suitable explicit feature maps allowing their application to large scale
problems. We investigate how convolution kernels for structured data are
composed from base kernels and construct corresponding feature maps. On this
basis we propose exact and approximative feature maps for widely used graph
kernels based on the kernel trick. We analyze for which kernels and graph
properties computation by explicit feature maps is feasible and actually more
efficient. In particular, we derive approximative, explicit feature maps for
state-of-the-art kernels supporting real-valued attributes including the
GraphHopper and graph invariant kernels. In extensive experiments we show that
our approaches often achieve a classification accuracy close to the exact
methods based on the kernel trick, but require only a fraction of their running
time. Moreover, we propose and analyze algorithms for computing random walk,
shortest-path and subgraph matching kernels by explicit and implicit feature
maps. Our theoretical results are confirmed experimentally by observing a phase
transition when comparing running time with respect to label diversity, walk
lengths and subgraph size, respectively.
|
Mixing Complexity and its Applications to Neural Networks | We suggest analyzing neural networks through the prism of space constraints.
We observe that most training algorithms applied in practice use bounded
memory, which enables us to use a new notion introduced in the study of
space-time tradeoffs that we call mixing complexity. This notion was devised in
order to measure the (in)ability to learn using a bounded-memory algorithm. In
this paper we describe how we use mixing complexity to obtain new results on
what can and cannot be learned using neural networks.
|
Distributed Bayesian Matrix Factorization with Limited Communication | Bayesian matrix factorization (BMF) is a powerful tool for producing low-rank
representations of matrices and for predicting missing values and providing
confidence intervals. Scaling up the posterior inference for massive-scale
matrices is challenging and requires distributing both data and computation
over many workers, making communication the main computational bottleneck.
Embarrassingly parallel inference would remove the communication needed, by
using completely independent computations on different data subsets, but it
suffers from the inherent unidentifiability of BMF solutions. We introduce a
hierarchical decomposition of the joint posterior distribution, which couples
the subset inferences, allowing for embarrassingly parallel computations in a
sequence of at most three stages. Using an efficient approximate
implementation, we show improvements empirically on both real and simulated
data. Our distributed approach is able to achieve a speed-up of almost an order
of magnitude over the full posterior, with a negligible effect on predictive
accuracy. Our method outperforms state-of-the-art embarrassingly parallel MCMC
methods in accuracy, and achieves results competitive to other available
distributed and parallel implementations of BMF.
|
Wireless Interference Identification with Convolutional Neural Networks | The steadily growing use of license-free frequency bands requires reliable
coexistence management for deterministic medium utilization. For interference
mitigation, proper wireless interference identification (WII) is essential. In
this work we propose the first WII approach based upon deep convolutional
neural networks (CNNs). The CNN naively learns its features through
self-optimization during an extensive data-driven GPU-based training process.
We propose a CNN example which is based upon sensing snapshots with a limited
duration of 12.8 {\mu}s and an acquisition bandwidth of 10 MHz. The CNN differs
between 15 classes. They represent packet transmissions of IEEE 802.11 b/g,
IEEE 802.15.4 and IEEE 802.15.1 with overlapping frequency channels within the
2.4 GHz ISM band. We show that the CNN outperforms state-of-the-art WII
approaches and has a classification accuracy greater than 95% for
signal-to-noise ratio of at least -5 dB.
|
Predicting Rankings of Software Verification Competitions | Software verification competitions, such as the annual SV-COMP, evaluate
software verification tools with respect to their effectivity and efficiency.
Typically, the outcome of a competition is a (possibly category-specific)
ranking of the tools. For many applications, such as building portfolio
solvers, it would be desirable to have an idea of the (relative) performance of
verification tools on a given verification task beforehand, i.e., prior to
actually running all tools on the task.
In this paper, we present a machine learning approach to predicting rankings
of tools on verification tasks. The method builds upon so-called label ranking
algorithms, which we complement with appropriate kernels providing a similarity
measure for verification tasks. Our kernels employ a graph representation for
software source code that mixes elements of control flow and program dependence
graphs with abstract syntax trees. Using data sets from SV-COMP, we demonstrate
our rank prediction technique to generalize well and achieve a rather high
predictive accuracy. In particular, our method outperforms a recently proposed
feature-based approach of Demyanova et al. (when applied to rank predictions).
|
Attentive Recurrent Comparators | Rapid learning requires flexible representations to quickly adopt to new
evidence. We develop a novel class of models called Attentive Recurrent
Comparators (ARCs) that form representations of objects by cycling through them
and making observations. Using the representations extracted by ARCs, we
develop a way of approximating a \textit{dynamic representation space} and use
it for one-shot learning. In the task of one-shot classification on the
Omniglot dataset, we achieve the state of the art performance with an error
rate of 1.5\%. This represents the first super-human result achieved for this
task with a generic model that uses only pixel information.
|
A Generic Online Parallel Learning Framework for Large Margin Models | To speed up the training process, many existing systems use parallel
technology for online learning algorithms. However, most research mainly focus
on stochastic gradient descent (SGD) instead of other algorithms. We propose a
generic online parallel learning framework for large margin models, and also
analyze our framework on popular large margin algorithms, including MIRA and
Structured Perceptron. Our framework is lock-free and easy to implement on
existing systems. Experiments show that systems with our framework can gain
near linear speed up by increasing running threads, and with no loss in
accuracy.
|
A Robust Adaptive Stochastic Gradient Method for Deep Learning | Stochastic gradient algorithms are the main focus of large-scale optimization
problems and led to important successes in the recent advancement of the deep
learning algorithms. The convergence of SGD depends on the careful choice of
learning rate and the amount of the noise in stochastic estimates of the
gradients. In this paper, we propose an adaptive learning rate algorithm, which
utilizes stochastic curvature information of the loss function for
automatically tuning the learning rates. The information about the element-wise
curvature of the loss function is estimated from the local statistics of the
stochastic first order gradients. We further propose a new variance reduction
technique to speed up the convergence. In our experiments with deep neural
networks, we obtained better performance compared to the popular stochastic
gradient algorithms.
|
Unsupervised Steganalysis Based on Artificial Training Sets | In this paper, an unsupervised steganalysis method that combines artificial
training setsand supervised classification is proposed. We provide a formal
framework for unsupervisedclassification of stego and cover images in the
typical situation of targeted steganalysis (i.e.,for a known algorithm and
approximate embedding bit rate). We also present a completeset of experiments
using 1) eight different image databases, 2) image features based on
RichModels, and 3) three different embedding algorithms: Least Significant Bit
(LSB) matching,Highly undetectable steganography (HUGO) and Wavelet Obtained
Weights (WOW). Weshow that the experimental results outperform previous methods
based on Rich Models inthe majority of the tested cases. At the same time, the
proposed approach bypasses theproblem of Cover Source Mismatch -when the
embedding algorithm and bit rate are known-, since it removes the need of a
training database when we have a large enough testing set.Furthermore, we
provide a generic proof of the proposed framework in the machine
learningcontext. Hence, the results of this paper could be extended to other
classification problemssimilar to steganalysis.
|
Opening the Black Box of Deep Neural Networks via Information | Despite their great success, there is still no comprehensive theoretical
understanding of learning with Deep Neural Networks (DNNs) or their inner
organization. Previous work proposed to analyze DNNs in the \textit{Information
Plane}; i.e., the plane of the Mutual Information values that each layer
preserves on the input and output variables. They suggested that the goal of
the network is to optimize the Information Bottleneck (IB) tradeoff between
compression and prediction, successively, for each layer.
In this work we follow up on this idea and demonstrate the effectiveness of
the Information-Plane visualization of DNNs. Our main results are: (i) most of
the training epochs in standard DL are spent on {\emph compression} of the
input to efficient representation and not on fitting the training labels. (ii)
The representation compression phase begins when the training errors becomes
small and the Stochastic Gradient Decent (SGD) epochs change from a fast drift
to smaller training error into a stochastic relaxation, or random diffusion,
constrained by the training error value. (iii) The converged layers lie on or
very close to the Information Bottleneck (IB) theoretical bound, and the maps
from the input to any hidden layer and from this hidden layer to the output
satisfy the IB self-consistent equations. This generalization through noise
mechanism is unique to Deep Neural Networks and absent in one layer networks.
(iv) The training time is dramatically reduced when adding more hidden layers.
Thus the main advantage of the hidden layers is computational. This can be
explained by the reduced relaxation time, as this it scales super-linearly
(exponentially for simple diffusion) with the information compression from the
previous layer.
|
Robust Communication-Optimal Distributed Clustering Algorithms | In this work, we study the $k$-median and $k$-means clustering problems when
the data is distributed across many servers and can contain outliers. While
there has been a lot of work on these problems for worst-case instances, we
focus on gaining a finer understanding through the lens of beyond worst-case
analysis. Our main motivation is the following: for many applications such as
clustering proteins by function or clustering communities in a social network,
there is some unknown target clustering, and the hope is that running a
$k$-median or $k$-means algorithm will produce clusterings which are close to
matching the target clustering. Worst-case results can guarantee constant
factor approximations to the optimal $k$-median or $k$-means objective value,
but not closeness to the target clustering.
Our first result is a distributed algorithm which returns a near-optimal
clustering assuming a natural notion of stability, namely, approximation
stability [Balcan et. al 2013], even when a constant fraction of the data are
outliers. The communication complexity is $\tilde O(sk+z)$ where $s$ is the
number of machines, $k$ is the number of clusters, and $z$ is the number of
outliers.
Next, we show this amount of communication cannot be improved even in the
setting when the input satisfies various non-worst-case assumptions. We give a
matching $\Omega(sk+z)$ lower bound on the communication required both for
approximating the optimal $k$-means or $k$-median cost up to any constant, and
for returning a clustering that is close to the target clustering in Hamming
distance. These lower bounds hold even when the data satisfies approximation
stability or other common notions of stability, and the cluster sizes are
balanced. Therefore, $\Omega(sk+z)$ is a communication bottleneck, even for
real-world instances.
|
Meta Networks | Neural networks have been successfully applied in applications with a large
amount of labeled data. However, the task of rapid generalization on new
concepts with small training data while preserving performances on previously
learned ones still presents a significant challenge to neural network models.
In this work, we introduce a novel meta learning method, Meta Networks
(MetaNet), that learns a meta-level knowledge across tasks and shifts its
inductive biases via fast parameterization for rapid generalization. When
evaluated on Omniglot and Mini-ImageNet benchmarks, our MetaNet models achieve
a near human-level performance and outperform the baseline approaches by up to
6% accuracy. We demonstrate several appealing properties of MetaNet relating to
generalization and continual learning.
|
Encrypted accelerated least squares regression | Information that is stored in an encrypted format is, by definition, usually
not amenable to statistical analysis or machine learning methods. In this paper
we present detailed analysis of coordinate and accelerated gradient descent
algorithms which are capable of fitting least squares and penalised ridge
regression models, using data encrypted under a fully homomorphic encryption
scheme. Gradient descent is shown to dominate in terms of encrypted
computational speed, and theoretical results are proven to give parameter
bounds which ensure correctness of decryption. The characteristics of encrypted
computation are empirically shown to favour a non-standard acceleration
technique. This demonstrates the possibility of approximating conventional
statistical regression methods using encrypted data without compromising
privacy.
|
Exact Topology Reconstruction of Radial Dynamical Systems with
Applications to Distribution System of the Power Grid | In this article we present a method to reconstruct the interconnectedness of
dynamically related stochastic processes, where the interactions are
bi-directional and the underlying topology is a tree. Our approach is based on
multivariate Wiener filtering which recovers spurious edges apart from the true
edges in the topology reconstruction. The main contribution of this work is to
show that all spurious links obtained using Wiener filtering can be eliminated
if the underlying topology is a tree based on which we present a three stage
network reconstruction procedure for trees. We illustrate the effectiveness of
the method developed by applying it on a typical distribution system of the
electric grid.
|
Learning the Structure of Generative Models without Labeled Data | Curating labeled training data has become the primary bottleneck in machine
learning. Recent frameworks address this bottleneck with generative models to
synthesize labels at scale from weak supervision sources. The generative
model's dependency structure directly affects the quality of the estimated
labels, but selecting a structure automatically without any labeled data is a
distinct challenge. We propose a structure estimation method that maximizes the
$\ell_1$-regularized marginal pseudolikelihood of the observed data. Our
analysis shows that the amount of unlabeled data required to identify the true
structure scales sublinearly in the number of possible dependencies for a broad
class of models. Simulations show that our method is 100$\times$ faster than a
maximum likelihood approach and selects $1/4$ as many extraneous dependencies.
We also show that our method provides an average of 1.5 F1 points of
improvement over existing, user-developed information extraction applications
on real-world data such as PubMed journal abstracts.
|
Binarized Convolutional Landmark Localizers for Human Pose Estimation
and Face Alignment with Limited Resources | Our goal is to design architectures that retain the groundbreaking
performance of CNNs for landmark localization and at the same time are
lightweight, compact and suitable for applications with limited computational
resources. To this end, we make the following contributions: (a) we are the
first to study the effect of neural network binarization on localization tasks,
namely human pose estimation and face alignment. We exhaustively evaluate
various design choices, identify performance bottlenecks, and more importantly
propose multiple orthogonal ways to boost performance. (b) Based on our
analysis, we propose a novel hierarchical, parallel and multi-scale residual
architecture that yields large performance improvement over the standard
bottleneck block while having the same number of parameters, thus bridging the
gap between the original network and its binarized counterpart. (c) We perform
a large number of ablation studies that shed light on the properties and the
performance of the proposed block. (d) We present results for experiments on
the most challenging datasets for human pose estimation and face alignment,
reporting in many cases state-of-the-art performance. Code can be downloaded
from https://www.adrianbulat.com/binary-cnn-landmarks
|
Using Synthetic Data to Train Neural Networks is Model-Based Reasoning | We draw a formal connection between using synthetic training data to optimize
neural network parameters and approximate, Bayesian, model-based reasoning. In
particular, training a neural network using synthetic data can be viewed as
learning a proposal distribution generator for approximate inference in the
synthetic-data generative model. We demonstrate this connection in a
recognition task where we develop a novel Captcha-breaking architecture and
train it using synthetic data, demonstrating both state-of-the-art performance
and a way of computing task-specific posterior uncertainty. Using a neural
network trained this way, we also demonstrate successful breaking of real-world
Captchas currently used by Facebook and Wikipedia. Reasoning from these
empirical results and drawing connections with Bayesian modeling, we discuss
the robustness of synthetic data results and suggest important considerations
for ensuring good neural network generalization when training with synthetic
data.
|
How to Escape Saddle Points Efficiently | This paper shows that a perturbed form of gradient descent converges to a
second-order stationary point in a number iterations which depends only
poly-logarithmically on dimension (i.e., it is almost "dimension-free"). The
convergence rate of this procedure matches the well-known convergence rate of
gradient descent to first-order stationary points, up to log factors. When all
saddle points are non-degenerate, all second-order stationary points are local
minima, and our result thus shows that perturbed gradient descent can escape
saddle points almost for free. Our results can be directly applied to many
machine learning applications, including deep learning. As a particular
concrete example of such an application, we show that our results can be used
directly to establish sharp global convergence rates for matrix factorization.
Our results rely on a novel characterization of the geometry around saddle
points, which may be of independent interest to the non-convex optimization
community.
|
Being Robust (in High Dimensions) Can Be Practical | Robust estimation is much more challenging in high dimensions than it is in
one dimension: Most techniques either lead to intractable optimization problems
or estimators that can tolerate only a tiny fraction of errors. Recent work in
theoretical computer science has shown that, in appropriate distributional
models, it is possible to robustly estimate the mean and covariance with
polynomial time algorithms that can tolerate a constant fraction of
corruptions, independent of the dimension. However, the sample and time
complexity of these algorithms is prohibitively large for high-dimensional
applications. In this work, we address both of these issues by establishing
sample complexity bounds that are optimal, up to logarithmic factors, as well
as giving various refinements that allow the algorithms to tolerate a much
larger fraction of corruptions. Finally, we show on both synthetic and real
data that our algorithms have state-of-the-art performance and suddenly make
high-dimensional robust estimation a realistic possibility.
|
Toward Controlled Generation of Text | Generic generation and manipulation of text is challenging and has limited
success compared to recent deep generative modeling in visual domain. This
paper aims at generating plausible natural language sentences, whose attributes
are dynamically controlled by learning disentangled latent representations with
designated semantics. We propose a new neural generative model which combines
variational auto-encoders and holistic attribute discriminators for effective
imposition of semantic structures. With differentiable approximation to
discrete text samples, explicit constraints on independent attribute controls,
and efficient collaborative learning of generator and discriminators, our model
learns highly interpretable representations from even only word annotations,
and produces realistic sentences with desired attributes. Quantitative
evaluation validates the accuracy of sentence and attribute generation.
|
A Laplacian Framework for Option Discovery in Reinforcement Learning | Representation learning and option discovery are two of the biggest
challenges in reinforcement learning (RL). Proto-value functions (PVFs) are a
well-known approach for representation learning in MDPs. In this paper we
address the option discovery problem by showing how PVFs implicitly define
options. We do it by introducing eigenpurposes, intrinsic reward functions
derived from the learned representations. The options discovered from
eigenpurposes traverse the principal directions of the state space. They are
useful for multiple tasks because they are discovered without taking the
environment's rewards into consideration. Moreover, different options act at
different time scales, making them helpful for exploration. We demonstrate
features of eigenpurposes in traditional tabular domains as well as in Atari
2600 games.
|
Self-Paced Multitask Learning with Shared Knowledge | This paper introduces self-paced task selection to multitask learning, where
instances from more closely related tasks are selected in a progression of
easier-to-harder tasks, to emulate an effective human education strategy, but
applied to multitask machine learning. We develop the mathematical foundation
for the approach based on iterative selection of the most appropriate task,
learning the task parameters, and updating the shared knowledge, optimizing a
new bi-convex loss function. This proposed method applies quite generally,
including to multitask feature learning, multitask learning with alternating
structure optimization, etc. Results show that in each of the above
formulations self-paced (easier-to-harder) task selection outperforms the
baseline version of these methods in all the experiments.
|
Compositional Falsification of Cyber-Physical Systems with Machine
Learning Components | Cyber-physical systems (CPS), such as automotive systems, are starting to
include sophisticated machine learning (ML) components. Their correctness,
therefore, depends on properties of the inner ML modules. While learning
algorithms aim to generalize from examples, they are only as good as the
examples provided, and recent efforts have shown that they can produce
inconsistent output under small adversarial perturbations. This raises the
question: can the output from learning components can lead to a failure of the
entire CPS? In this work, we address this question by formulating it as a
problem of falsifying signal temporal logic (STL) specifications for CPS with
ML components. We propose a compositional falsification framework where a
temporal logic falsifier and a machine learning analyzer cooperate with the aim
of finding falsifying executions of the considered model. The efficacy of the
proposed technique is shown on an automatic emergency braking system model with
a perception component based on deep neural networks.
|
Belief Propagation in Conditional RBMs for Structured Prediction | Restricted Boltzmann machines~(RBMs) and conditional RBMs~(CRBMs) are popular
models for a wide range of applications. In previous work, learning on such
models has been dominated by contrastive divergence~(CD) and its variants.
Belief propagation~(BP) algorithms are believed to be slow for structured
prediction on conditional RBMs~(e.g., Mnih et al. [2011]), and not as good as
CD when applied in learning~(e.g., Larochelle et al. [2012]). In this work, we
present a matrix-based implementation of belief propagation algorithms on
CRBMs, which is easily scalable to tens of thousands of visible and hidden
units. We demonstrate that, in both maximum likelihood and max-margin learning,
training conditional RBMs with BP as the inference routine can provide
significantly better results than current state-of-the-art CD methods on
structured prediction problems. We also include practical guidelines on
training CRBMs with BP, and some insights on the interaction of learning and
inference algorithms for CRBMs.
|
Optimization of distributions differences for classification | In this paper we introduce a new classification algorithm called Optimization
of Distributions Differences (ODD). The algorithm aims to find a transformation
from the feature space to a new space where the instances in the same class are
as close as possible to one another while the gravity centers of these classes
are as far as possible from one another. This aim is formulated as a
multiobjective optimization problem that is solved by a hybrid of an
evolutionary strategy and the Quasi-Newton method. The choice of the
transformation function is flexible and could be any continuous space function.
We experiment with a linear and a non-linear transformation in this paper. We
show that the algorithm can outperform 6 other state-of-the-art classification
methods, namely naive Bayes, support vector machines, linear discriminant
analysis, multi-layer perceptrons, decision trees, and k-nearest neighbors, in
12 standard classification datasets. Our results show that the method is less
sensitive to the imbalanced number of instances comparing to these methods. We
also show that ODD maintains its performance better than other classification
methods in these datasets, hence, offers a better generalization ability.
|
Co-Clustering for Multitask Learning | This paper presents a new multitask learning framework that learns a shared
representation among the tasks, incorporating both task and feature clusters.
The jointly-induced clusters yield a shared latent subspace where task
relationships are learned more effectively and more generally than in
state-of-the-art multitask learning methods. The proposed general framework
enables the derivation of more specific or restricted state-of-the-art
multitask methods. The paper also proposes a highly-scalable multitask learning
algorithm, based on the new framework, using conjugate gradient descent and
generalized \textit{Sylvester equations}. Experimental results on synthetic and
benchmark datasets show that the proposed method systematically outperforms
several state-of-the-art multitask learning methods.
|
Scalable Deep Traffic Flow Neural Networks for Urban Traffic Congestion
Prediction | Tracking congestion throughout the network road is a critical component of
Intelligent transportation network management systems. Understanding how the
traffic flows and short-term prediction of congestion occurrence due to
rush-hour or incidents can be beneficial to such systems to effectively manage
and direct the traffic to the most appropriate detours. Many of the current
traffic flow prediction systems are designed by utilizing a central processing
component where the prediction is carried out through aggregation of the
information gathered from all measuring stations. However, centralized systems
are not scalable and fail provide real-time feedback to the system whereas in a
decentralized scheme, each node is responsible to predict its own short-term
congestion based on the local current measurements in neighboring nodes.
We propose a decentralized deep learning-based method where each node
accurately predicts its own congestion state in real-time based on the
congestion state of the neighboring stations. Moreover, historical data from
the deployment site is not required, which makes the proposed method more
suitable for newly installed stations. In order to achieve higher performance,
we introduce a regularized Euclidean loss function that favors high congestion
samples over low congestion samples to avoid the impact of the unbalanced
training dataset. A novel dataset for this purpose is designed based on the
traffic data obtained from traffic control stations in northern California.
Extensive experiments conducted on the designed benchmark reflect a successful
congestion prediction.
|
Active Learning for Cost-Sensitive Classification | We design an active learning algorithm for cost-sensitive multiclass
classification: problems where different errors have different costs. Our
algorithm, COAL, makes predictions by regressing to each label's cost and
predicting the smallest. On a new example, it uses a set of regressors that
perform well on past data to estimate possible costs for each label. It queries
only the labels that could be the best, ignoring the sure losers. We prove COAL
can be efficiently implemented for any regression family that admits squared
loss optimization; it also enjoys strong guarantees with respect to predictive
performance and labeling effort. We empirically compare COAL to passive
learning and several active learning baselines, showing significant
improvements in labeling effort and test cost on real-world datasets.
|
Unsupervised Basis Function Adaptation for Reinforcement Learning | When using reinforcement learning (RL) algorithms to evaluate a policy it is
common, given a large state space, to introduce some form of approximation
architecture for the value function (VF). The exact form of this architecture
can have a significant effect on the accuracy of the VF estimate, however, and
determining a suitable approximation architecture can often be a highly complex
task. Consequently there is a large amount of interest in the potential for
allowing RL algorithms to adaptively generate approximation architectures.
We investigate a method of adapting approximation architectures which uses
feedback regarding the frequency with which an agent has visited certain states
to guide which areas of the state space to approximate with greater detail.
This method is "unsupervised" in the sense that it makes no direct reference to
reward or the VF estimate. We introduce an algorithm based upon this idea which
adapts a state aggregation approximation architecture on-line.
A common method of scoring a VF estimate is to weight the squared Bellman
error of each state-action by the probability of that state-action occurring.
Adopting this scoring method, and assuming $S$ states, we demonstrate
theoretically that - provided (1) the number of cells $X$ in the state
aggregation architecture is of order $\sqrt{S}\log_2{S}\ln{S}$ or greater, (2)
the policy and transition function are close to deterministic, and (3) the
prior for the transition function is uniformly distributed - our algorithm,
used in conjunction with a suitable RL algorithm, can guarantee a score which
is arbitrarily close to zero as $S$ becomes large. It is able to do this
despite having only $O(X \log_2S)$ space complexity and negligible time
complexity. The results take advantage of certain properties of the stationary
distributions of Markov chains.
|
Deeply AggreVaTeD: Differentiable Imitation Learning for Sequential
Prediction | Researchers have demonstrated state-of-the-art performance in sequential
decision making problems (e.g., robotics control, sequential prediction) with
deep neural network models. One often has access to near-optimal oracles that
achieve good performance on the task during training. We demonstrate that
AggreVaTeD --- a policy gradient extension of the Imitation Learning (IL)
approach of (Ross & Bagnell, 2014) --- can leverage such an oracle to achieve
faster and better solutions with less training data than a less-informed
Reinforcement Learning (RL) technique. Using both feedforward and recurrent
neural network predictors, we present stochastic gradient procedures on a
sequential prediction task, dependency-parsing from raw image data, as well as
on various high dimensional robotics control problems. We also provide a
comprehensive theoretical study of IL that demonstrates we can expect up to
exponentially lower sample complexity for learning with AggreVaTeD than with RL
algorithms, which backs our empirical findings. Our results and theory indicate
that the proposed approach can achieve superior performance with respect to the
oracle when the demonstrator is sub-optimal.
|
Learning Robot Activities from First-Person Human Videos Using
Convolutional Future Regression | We design a new approach that allows robot learning of new activities from
unlabeled human example videos. Given videos of humans executing the same
activity from a human's viewpoint (i.e., first-person videos), our objective is
to make the robot learn the temporal structure of the activity as its future
regression network, and learn to transfer such model for its own motor
execution. We present a new deep learning model: We extend the state-of-the-art
convolutional object detection network for the representation/estimation of
human hands in training videos, and newly introduce the concept of using a
fully convolutional network to regress (i.e., predict) the intermediate scene
representation corresponding to the future frame (e.g., 1-2 seconds later).
Combining these allows direct prediction of future locations of human hands and
objects, which enables the robot to infer the motor control plan using our
manipulation network. We experimentally confirm that our approach makes
learning of robot activities from unlabeled human interaction videos possible,
and demonstrate that our robot is able to execute the learned collaborative
activities in real-time directly based on its camera input.
|
Adversarial Examples for Semantic Image Segmentation | Machine learning methods in general and Deep Neural Networks in particular
have shown to be vulnerable to adversarial perturbations. So far this
phenomenon has mainly been studied in the context of whole-image
classification. In this contribution, we analyse how adversarial perturbations
can affect the task of semantic segmentation. We show how existing adversarial
attackers can be transferred to this task and that it is possible to create
imperceptible adversarial perturbations that lead a deep network to misclassify
almost all pixels of a chosen class while leaving network prediction nearly
unchanged outside this class.
|
Differentially Private Bayesian Learning on Distributed Data | Many applications of machine learning, for example in health care, would
benefit from methods that can guarantee privacy of data subjects. Differential
privacy (DP) has become established as a standard for protecting learning
results. The standard DP algorithms require a single trusted party to have
access to the entire data, which is a clear weakness. We consider DP Bayesian
learning in a distributed setting, where each party only holds a single sample
or a few samples of the data. We propose a learning strategy based on a secure
multi-party sum function for aggregating summaries from data holders and the
Gaussian mechanism for DP. Our method builds on an asymptotically optimal and
practically efficient DP Bayesian inference with rapidly diminishing extra
cost.
|
On the Behavior of Convolutional Nets for Feature Extraction | Deep neural networks are representation learning techniques. During training,
a deep net is capable of generating a descriptive language of unprecedented
size and detail in machine learning. Extracting the descriptive language coded
within a trained CNN model (in the case of image data), and reusing it for
other purposes is a field of interest, as it provides access to the visual
descriptors previously learnt by the CNN after processing millions of images,
without requiring an expensive training phase. Contributions to this field
(commonly known as feature representation transfer or transfer learning) have
been purely empirical so far, extracting all CNN features from a single layer
close to the output and testing their performance by feeding them to a
classifier. This approach has provided consistent results, although its
relevance is limited to classification tasks. In a completely different
approach, in this paper we statistically measure the discriminative power of
every single feature found within a deep CNN, when used for characterizing
every class of 11 datasets. We seek to provide new insights into the behavior
of CNN features, particularly the ones from convolutional layers, as this can
be relevant for their application to knowledge representation and reasoning.
Our results confirm that low and middle level features may behave differently
to high level features, but only under certain conditions. We find that all CNN
features can be used for knowledge representation purposes both by their
presence or by their absence, doubling the information a single CNN feature may
provide. We also study how much noise these features may include, and propose a
thresholding approach to discard most of it. All these insights have a direct
application to the generation of CNN embedding spaces.
|
Dynamic State Warping | The ubiquity of sequences in many domains enhances significant recent
interest in sequence learning, for which a basic problem is how to measure the
distance between sequences. Dynamic time warping (DTW) aligns two sequences by
nonlinear local warping and returns a distance value. DTW shows superior
ability in many applications, e.g. video, image, etc. However, in DTW, two
points are paired essentially based on point-to-point Euclidean distance (ED)
without considering the autocorrelation of sequences. Thus, points with
different semantic meanings, e.g. peaks and valleys, may be matched providing
their coordinate values are similar. As a result, DTW is sensitive to noise and
poorly interpretable. This paper proposes an efficient and flexible sequence
alignment algorithm, dynamic state warping (DSW). DSW converts each time point
into a latent state, which endows point-wise autocorrelation information.
Alignment is performed by using the state sequences. Thus DSW is able to yield
alignment that is semantically more interpretable than that of DTW. Using one
nearest neighbor classifier, DSW shows significant improvement on
classification accuracy in comparison to ED (70/85 wins) and DTW (74/85 wins).
We also empirically demonstrate that DSW is more robust and scales better to
long sequences than ED and DTW.
|
Learning Identifiable Gaussian Bayesian Networks in Polynomial Time and
Sample Complexity | Learning the directed acyclic graph (DAG) structure of a Bayesian network
from observational data is a notoriously difficult problem for which many
hardness results are known. In this paper we propose a provably polynomial-time
algorithm for learning sparse Gaussian Bayesian networks with equal noise
variance --- a class of Bayesian networks for which the DAG structure can be
uniquely identified from observational data --- under high-dimensional
settings. We show that $O(k^4 \log p)$ number of samples suffices for our
method to recover the true DAG structure with high probability, where $p$ is
the number of variables and $k$ is the maximum Markov blanket size. We obtain
our theoretical guarantees under a condition called Restricted Strong Adjacency
Faithfulness, which is strictly weaker than strong faithfulness --- a condition
that other methods based on conditional independence testing need for their
success. The sample complexity of our method matches the information-theoretic
limits in terms of the dependence on $p$. We show that our method out-performs
existing state-of-the-art methods for learning Gaussian Bayesian networks in
terms of recovering the true DAG structure while being comparable in speed to
heuristic methods.
|
Stochastic Separation Theorems | The problem of non-iterative one-shot and non-destructive correction of
unavoidable mistakes arises in all Artificial Intelligence applications in the
real world. Its solution requires robust separation of samples with errors from
samples where the system works properly. We demonstrate that in (moderately)
high dimension this separation could be achieved with probability close to one
by linear discriminants. Surprisingly, separation of a new image from a very
large set of known images is almost always possible even in moderately high
dimensions by linear functionals, and coefficients of these functionals can be
found explicitly. Based on fundamental properties of measure concentration, we
show that for $M<a\exp(b{n})$ random $M$-element sets in $\mathbb{R}^n$ are
linearly separable with probability $p$, $p>1-\vartheta$, where $1>\vartheta>0$
is a given small constant. Exact values of $a,b>0$ depend on the probability
distribution that determines how the random $M$-element sets are drawn, and on
the constant $\vartheta$. These {\em stochastic separation theorems} provide a
new instrument for the development, analysis, and assessment of machine
learning methods and algorithms in high dimension. Theoretical statements are
illustrated with numerical examples.
|
Learning Graphical Games from Behavioral Data: Sufficient and Necessary
Conditions | In this paper we obtain sufficient and necessary conditions on the number of
samples required for exact recovery of the pure-strategy Nash equilibria (PSNE)
set of a graphical game from noisy observations of joint actions. We consider
sparse linear influence games --- a parametric class of graphical games with
linear payoffs, and represented by directed graphs of n nodes (players) and
in-degree of at most k. We show that one can efficiently recover the PSNE set
of a linear influence game with $O(k^2 \log n)$ samples, under very general
observation models. On the other hand, we show that $\Omega(k \log n)$ samples
are necessary for any procedure to recover the PSNE set from observations of
joint actions.
|
Denoising Adversarial Autoencoders | Unsupervised learning is of growing interest because it unlocks the potential
held in vast amounts of unlabelled data to learn useful representations for
inference. Autoencoders, a form of generative model, may be trained by learning
to reconstruct unlabelled input data from a latent representation space. More
robust representations may be produced by an autoencoder if it learns to
recover clean input samples from corrupted ones. Representations may be further
improved by introducing regularisation during training to shape the
distribution of the encoded data in latent space. We suggest denoising
adversarial autoencoders, which combine denoising and regularisation, shaping
the distribution of latent space using adversarial training. We introduce a
novel analysis that shows how denoising may be incorporated into the training
and sampling of adversarial autoencoders. Experiments are performed to assess
the contributions that denoising makes to the learning of representations for
classification and sample synthesis. Our results suggest that autoencoders
trained using a denoising criterion achieve higher classification performance,
and can synthesise samples that are more consistent with the input data than
those trained without a corruption process.
|
Virtual vs. Real: Trading Off Simulations and Physical Experiments in
Reinforcement Learning with Bayesian Optimization | In practice, the parameters of control policies are often tuned manually.
This is time-consuming and frustrating. Reinforcement learning is a promising
alternative that aims to automate this process, yet often requires too many
experiments to be practical. In this paper, we propose a solution to this
problem by exploiting prior knowledge from simulations, which are readily
available for most robotic platforms. Specifically, we extend Entropy Search, a
Bayesian optimization algorithm that maximizes information gain from each
experiment, to the case of multiple information sources. The result is a
principled way to automatically combine cheap, but inaccurate information from
simulations with expensive and accurate physical experiments in a
cost-effective manner. We apply the resulting method to a cart-pole system,
which confirms that the algorithm can find good control policies with fewer
experiments than standard Bayesian optimization on the physical system only.
|
Machine Learning on Sequential Data Using a Recurrent Weighted Average | Recurrent Neural Networks (RNN) are a type of statistical model designed to
handle sequential data. The model reads a sequence one symbol at a time. Each
symbol is processed based on information collected from the previous symbols.
With existing RNN architectures, each symbol is processed using only
information from the previous processing step. To overcome this limitation, we
propose a new kind of RNN model that computes a recurrent weighted average
(RWA) over every past processing step. Because the RWA can be computed as a
running average, the computational overhead scales like that of any other RNN
architecture. The approach essentially reformulates the attention mechanism
into a stand-alone model. The performance of the RWA model is assessed on the
variable copy problem, the adding problem, classification of artificial
grammar, classification of sequences by length, and classification of the MNIST
images (where the pixels are read sequentially one at a time). On almost every
task, the RWA model is found to outperform a standard LSTM model.
|
EX2: Exploration with Exemplar Models for Deep Reinforcement Learning | Deep reinforcement learning algorithms have been shown to learn complex tasks
using highly general policy classes. However, sparse reward problems remain a
significant challenge. Exploration methods based on novelty detection have been
particularly successful in such settings but typically require generative or
predictive models of the observations, which can be difficult to train when the
observations are very high-dimensional and complex, as in the case of raw
images. We propose a novelty detection algorithm for exploration that is based
entirely on discriminatively trained exemplar models, where classifiers are
trained to discriminate each visited state against all others. Intuitively,
novel states are easier to distinguish against other states seen during
training. We show that this kind of discriminative modeling corresponds to
implicit density estimation, and that it can be combined with count-based
exploration to produce competitive results on a range of popular benchmark
tasks, including state-of-the-art results on challenging egocentric
observations in the vizDoom benchmark.
|
Multi-step Reinforcement Learning: A Unifying Algorithm | Unifying seemingly disparate algorithmic ideas to produce better performing
algorithms has been a longstanding goal in reinforcement learning. As a primary
example, TD($\lambda$) elegantly unifies one-step TD prediction with Monte
Carlo methods through the use of eligibility traces and the trace-decay
parameter $\lambda$. Currently, there are a multitude of algorithms that can be
used to perform TD control, including Sarsa, $Q$-learning, and Expected Sarsa.
These methods are often studied in the one-step case, but they can be extended
across multiple time steps to achieve better performance. Each of these
algorithms is seemingly distinct, and no one dominates the others for all
problems. In this paper, we study a new multi-step action-value algorithm
called $Q(\sigma)$ which unifies and generalizes these existing algorithms,
while subsuming them as special cases. A new parameter, $\sigma$, is introduced
to allow the degree of sampling performed by the algorithm at each step during
its backup to be continuously varied, with Sarsa existing at one extreme (full
sampling), and Expected Sarsa existing at the other (pure expectation).
$Q(\sigma)$ is generally applicable to both on- and off-policy learning, but in
this work we focus on experiments in the on-policy case. Our results show that
an intermediate value of $\sigma$, which results in a mixture of the existing
algorithms, performs better than either extreme. The mixture can also be varied
dynamically which can result in even greater performance.
|
Generative Poisoning Attack Method Against Neural Networks | Poisoning attack is identified as a severe security threat to machine
learning algorithms. In many applications, for example, deep neural network
(DNN) models collect public data as the inputs to perform re-training, where
the input data can be poisoned. Although poisoning attack against support
vector machines (SVM) has been extensively studied before, there is still very
limited knowledge about how such attack can be implemented on neural networks
(NN), especially DNNs. In this work, we first examine the possibility of
applying traditional gradient-based method (named as the direct gradient
method) to generate poisoned data against NNs by leveraging the gradient of the
target model w.r.t. the normal data. We then propose a generative method to
accelerate the generation rate of the poisoned data: an auto-encoder
(generator) used to generate poisoned data is updated by a reward function of
the loss, and the target NN model (discriminator) receives the poisoned data to
calculate the loss w.r.t. the normal data. Our experiment results show that the
generative method can speed up the poisoned data generation rate by up to
239.38x compared with the direct gradient method, with slightly lower model
accuracy degradation. A countermeasure is also designed to detect such
poisoning attack methods by checking the loss of the target model.
|
Contextual Linear Bandits under Noisy Features: Towards Bayesian Oracles | We study contextual linear bandit problems under uncertainty on features;
they are noisy with missing entries. To address the challenges from the noise,
we analyze Bayesian oracles given observed noisy features. Our Bayesian
analysis finds that the optimal hypothesis can be far from the underlying
realizability function, depending on noise characteristics, which is highly
non-intuitive and does not occur for classical noiseless setups. This implies
that classical approaches cannot guarantee a non-trivial regret bound. We thus
propose an algorithm aiming at the Bayesian oracle from observed information
under this model, achieving $\tilde{O}(d\sqrt{T})$ regret bound with respect to
feature dimension $d$ and time horizon $T$. We demonstrate the proposed
algorithm using synthetic and real-world datasets.
|
Convex Geometry of the Generalized Matrix-Fractional Function | Generalized matrix-fractional (GMF) functions are a class of matrix support
functions introduced by Burke and Hoheisel as a tool for unifying a range of
seemingly divergent matrix optimization problems associated with inverse
problems, regularization and learning. In this paper we dramatically simplify
the support function representation for GMF functions as well as the
representation of their subdifferentials. These new representations allow the
ready computation of a range of important related geometric objects whose
formulations were previously unavailable.
|
Axiomatic Attribution for Deep Networks | We study the problem of attributing the prediction of a deep network to its
input features, a problem previously studied by several other works. We
identify two fundamental axioms---Sensitivity and Implementation Invariance
that attribution methods ought to satisfy. We show that they are not satisfied
by most known attribution methods, which we consider to be a fundamental
weakness of those methods. We use the axioms to guide the design of a new
attribution method called Integrated Gradients. Our method requires no
modification to the original network and is extremely simple to implement; it
just needs a few calls to the standard gradient operator. We apply this method
to a couple of image models, a couple of text models and a chemistry model,
demonstrating its ability to debug networks, to extract rules from a network,
and to enable users to engage with models better.
|
Recurrent Poisson Factorization for Temporal Recommendation | Poisson factorization is a probabilistic model of users and items for
recommendation systems, where the so-called implicit consumer data is modeled
by a factorized Poisson distribution. There are many variants of Poisson
factorization methods who show state-of-the-art performance on real-world
recommendation tasks. However, most of them do not explicitly take into account
the temporal behavior and the recurrent activities of users which is essential
to recommend the right item to the right user at the right time. In this paper,
we introduce Recurrent Poisson Factorization (RPF) framework that generalizes
the classical PF methods by utilizing a Poisson process for modeling the
implicit feedback. RPF treats time as a natural constituent of the model and
brings to the table a rich family of time-sensitive factorization models. To
elaborate, we instantiate several variants of RPF who are capable of handling
dynamic user preferences and item specification (DRPF), modeling the
social-aspect of product adoption (SRPF), and capturing the consumption
heterogeneity among users and items (HRPF). We also develop a variational
algorithm for approximate posterior inference that scales up to massive data
sets. Furthermore, we demonstrate RPF's superior performance over many
state-of-the-art methods on synthetic dataset, and large scale real-world
datasets on music streaming logs, and user-item interactions in M-Commerce
platforms.
|
Learning Deep Matrix Representations | We present a new distributed representation in deep neural nets wherein the
information is represented in native form as a matrix. This differs from
current neural architectures that rely on vector representations. We consider
matrices as central to the architecture and they compose the input, hidden and
output layers. The model representation is more compact and elegant -- the
number of parameters grows only with the largest dimension of the incoming
layer rather than the number of hidden units. We derive several new deep
networks: (i) feed-forward nets that map an input matrix into an output matrix,
(ii) recurrent nets which map a sequence of input matrices into a sequence of
output matrices. We also reinterpret existing models for (iii) memory-augmented
networks and (iv) graphs using matrix notations. For graphs we demonstrate how
the new notations lead to simple but effective extensions with multiple
attentions. Extensive experiments on handwritten digits recognition, face
reconstruction, sequence to sequence learning, EEG classification, and
graph-based node classification demonstrate the efficacy and compactness of the
matrix architectures.
|
Chain-NN: An Energy-Efficient 1D Chain Architecture for Accelerating
Deep Convolutional Neural Networks | Deep convolutional neural networks (CNN) have shown their good performances
in many computer vision tasks. However, the high computational complexity of
CNN involves a huge amount of data movements between the computational
processor core and memory hierarchy which occupies the major of the power
consumption. This paper presents Chain-NN, a novel energy-efficient 1D chain
architecture for accelerating deep CNNs. Chain-NN consists of the dedicated
dual-channel process engines (PE). In Chain-NN, convolutions are done by the 1D
systolic primitives composed of a group of adjacent PEs. These systolic
primitives, together with the proposed column-wise scan input pattern, can
fully reuse input operand to reduce the memory bandwidth requirement for energy
saving. Moreover, the 1D chain architecture allows the systolic primitives to
be easily reconfigured according to specific CNN parameters with fewer design
complexity. The synthesis and layout of Chain-NN is under TSMC 28nm process. It
costs 3751k logic gates and 352KB on-chip memory. The results show a 576-PE
Chain-NN can be scaled up to 700MHz. This achieves a peak throughput of
806.4GOPS with 567.5mW and is able to accelerate the five convolutional layers
in AlexNet at a frame rate of 326.2fps. 1421.0GOPS/W power efficiency is at
least 2.5 to 4.1x times better than the state-of-the-art works.
|
Adversarial Generation of Real-time Feedback with Neural Networks for
Simulation-based Training | Simulation-based training (SBT) is gaining popularity as a low-cost and
convenient training technique in a vast range of applications. However, for a
SBT platform to be fully utilized as an effective training tool, it is
essential that feedback on performance is provided automatically in real-time
during training. It is the aim of this paper to develop an efficient and
effective feedback generation method for the provision of real-time feedback in
SBT. Existing methods either have low effectiveness in improving novice skills
or suffer from low efficiency, resulting in their inability to be used in
real-time. In this paper, we propose a neural network based method to generate
feedback using the adversarial technique. The proposed method utilizes a
bounded adversarial update to minimize a L1 regularized loss via
back-propagation. We empirically show that the proposed method can be used to
generate simple, yet effective feedback. Also, it was observed to have high
effectiveness and efficiency when compared to existing methods, thus making it
a promising option for real-time feedback generation in SBT.
|
Addressing Appearance Change in Outdoor Robotics with Adversarial Domain
Adaptation | Appearance changes due to weather and seasonal conditions represent a strong
impediment to the robust implementation of machine learning systems in outdoor
robotics. While supervised learning optimises a model for the training domain,
it will deliver degraded performance in application domains that underlie
distributional shifts caused by these changes. Traditionally, this problem has
been addressed via the collection of labelled data in multiple domains or by
imposing priors on the type of shift between both domains. We frame the problem
in the context of unsupervised domain adaptation and develop a framework for
applying adversarial techniques to adapt popular, state-of-the-art network
architectures with the additional objective to align features across domains.
Moreover, as adversarial training is notoriously unstable, we first perform an
extensive ablation study, adapting many techniques known to stabilise
generative adversarial networks, and evaluate on a surrogate classification
task with the same appearance change. The distilled insights are applied to the
problem of free-space segmentation for motion planning in autonomous driving.
|
Sharp bounds for population recovery | The population recovery problem is a basic problem in noisy unsupervised
learning that has attracted significant research attention in recent years
[WY12,DRWY12, MS13, BIMP13, LZ15,DST16]. A number of different variants of this
problem have been studied, often under assumptions on the unknown distribution
(such as that it has restricted support size). In this work we study the sample
complexity and algorithmic complexity of the most general version of the
problem, under both bit-flip noise and erasure noise model. We give essentially
matching upper and lower sample complexity bounds for both noise models, and
efficient algorithms matching these sample complexity bounds up to polynomial
factors.
|
Machine Learning Friendly Set Version of Johnson-Lindenstrauss Lemma | In this paper we make a novel use of the Johnson-Lindenstrauss Lemma. The
Lemma has an existential form saying that there exists a JL transformation $f$
of the data points into lower dimensional space such that all of them fall into
predefined error range $\delta$.
We formulate in this paper a theorem stating that we can choose the target
dimensionality in a random projection type JL linear transformation in such a
way that with probability $1-\epsilon$ all of them fall into predefined error
range $\delta$ for any user-predefined failure probability $\epsilon$.
This result is important for applications such a data clustering where we
want to have a priori dimensionality reducing transformation instead of trying
out a (large) number of them, as with traditional Johnson-Lindenstrauss Lemma.
In particular, we take a closer look at the $k$-means algorithm and prove that
a good solution in the projected space is also a good solution in the original
space. Furthermore, under proper assumptions local optima in the original space
are also ones in the projected space. We define also conditions for which
clusterability property of the original space is transmitted to the projected
space, so that special case algorithms for the original space are also
applicable in the projected space.
|
Using Graphs of Classifiers to Impose Declarative Constraints on
Semi-supervised Learning | We propose a general approach to modeling semi-supervised learning (SSL)
algorithms. Specifically, we present a declarative language for modeling both
traditional supervised classification tasks and many SSL heuristics, including
both well-known heuristics such as co-training and novel domain-specific
heuristics. In addition to representing individual SSL heuristics, we show that
multiple heuristics can be automatically combined using Bayesian optimization
methods. We experiment with two classes of tasks, link-based text
classification and relation extraction. We show modest improvements on
well-studied link-based classification benchmarks, and state-of-the-art results
on relation-extraction tasks for two realistic domains.
|
LR-GAN: Layered Recursive Generative Adversarial Networks for Image
Generation | We present LR-GAN: an adversarial image generation model which takes scene
structure and context into account. Unlike previous generative adversarial
networks (GANs), the proposed GAN learns to generate image background and
foregrounds separately and recursively, and stitch the foregrounds on the
background in a contextually relevant manner to produce a complete natural
image. For each foreground, the model learns to generate its appearance, shape
and pose. The whole model is unsupervised, and is trained in an end-to-end
manner with gradient descent methods. The experiments demonstrate that LR-GAN
can generate more natural images with objects that are more human recognizable
than DCGAN.
|
Graph sampling with determinantal processes | We present a new random sampling strategy for k-bandlimited signals defined
on graphs, based on determinantal point processes (DPP). For small graphs, ie,
in cases where the spectrum of the graph is accessible, we exhibit a DPP
sampling scheme that enables perfect recovery of bandlimited signals. For large
graphs, ie, in cases where the graph's spectrum is not accessible, we
investigate, both theoretically and empirically, a sub-optimal but much faster
DPP based on loop-erased random walks on the graph. Preliminary experiments
show promising results especially in cases where the number of measurements
should stay as small as possible and for graphs that have a strong community
structure. Our sampling scheme is efficient and can be applied to graphs with
up to $10^6$ nodes.
|
A Theory of Output-Side Unsupervised Domain Adaptation | When learning a mapping from an input space to an output space, the
assumption that the sample distribution of the training data is the same as
that of the test data is often violated. Unsupervised domain shift methods
adapt the learned function in order to correct for this shift. Previous work
has focused on utilizing unlabeled samples from the target distribution. We
consider the complementary problem in which the unlabeled samples are given
post mapping, i.e., we are given the outputs of the mapping of unknown samples
from the shifted domain. Two other variants are also studied: the two sided
version, in which unlabeled samples are give from both the input and the output
spaces, and the Domain Transfer problem, which was recently formalized. In all
cases, we derive generalization bounds that employ discrepancy terms.
|
Improving Regret Bounds for Combinatorial Semi-Bandits with
Probabilistically Triggered Arms and Its Applications | We study combinatorial multi-armed bandit with probabilistically triggered
arms (CMAB-T) and semi-bandit feedback. We resolve a serious issue in the prior
CMAB-T studies where the regret bounds contain a possibly exponentially large
factor of $1/p^*$, where $p^*$ is the minimum positive probability that an arm
is triggered by any action. We address this issue by introducing a triggering
probability modulated (TPM) bounded smoothness condition into the general
CMAB-T framework, and show that many applications such as influence
maximization bandit and combinatorial cascading bandit satisfy this TPM
condition. As a result, we completely remove the factor of $1/p^*$ from the
regret bounds, achieving significantly better regret bounds for influence
maximization and cascading bandits than before. Finally, we provide lower bound
results showing that the factor $1/p^*$ is unavoidable for general CMAB-T
problems, suggesting that the TPM condition is crucial in removing this factor.
|
Neural Machine Translation and Sequence-to-sequence Models: A Tutorial | This tutorial introduces a new and powerful set of techniques variously
called "neural machine translation" or "neural sequence-to-sequence models".
These techniques have been used in a number of tasks regarding the handling of
human language, and can be a powerful tool in the toolbox of anyone who wants
to model sequential data of some sort. The tutorial assumes that the reader
knows the basics of math and programming, but does not assume any particular
experience with neural networks or natural language processing. It attempts to
explain the intuition behind the various methods covered, then delves into them
with enough mathematical detail to understand them concretely, and culiminates
with a suggestion for an implementation exercise, where readers can test that
they understood the content in practice.
|
Data-Dependent Stability of Stochastic Gradient Descent | We establish a data-dependent notion of algorithmic stability for Stochastic
Gradient Descent (SGD), and employ it to develop novel generalization bounds.
This is in contrast to previous distribution-free algorithmic stability results
for SGD which depend on the worst-case constants. By virtue of the
data-dependent argument, our bounds provide new insights into learning with SGD
on convex and non-convex problems. In the convex case, we show that the bound
on the generalization error depends on the risk at the initialization point. In
the non-convex case, we prove that the expected curvature of the objective
function around the initialization point has crucial influence on the
generalization error. In both cases, our results suggest a simple data-driven
strategy to stabilize SGD by pre-screening its initialization. As a corollary,
our results allow us to show optimistic generalization bounds that exhibit fast
convergence rates for SGD subject to a vanishing empirical risk and low noise
of stochastic gradient.
|
Multi-Objective Non-parametric Sequential Prediction | Online-learning research has mainly been focusing on minimizing one objective
function. In many real-world applications, however, several objective functions
have to be considered simultaneously. Recently, an algorithm for dealing with
several objective functions in the i.i.d. case has been presented. In this
paper, we extend the multi-objective framework to the case of stationary and
ergodic processes, thus allowing dependencies among observations. We first
identify an asymptomatic lower bound for any prediction strategy and then
present an algorithm whose predictions achieve the optimal solution while
fulfilling any continuous and convex constraining criterion.
|
Third-Person Imitation Learning | Reinforcement learning (RL) makes it possible to train agents capable of
achieving sophisticated goals in complex and uncertain environments. A key
difficulty in reinforcement learning is specifying a reward function for the
agent to optimize. Traditionally, imitation learning in RL has been used to
overcome this problem. Unfortunately, hitherto imitation learning methods tend
to require that demonstrations are supplied in the first-person: the agent is
provided with a sequence of states and a specification of the actions that it
should have taken. While powerful, this kind of imitation learning is limited
by the relatively hard problem of collecting first-person demonstrations.
Humans address this problem by learning from third-person demonstrations: they
observe other humans perform tasks, infer the task, and accomplish the same
task themselves.
In this paper, we present a method for unsupervised third-person imitation
learning. Here third-person refers to training an agent to correctly achieve a
simple goal in a simple environment when it is provided a demonstration of a
teacher achieving the same goal but from a different viewpoint; and
unsupervised refers to the fact that the agent receives only these third-person
demonstrations, and is not provided a correspondence between teacher states and
student states. Our methods primary insight is that recent advances from domain
confusion can be utilized to yield domain agnostic features which are crucial
during the training process. To validate our approach, we report successful
experiments on learning from third-person demonstrations in a pointmass domain,
a reacher domain, and inverted pendulum.
|
Measuring Sample Quality with Kernels | Approximate Markov chain Monte Carlo (MCMC) offers the promise of more rapid
sampling at the cost of more biased inference. Since standard MCMC diagnostics
fail to detect these biases, researchers have developed computable Stein
discrepancy measures that provably determine the convergence of a sample to its
target distribution. This approach was recently combined with the theory of
reproducing kernels to define a closed-form kernel Stein discrepancy (KSD)
computable by summing kernel evaluations across pairs of sample points. We
develop a theory of weak convergence for KSDs based on Stein's method,
demonstrate that commonly used KSDs fail to detect non-convergence even for
Gaussian targets, and show that kernels with slowly decaying tails provably
determine convergence for a large class of target distributions. The resulting
convergence-determining KSDs are suitable for comparing biased, exact, and
deterministic sample sequences and simpler to compute and parallelize than
alternative Stein discrepancies. We use our tools to compare biased samplers,
select sampler hyperparameters, and improve upon existing KSD approaches to
one-sample hypothesis testing and sample quality improvement.
|
Surprise-Based Intrinsic Motivation for Deep Reinforcement Learning | Exploration in complex domains is a key challenge in reinforcement learning,
especially for tasks with very sparse rewards. Recent successes in deep
reinforcement learning have been achieved mostly using simple heuristic
exploration strategies such as $\epsilon$-greedy action selection or Gaussian
control noise, but there are many tasks where these methods are insufficient to
make any learning progress. Here, we consider more complex heuristics:
efficient and scalable exploration strategies that maximize a notion of an
agent's surprise about its experiences via intrinsic motivation. We propose to
learn a model of the MDP transition probabilities concurrently with the policy,
and to form intrinsic rewards that approximate the KL-divergence of the true
transition probabilities from the learned model. One of our approximations
results in using surprisal as intrinsic motivation, while the other gives the
$k$-step learning progress. We show that our incentives enable agents to
succeed in a wide range of environments with high-dimensional state spaces and
very sparse rewards, including continuous control tasks and games in the Atari
RAM domain, outperforming several other heuristic exploration techniques.
|
A Correlative Denoising Autoencoder to Model Social Influence for Top-N
Recommender System | In recent years, there are numerous works been proposed to leverage the
techniques of deep learning to improve social-aware recommendation performance.
In most cases, it requires a larger number of data to train a robust deep
learning model, which contains a lot of parameters to fit training data.
However, both data of user ratings and social networks are facing critical
sparse problem, which makes it not easy to train a robust deep neural network
model. Towards this problem, we propose a novel Correlative Denoising
Autoencoder (CoDAE) method by taking correlations between users with multiple
roles into account to learn robust representations from sparse inputs of
ratings and social networks for recommendation. We develop the CoDAE model by
utilizing three separated autoencoders to learn user features with roles of
rater, truster and trustee, respectively. Especially, on account of that each
input unit of user vectors with roles of truster and trustee is corresponding
to a particular user, we propose to utilize shared parameters to learn common
information of the units that corresponding to same users. Moreover, we propose
a related regularization term to learn correlations between user features that
learnt by the three subnetworks of CoDAE model. We further conduct a series of
experiments to evaluate the proposed method on two public datasets for Top-N
recommendation task. The experimental results demonstrate that the proposed
model outperforms state-of-the-art algorithms on rank-sensitive metrics of MAP
and NDCG.
|
Building a Regular Decision Boundary with Deep Networks | In this work, we build a generic architecture of Convolutional Neural
Networks to discover empirical properties of neural networks. Our first
contribution is to introduce a state-of-the-art framework that depends upon few
hyper parameters and to study the network when we vary them. It has no max
pooling, no biases, only 13 layers, is purely convolutional and yields up to
95.4% and 79.6% accuracy respectively on CIFAR10 and CIFAR100. We show that the
nonlinearity of a deep network does not need to be continuous, non expansive or
point-wise, to achieve good performance. We show that increasing the width of
our network permits being competitive with very deep networks. Our second
contribution is an analysis of the contraction and separation properties of
this network. Indeed, a 1-nearest neighbor classifier applied on deep features
progressively improves with depth, which indicates that the representation is
progressively more regular. Besides, we defined and analyzed local support
vectors that separate classes locally. All our experiments are reproducible and
code is available online, based on TensorFlow.
|
Mean teachers are better role models: Weight-averaged consistency
targets improve semi-supervised deep learning results | The recently proposed Temporal Ensembling has achieved state-of-the-art
results in several semi-supervised learning benchmarks. It maintains an
exponential moving average of label predictions on each training example, and
penalizes predictions that are inconsistent with this target. However, because
the targets change only once per epoch, Temporal Ensembling becomes unwieldy
when learning large datasets. To overcome this problem, we propose Mean
Teacher, a method that averages model weights instead of label predictions. As
an additional benefit, Mean Teacher improves test accuracy and enables training
with fewer labels than Temporal Ensembling. Without changing the network
architecture, Mean Teacher achieves an error rate of 4.35% on SVHN with 250
labels, outperforming Temporal Ensembling trained with 1000 labels. We also
show that a good network architecture is crucial to performance. Combining Mean
Teacher and Residual Networks, we improve the state of the art on CIFAR-10 with
4000 labels from 10.55% to 6.28%, and on ImageNet 2012 with 10% of the labels
from 35.24% to 9.11%.
|
Sample-level Deep Convolutional Neural Networks for Music Auto-tagging
Using Raw Waveforms | Recently, the end-to-end approach that learns hierarchical representations
from raw data using deep convolutional neural networks has been successfully
explored in the image, text and speech domains. This approach was applied to
musical signals as well but has been not fully explored yet. To this end, we
propose sample-level deep convolutional neural networks which learn
representations from very small grains of waveforms (e.g. 2 or 3 samples)
beyond typical frame-level input representations. Our experiments show how deep
architectures with sample-level filters improve the accuracy in music
auto-tagging and they provide results comparable to previous state-of-the-art
performances for the Magnatagatune dataset and Million Song Dataset. In
addition, we visualize filters learned in a sample-level DCNN in each layer to
identify hierarchically learned features and show that they are sensitive to
log-scaled frequency along layer, such as mel-frequency spectrogram that is
widely used in music classification systems.
|
Multi-Level and Multi-Scale Feature Aggregation Using Pre-trained
Convolutional Neural Networks for Music Auto-tagging | Music auto-tagging is often handled in a similar manner to image
classification by regarding the 2D audio spectrogram as image data. However,
music auto-tagging is distinguished from image classification in that the tags
are highly diverse and have different levels of abstractions. Considering this
issue, we propose a convolutional neural networks (CNN)-based architecture that
embraces multi-level and multi-scaled features. The architecture is trained in
three steps. First, we conduct supervised feature learning to capture local
audio features using a set of CNNs with different input sizes. Second, we
extract audio features from each layer of the pre-trained convolutional
networks separately and aggregate them altogether given a long audio clip.
Finally, we put them into fully-connected networks and make final predictions
of the tags. Our experiments show that using the combination of multi-level and
multi-scale features is highly effective in music auto-tagging and the proposed
method outperforms previous state-of-the-arts on the MagnaTagATune dataset and
the Million Song Dataset. We further show that the proposed architecture is
useful in transfer learning.
|
Orthogonalized ALS: A Theoretically Principled Tensor Decomposition
Algorithm for Practical Use | The popular Alternating Least Squares (ALS) algorithm for tensor
decomposition is efficient and easy to implement, but often converges to poor
local optima---particularly when the weights of the factors are non-uniform. We
propose a modification of the ALS approach that is as efficient as standard
ALS, but provably recovers the true factors with random initialization under
standard incoherence assumptions on the factors of the tensor. We demonstrate
the significant practical superiority of our approach over traditional ALS for
a variety of tasks on synthetic data---including tensor factorization on exact,
noisy and over-complete tensors, as well as tensor completion---and for
computing word embeddings from a third-order word tri-occurrence tensor.
|
All You Need is Beyond a Good Init: Exploring Better Solution for
Training Extremely Deep Convolutional Neural Networks with Orthonormality and
Modulation | Deep neural network is difficult to train and this predicament becomes worse
as the depth increases. The essence of this problem exists in the magnitude of
backpropagated errors that will result in gradient vanishing or exploding
phenomenon. We show that a variant of regularizer which utilizes orthonormality
among different filter banks can alleviate this problem. Moreover, we design a
backward error modulation mechanism based on the quasi-isometry assumption
between two consecutive parametric layers. Equipped with these two ingredients,
we propose several novel optimization solutions that can be utilized for
training a specific-structured (repetitively triple modules of Conv-BNReLU)
extremely deep convolutional neural network (CNN) WITHOUT any shortcuts/
identity mappings from scratch. Experiments show that our proposed solutions
can achieve distinct improvements for a 44-layer and a 110-layer plain networks
on both the CIFAR-10 and ImageNet datasets. Moreover, we can successfully train
plain CNNs to match the performance of the residual counterparts.
Besides, we propose new principles for designing network structure from the
insights evoked by orthonormality. Combined with residual structure, we achieve
comparative performance on the ImageNet dataset.
|
Decomposable Submodular Function Minimization: Discrete and Continuous | This paper investigates connections between discrete and continuous
approaches for decomposable submodular function minimization. We provide
improved running time estimates for the state-of-the-art continuous algorithms
for the problem using combinatorial arguments. We also provide a systematic
experimental comparison of the two types of methods, based on a clear
distinction between level-0 and level-1 algorithms.
|
Generative and Discriminative Text Classification with Recurrent Neural
Networks | We empirically characterize the performance of discriminative and generative
LSTM models for text classification. We find that although RNN-based generative
models are more powerful than their bag-of-words ancestors (e.g., they account
for conditional dependencies across words in a document), they have higher
asymptotic error rates than discriminatively trained RNN models. However we
also find that generative models approach their asymptotic error rate more
rapidly than their discriminative counterparts---the same pattern that Ng &
Jordan (2001) proved holds for linear classification models that make more
naive conditional independence assumptions. Building on this finding, we
hypothesize that RNN-based generative classification models will be more robust
to shifts in the data distribution. This hypothesis is confirmed in a series of
experiments in zero-shot and continual learning settings that show that
generative models substantially outperform discriminative models.
|
Near-Optimal Closeness Testing of Discrete Histogram Distributions | We investigate the problem of testing the equivalence between two discrete
histograms. A {\em $k$-histogram} over $[n]$ is a probability distribution that
is piecewise constant over some set of $k$ intervals over $[n]$. Histograms
have been extensively studied in computer science and statistics. Given a set
of samples from two $k$-histogram distributions $p, q$ over $[n]$, we want to
distinguish (with high probability) between the cases that $p = q$ and
$\|p-q\|_1 \geq \epsilon$. The main contribution of this paper is a new
algorithm for this testing problem and a nearly matching information-theoretic
lower bound. Specifically, the sample complexity of our algorithm matches our
lower bound up to a logarithmic factor, improving on previous work by
polynomial factors in the relevant parameters. Our algorithmic approach applies
in a more general setting and yields improved sample upper bounds for testing
closeness of other structured distributions as well.
|
Metric Learning for Generalizing Spatial Relations to New Objects | Human-centered environments are rich with a wide variety of spatial relations
between everyday objects. For autonomous robots to operate effectively in such
environments, they should be able to reason about these relations and
generalize them to objects with different shapes and sizes. For example, having
learned to place a toy inside a basket, a robot should be able to generalize
this concept using a spoon and a cup. This requires a robot to have the
flexibility to learn arbitrary relations in a lifelong manner, making it
challenging for an expert to pre-program it with sufficient knowledge to do so
beforehand. In this paper, we address the problem of learning spatial relations
by introducing a novel method from the perspective of distance metric learning.
Our approach enables a robot to reason about the similarity between pairwise
spatial relations, thereby enabling it to use its previous knowledge when
presented with a new relation to imitate. We show how this makes it possible to
learn arbitrary spatial relations from non-expert users using a small number of
examples and in an interactive manner. Our extensive evaluation with real-world
data demonstrates the effectiveness of our method in reasoning about a
continuous spectrum of spatial relations and generalizing them to new objects.
|
Network Inference via the Time-Varying Graphical Lasso | Many important problems can be modeled as a system of interconnected
entities, where each entity is recording time-dependent observations or
measurements. In order to spot trends, detect anomalies, and interpret the
temporal dynamics of such data, it is essential to understand the relationships
between the different entities and how these relationships evolve over time. In
this paper, we introduce the time-varying graphical lasso (TVGL), a method of
inferring time-varying networks from raw time series data. We cast the problem
in terms of estimating a sparse time-varying inverse covariance matrix, which
reveals a dynamic network of interdependencies between the entities. Since
dynamic network inference is a computationally expensive task, we derive a
scalable message-passing algorithm based on the Alternating Direction Method of
Multipliers (ADMM) to solve this problem in an efficient way. We also discuss
several extensions, including a streaming algorithm to update the model and
incorporate new observations in real time. Finally, we evaluate our TVGL
algorithm on both real and synthetic datasets, obtaining interpretable results
and outperforming state-of-the-art baselines in terms of both accuracy and
scalability.
|
Multiplicative Normalizing Flows for Variational Bayesian Neural
Networks | We reinterpret multiplicative noise in neural networks as auxiliary random
variables that augment the approximate posterior in a variational setting for
Bayesian neural networks. We show that through this interpretation it is both
efficient and straightforward to improve the approximation by employing
normalizing flows while still allowing for local reparametrizations and a
tractable lower bound. In experiments we show that with this new approximation
we can significantly improve upon classical mean field for Bayesian neural
networks on both predictive accuracy as well as predictive uncertainty.
|
Max-value Entropy Search for Efficient Bayesian Optimization | Entropy Search (ES) and Predictive Entropy Search (PES) are popular and
empirically successful Bayesian Optimization techniques. Both rely on a
compelling information-theoretic motivation, and maximize the information
gained about the $\arg\max$ of the unknown function; yet, both are plagued by
the expensive computation for estimating entropies. We propose a new criterion,
Max-value Entropy Search (MES), that instead uses the information about the
maximum function value. We show relations of MES to other Bayesian optimization
methods, and establish a regret bound. We observe that MES maintains or
improves the good empirical performance of ES/PES, while tremendously
lightening the computational burden. In particular, MES is much more robust to
the number of samples used for computing the entropy, and hence more efficient
for higher dimensional problems.
|
Concentration Bounds for High Sensitivity Functions Through Differential
Privacy | A new line of work [Dwork et al. STOC 2015], [Hardt and Ullman FOCS 2014],
[Steinke and Ullman COLT 2015], [Bassily et al. STOC 2016] demonstrates how
differential privacy [Dwork et al. TCC 2006] can be used as a mathematical tool
for guaranteeing generalization in adaptive data analysis. Specifically, if a
differentially private analysis is applied on a sample S of i.i.d. examples to
select a low-sensitivity function f, then w.h.p. f(S) is close to its
expectation, although f is being chosen based on the data.
Very recently, Steinke and Ullman observed that these generalization
guarantees can be used for proving concentration bounds in the non-adaptive
setting, where the low-sensitivity function is fixed beforehand. In particular,
they obtain alternative proofs for classical concentration bounds for
low-sensitivity functions, such as the Chernoff bound and McDiarmid's
Inequality.
In this work, we set out to examine the situation for functions with
high-sensitivity, for which differential privacy does not imply generalization
guarantees under adaptive analysis. We show that differential privacy can be
used to prove concentration bounds for such functions in the non-adaptive
setting.
|
Batched High-dimensional Bayesian Optimization via Structural Kernel
Learning | Optimization of high-dimensional black-box functions is an extremely
challenging problem. While Bayesian optimization has emerged as a popular
approach for optimizing black-box functions, its applicability has been limited
to low-dimensional problems due to its computational and statistical challenges
arising from high-dimensional settings. In this paper, we propose to tackle
these challenges by (1) assuming a latent additive structure in the function
and inferring it properly for more efficient and effective BO, and (2)
performing multiple evaluations in parallel to reduce the number of iterations
required by the method. Our novel approach learns the latent structure with
Gibbs sampling and constructs batched queries using determinantal point
processes. Experimental validations on both synthetic and real-world functions
demonstrate that the proposed method outperforms the existing state-of-the-art
approaches.
|
Linear, Machine Learning and Probabilistic Approaches for Time Series
Analysis | In this paper we study different approaches for time series modeling. The
forecasting approaches using linear models, ARIMA alpgorithm, XGBoost machine
learning algorithm are described. Results of different model combinations are
shown. For probabilistic modeling the approaches using copulas and Bayesian
inference are considered.
|
Neural Episodic Control | Deep reinforcement learning methods attain super-human performance in a wide
range of environments. Such methods are grossly inefficient, often taking
orders of magnitudes more data than humans to achieve reasonable performance.
We propose Neural Episodic Control: a deep reinforcement learning agent that is
able to rapidly assimilate new experiences and act upon them. Our agent uses a
semi-tabular representation of the value function: a buffer of past experience
containing slowly changing state representations and rapidly updated estimates
of the value function. We show across a wide range of environments that our
agent learns significantly faster than other state-of-the-art, general purpose
deep reinforcement learning agents.
|
Activation Maximization Generative Adversarial Nets | Class labels have been empirically shown useful in improving the sample
quality of generative adversarial nets (GANs). In this paper, we mathematically
study the properties of the current variants of GANs that make use of class
label information. With class aware gradient and cross-entropy decomposition,
we reveal how class labels and associated losses influence GAN's training.
Based on that, we propose Activation Maximization Generative Adversarial
Networks (AM-GAN) as an advanced solution. Comprehensive experiments have been
conducted to validate our analysis and evaluate the effectiveness of our
solution, where AM-GAN outperforms other strong baselines and achieves
state-of-the-art Inception Score (8.91) on CIFAR-10. In addition, we
demonstrate that, with the Inception ImageNet classifier, Inception Score
mainly tracks the diversity of the generator, and there is, however, no
reliable evidence that it can reflect the true sample quality. We thus propose
a new metric, called AM Score, to provide a more accurate estimation of the
sample quality. Our proposed model also outperforms the baseline methods in the
new metric.
|
Combining Self-Supervised Learning and Imitation for Vision-Based Rope
Manipulation | Manipulation of deformable objects, such as ropes and cloth, is an important
but challenging problem in robotics. We present a learning-based system where a
robot takes as input a sequence of images of a human manipulating a rope from
an initial to goal configuration, and outputs a sequence of actions that can
reproduce the human demonstration, using only monocular images as input. To
perform this task, the robot learns a pixel-level inverse dynamics model of
rope manipulation directly from images in a self-supervised manner, using about
60K interactions with the rope collected autonomously by the robot. The human
demonstration provides a high-level plan of what to do and the low-level
inverse model is used to execute the plan. We show that by combining the high
and low-level plans, the robot can successfully manipulate a rope into a
variety of target shapes using only a sequence of human-provided images for
direction.
|
Cheshire: An Online Algorithm for Activity Maximization in Social
Networks | User engagement in social networks depends critically on the number of online
actions their users take in the network. Can we design an algorithm that finds
when to incentivize users to take actions to maximize the overall activity in a
social network? In this paper, we model the number of online actions over time
using multidimensional Hawkes processes, derive an alternate representation of
these processes based on stochastic differential equations (SDEs) with jumps
and, exploiting this alternate representation, address the above question from
the perspective of stochastic optimal control of SDEs with jumps. We find that
the optimal level of incentivized actions depends linearly on the current level
of overall actions. Moreover, the coefficients of this linear relationship can
be found by solving a matrix Riccati differential equation, which can be solved
efficiently, and a first order differential equation, which has a closed form
solution. As a result, we are able to design an efficient online algorithm,
Cheshire, to sample the optimal times of the users' incentivized actions.
Experiments on both synthetic and real data gathered from Twitter show that our
algorithm is able to consistently maximize the number of online actions more
effectively than the state of the art.
|
On the Expressive Power of Overlapping Architectures of Deep Learning | Expressive efficiency refers to the relation between two architectures A and
B, whereby any function realized by B could be replicated by A, but there
exists functions realized by A, which cannot be replicated by B unless its size
grows significantly larger. For example, it is known that deep networks are
exponentially efficient with respect to shallow networks, in the sense that a
shallow network must grow exponentially large in order to approximate the
functions represented by a deep network of polynomial size. In this work, we
extend the study of expressive efficiency to the attribute of network
connectivity and in particular to the effect of "overlaps" in the convolutional
process, i.e., when the stride of the convolution is smaller than its filter
size (receptive field). To theoretically analyze this aspect of network's
design, we focus on a well-established surrogate for ConvNets called
Convolutional Arithmetic Circuits (ConvACs), and then demonstrate empirically
that our results hold for standard ConvNets as well. Specifically, our analysis
shows that having overlapping local receptive fields, and more broadly denser
connectivity, results in an exponential increase in the expressive capacity of
neural networks. Moreover, while denser connectivity can increase the
expressive capacity, we show that the most common types of modern architectures
already exhibit exponential increase in expressivity, without relying on
fully-connected layers.
|
Guarantees for Greedy Maximization of Non-submodular Functions with
Applications | We investigate the performance of the standard Greedy algorithm for
cardinality constrained maximization of non-submodular nondecreasing set
functions. While there are strong theoretical guarantees on the performance of
Greedy for maximizing submodular functions, there are few guarantees for
non-submodular ones. However, Greedy enjoys strong empirical performance for
many important non-submodular functions, e.g., the Bayesian A-optimality
objective in experimental design. We prove theoretical guarantees supporting
the empirical performance. Our guarantees are characterized by a combination of
the (generalized) curvature $\alpha$ and the submodularity ratio $\gamma$. In
particular, we prove that Greedy enjoys a tight approximation guarantee of
$\frac{1}{\alpha}(1- e^{-\gamma\alpha})$ for cardinality constrained
maximization. In addition, we bound the submodularity ratio and curvature for
several important real-world objectives, including the Bayesian A-optimality
objective, the determinantal function of a square submatrix and certain linear
programs with combinatorial constraints. We experimentally validate our
theoretical findings for both synthetic and real-world applications.
|
Revisiting stochastic off-policy action-value gradients | Off-policy stochastic actor-critic methods rely on approximating the
stochastic policy gradient in order to derive an optimal policy. One may also
derive the optimal policy by approximating the action-value gradient. The use
of action-value gradients is desirable as policy improvement occurs along the
direction of steepest ascent. This has been studied extensively within the
context of natural gradient actor-critic algorithms and more recently within
the context of deterministic policy gradients. In this paper we briefly discuss
the off-policy stochastic counterpart to deterministic action-value gradients,
as well as an incremental approach for following the policy gradient in lieu of
the natural gradient.
|
Classification and clustering for observations of event time data using
non-homogeneous Poisson process models | Data of the form of event times arise in various applications. A simple model
for such data is a non-homogeneous Poisson process (NHPP) which is specified by
a rate function that depends on time. We consider the problem of having access
to multiple independent observations of event time data, observed on a common
interval, from which we wish to classify or cluster the observations according
to their rate functions. Each rate function is unknown but assumed to belong to
a finite number of rate functions each defining a distinct class. We model the
rate functions using a spline basis expansion, the coefficients of which need
to be estimated from data. The classification approach consists of using
training data for which the class membership is known, to calculate maximum
likelihood estimates of the coefficients for each group, then assigning test
observations to a group by a maximum likelihood criterion. For clustering, by
analogy to the Gaussian mixture model approach for Euclidean data, we consider
mixtures of NHPP and use the expectation-maximisation algorithm to estimate the
coefficients of the rate functions for the component models and group
membership probabilities for each observation. The classification and
clustering approaches perform well on both synthetic and real-world data sets.
Code associated with this paper is available at
https://github.com/duncan-barrack/NHPP .
|
Evaluation of Machine Learning Methods to Predict Coronary Artery
Disease Using Metabolomic Data | Metabolomic data can potentially enable accurate, non-invasive and low-cost
prediction of coronary artery disease. Regression-based analytical approaches
however might fail to fully account for interactions between metabolites, rely
on a priori selected input features and thus might suffer from poorer accuracy.
Supervised machine learning methods can potentially be used in order to fully
exploit the dimensionality and richness of the data. In this paper, we
systematically implement and evaluate a set of supervised learning methods (L1
regression, random forest classifier) and compare them to traditional
regression-based approaches for disease prediction using metabolomic data.
|
Contextual Motifs: Increasing the Utility of Motifs using Contextual
Data | Motifs are a powerful tool for analyzing physiological waveform data.
Standard motif methods, however, ignore important contextual information (e.g.,
what the patient was doing at the time the data were collected). We hypothesize
that these additional contextual data could increase the utility of motifs.
Thus, we propose an extension to motifs, contextual motifs, that incorporates
context. Recognizing that, oftentimes, context may be unobserved or
unavailable, we focus on methods to jointly infer motifs and context. Applied
to both simulated and real physiological data, our proposed approach improves
upon existing motif methods in terms of the discriminative utility of the
discovered motifs. In particular, we discovered contextual motifs in continuous
glucose monitor (CGM) data collected from patients with type 1 diabetes.
Compared to their contextless counterparts, these contextual motifs led to
better predictions of hypo- and hyperglycemic events. Our results suggest that
even when inferred, context is useful in both a long- and short-term prediction
horizon when processing and interpreting physiological waveform data.
|
Model-Based Multiple Instance Learning | While Multiple Instance (MI) data are point patterns -- sets or multi-sets of
unordered points -- appropriate statistical point pattern models have not been
used in MI learning. This article proposes a framework for model-based MI
learning using point process theory. Likelihood functions for point pattern
data derived from point process theory enable principled yet conceptually
transparent extensions of learning tasks, such as classification, novelty
detection and clustering, to point pattern data. Furthermore, tractable point
pattern models as well as solutions for learning and decision making from point
pattern data are developed.
|
On the Limits of Learning Representations with Label-Based Supervision | Advances in neural network based classifiers have transformed automatic
feature learning from a pipe dream of stronger AI to a routine and expected
property of practical systems. Since the emergence of AlexNet every winning
submission of the ImageNet challenge has employed end-to-end representation
learning, and due to the utility of good representations for transfer learning,
representation learning has become as an important and distinct task from
supervised learning. At present, this distinction is inconsequential, as
supervised methods are state-of-the-art in learning transferable
representations. But recent work has shown that generative models can also be
powerful agents of representation learning. Will the representations learned
from these generative methods ever rival the quality of those from their
supervised competitors? In this work, we argue in the affirmative, that from an
information theoretic perspective, generative models have greater potential for
representation learning. Based on several experimentally validated assumptions,
we show that supervised learning is upper bounded in its capacity for
representation learning in ways that certain generative models, such as
Generative Adversarial Networks (GANs) are not. We hope that our analysis will
provide a rigorous motivation for further exploration of generative
representation learning.
|
Distance Metric Learning using Graph Convolutional Networks: Application
to Functional Brain Networks | Evaluating similarity between graphs is of major importance in several
computer vision and pattern recognition problems, where graph representations
are often used to model objects or interactions between elements. The choice of
a distance or similarity metric is, however, not trivial and can be highly
dependent on the application at hand. In this work, we propose a novel metric
learning method to evaluate distance between graphs that leverages the power of
convolutional neural networks, while exploiting concepts from spectral graph
theory to allow these operations on irregular graphs. We demonstrate the
potential of our method in the field of connectomics, where neuronal pathways
or functional connections between brain regions are commonly modelled as
graphs. In this problem, the definition of an appropriate graph similarity
function is critical to unveil patterns of disruptions associated with certain
brain disorders. Experimental results on the ABIDE dataset show that our method
can learn a graph similarity metric tailored for a clinical application,
improving the performance of a simple k-nn classifier by 11.9% compared to a
traditional distance metric.
|
Raw Waveform-based Speech Enhancement by Fully Convolutional Networks | This study proposes a fully convolutional network (FCN) model for raw
waveform-based speech enhancement. The proposed system performs speech
enhancement in an end-to-end (i.e., waveform-in and waveform-out) manner, which
dif-fers from most existing denoising methods that process the magnitude
spectrum (e.g., log power spectrum (LPS)) only. Because the fully connected
layers, which are involved in deep neural networks (DNN) and convolutional
neural networks (CNN), may not accurately characterize the local information of
speech signals, particularly with high frequency components, we employed fully
convolutional layers to model the waveform. More specifically, FCN consists of
only convolutional layers and thus the local temporal structures of speech
signals can be efficiently and effectively preserved with relatively few
weights. Experimental results show that DNN- and CNN-based models have limited
capability to restore high frequency components of waveforms, thus leading to
decreased intelligibility of enhanced speech. By contrast, the proposed FCN
model can not only effectively recover the waveforms but also outperform the
LPS-based DNN baseline in terms of short-time objective intelligibility (STOI)
and perceptual evaluation of speech quality (PESQ). In addition, the number of
model parameters in FCN is approximately only 0.2% compared with that in both
DNN and CNN.
|
Triple Generative Adversarial Nets | Generative Adversarial Nets (GANs) have shown promise in image generation and
semi-supervised learning (SSL). However, existing GANs in SSL have two
problems: (1) the generator and the discriminator (i.e. the classifier) may not
be optimal at the same time; and (2) the generator cannot control the semantics
of the generated samples. The problems essentially arise from the two-player
formulation, where a single discriminator shares incompatible roles of
identifying fake samples and predicting labels and it only estimates the data
without considering the labels. To address the problems, we present triple
generative adversarial net (Triple-GAN), which consists of three players---a
generator, a discriminator and a classifier. The generator and the classifier
characterize the conditional distributions between images and labels, and the
discriminator solely focuses on identifying fake image-label pairs. We design
compatible utilities to ensure that the distributions characterized by the
classifier and the generator both converge to the data distribution. Our
results on various datasets demonstrate that Triple-GAN as a unified model can
simultaneously (1) achieve the state-of-the-art classification results among
deep generative models, and (2) disentangle the classes and styles of the input
and transfer smoothly in the data space via interpolation in the latent space
class-conditionally.
|
Deep Robust Kalman Filter | A Robust Markov Decision Process (RMDP) is a sequential decision making model
that accounts for uncertainty in the parameters of dynamic systems. This
uncertainty introduces difficulties in learning an optimal policy, especially
for environments with large state spaces. We propose two algorithms, RTD-DQN
and Deep-RoK, for solving large-scale RMDPs using nonlinear approximation
schemes such as deep neural networks. The RTD-DQN algorithm incorporates the
robust Bellman temporal difference error into a robust loss function, yielding
robust policies for the agent. The Deep-RoK algorithm is a robust Bayesian
method, based on the Extended Kalman Filter (EKF), that accounts for both the
uncertainty in the weights of the approximated value function and the
uncertainty in the transition probabilities, improving the robustness of the
agent. We provide theoretical results for our approach and test the proposed
algorithms on a continuous state domain.
|
Convolutional Recurrent Neural Networks for Bird Audio Detection | Bird sounds possess distinctive spectral structure which may exhibit small
shifts in spectrum depending on the bird species and environmental conditions.
In this paper, we propose using convolutional recurrent neural networks on the
task of automated bird audio detection in real-life environments. In the
proposed method, convolutional layers extract high dimensional, local frequency
shift invariant features, while recurrent layers capture longer term
dependencies between the features extracted from short time frames. This method
achieves 88.5% Area Under ROC Curve (AUC) score on the unseen evaluation data
and obtains the second place in the Bird Audio Detection challenge.
|
Qualitative Assessment of Recurrent Human Motion | Smartphone applications designed to track human motion in combination with
wearable sensors, e.g., during physical exercising, raised huge attention
recently. Commonly, they provide quantitative services, such as personalized
training instructions or the counting of distances. But qualitative monitoring
and assessment is still missing, e.g., to detect malpositions, to prevent
injuries, or to optimize training success. We address this issue by presenting
a concept for qualitative as well as generic assessment of recurrent human
motion by processing multi-dimensional, continuous time series tracked with
motion sensors. Therefore, our segmentation procedure extracts individual
events of specific length and we propose expressive features to accomplish a
qualitative motion assessment by supervised classification. We verified our
approach within a comprehensive study encompassing 27 athletes undertaking
different body weight exercises. We are able to recognize six different
exercise types with a success rate of 100% and to assess them qualitatively
with an average success rate of 99.3%.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.