title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
On gradient regularizers for MMD GANs | We propose a principled method for gradient-based regularization of the
critic of GAN-like models trained by adversarially optimizing the kernel of a
Maximum Mean Discrepancy (MMD). We show that controlling the gradient of the
critic is vital to having a sensible loss function, and devise a method to
enforce exact, analytical gradient constraints at no additional cost compared
to existing approximate techniques based on additive regularizers. The new loss
function is provably continuous, and experiments show that it stabilizes and
accelerates training, giving image generation models that outperform
state-of-the art methods on $160 \times 160$ CelebA and $64 \times 64$
unconditional ImageNet.
|
Human-in-the-Loop Interpretability Prior | We often desire our models to be interpretable as well as accurate. Prior
work on optimizing models for interpretability has relied on easy-to-quantify
proxies for interpretability, such as sparsity or the number of operations
required. In this work, we optimize for interpretability by directly including
humans in the optimization loop. We develop an algorithm that minimizes the
number of user studies to find models that are both predictive and
interpretable and demonstrate our approach on several data sets. Our human
subjects results show trends towards different proxy notions of
interpretability on different datasets, which suggests that different proxies
are preferred on different tasks.
|
Adversarial Regularizers in Inverse Problems | Inverse Problems in medical imaging and computer vision are traditionally
solved using purely model-based methods. Among those variational regularization
models are one of the most popular approaches. We propose a new framework for
applying data-driven approaches to inverse problems, using a neural network as
a regularization functional. The network learns to discriminate between the
distribution of ground truth images and the distribution of unregularized
reconstructions. Once trained, the network is applied to the inverse problem by
solving the corresponding variational problem. Unlike other data-based
approaches for inverse problems, the algorithm can be applied even if only
unsupervised training data is available. Experiments demonstrate the potential
of the framework for denoising on the BSDS dataset and for computed tomography
reconstruction on the LIDC dataset.
|
Focal onset seizure prediction using convolutional networks | Objective: This work investigates the hypothesis that focal seizures can be
predicted using scalp electroencephalogram (EEG) data. Our first aim is to
learn features that distinguish between the interictal and preictal regions.
The second aim is to define a prediction horizon in which the prediction is as
accurate and as early as possible, clearly two competing objectives. Methods:
Convolutional filters on the wavelet transformation of the EEG signal are used
to define and learn quantitative signatures for each period: interictal,
preictal, and ictal. The optimal seizure prediction horizon is also learned
from the data as opposed to making an a priori assumption. Results:
Computational solutions to the optimization problem indicate a ten-minute
seizure prediction horizon. This result is verified by measuring
Kullback-Leibler divergence on the distributions of the automatically extracted
features. Conclusion: The results on the EEG database of 204 recordings
demonstrate that (i) the preictal phase transition occurs approximately ten
minutes before seizure onset, and (ii) the prediction results on the test set
are promising, with a sensitivity of 87.8% and a low false prediction rate of
0.142 FP/h. Our results significantly outperform a random predictor and other
seizure prediction algorithms. Significance: We demonstrate that a robust set
of features can be learned from scalp EEG that characterize the preictal state
of focal seizures.
|
Playing hard exploration games by watching YouTube | Deep reinforcement learning methods traditionally struggle with tasks where
environment rewards are particularly sparse. One successful method of guiding
exploration in these domains is to imitate trajectories provided by a human
demonstrator. However, these demonstrations are typically collected under
artificial conditions, i.e. with access to the agent's exact environment setup
and the demonstrator's action and reward trajectories. Here we propose a
two-stage method that overcomes these limitations by relying on noisy,
unaligned footage without access to such data. First, we learn to map unaligned
videos from multiple sources to a common representation using self-supervised
objectives constructed over both time and modality (i.e. vision and sound).
Second, we embed a single YouTube video in this representation to construct a
reward function that encourages an agent to imitate human gameplay. This method
of one-shot imitation allows our agent to convincingly exceed human-level
performance on the infamously hard exploration games Montezuma's Revenge,
Pitfall! and Private Eye for the first time, even if the agent is not presented
with any environment rewards.
|
Observe and Look Further: Achieving Consistent Performance on Atari | Despite significant advances in the field of deep Reinforcement Learning
(RL), today's algorithms still fail to learn human-level policies consistently
over a set of diverse tasks such as Atari 2600 games. We identify three key
challenges that any algorithm needs to master in order to perform well on all
games: processing diverse reward distributions, reasoning over long time
horizons, and exploring efficiently. In this paper, we propose an algorithm
that addresses each of these challenges and is able to learn human-level
policies on nearly all Atari games. A new transformed Bellman operator allows
our algorithm to process rewards of varying densities and scales; an auxiliary
temporal consistency loss allows us to train stably using a discount factor of
$\gamma = 0.999$ (instead of $\gamma = 0.99$) extending the effective planning
horizon by an order of magnitude; and we ease the exploration problem by using
human demonstrations that guide the agent towards rewarding states. When tested
on a set of 42 Atari games, our algorithm exceeds the performance of an average
human on 40 games using a common set of hyper parameters. Furthermore, it is
the first deep RL algorithm to solve the first level of Montezuma's Revenge.
|
Adversarial Noise Attacks of Deep Learning Architectures -- Stability
Analysis via Sparse Modeled Signals | Despite their impressive performance, deep convolutional neural networks
(CNNs) have been shown to be sensitive to small adversarial perturbations.
These nuisances, which one can barely notice, are powerful enough to fool
sophisticated and well performing classifiers, leading to ridiculous
misclassification results. In this paper we analyze the stability of
state-of-the-art deep-learning classification machines to adversarial
perturbations, where we assume that the signals belong to the (possibly
multi-layer) sparse representation model. We start with convolutional sparsity
and then proceed to its multi-layered version, which is tightly connected to
CNNs. Our analysis links between the stability of the classification to noise
and the underlying structure of the signal, quantified by the sparsity of its
representation under a fixed dictionary. In addition, we offer similar
stability theorems for two practical pursuit algorithms, which are posed as two
different deep-learning architectures - the layered Thresholding and the
layered Basis Pursuit. Our analysis establishes the better robustness of the
later to adversarial attacks. We corroborate these theoretical results by
numerical experiments on three datasets: MNIST, CIFAR-10 and CIFAR-100.
|
Deep Neural Networks for Swept Volume Prediction Between Configurations | Swept Volume (SV), the volume displaced by an object when it is moving along
a trajectory, is considered a useful metric for motion planning. First, SV has
been used to identify collisions along a trajectory, because it directly
measures the amount of space required for an object to move. Second, in
sampling-based motion planning, SV is an ideal distance metric, because it
correlates to the likelihood of success of the expensive local planning step
between two sampled configurations. However, in both of these applications,
traditional SV algorithms are too computationally expensive for efficient
motion planning. In this work, we train Deep Neural Networks (DNNs) to learn
the size of SV for specific robot geometries. Results for two robots, a 6
degree of freedom (DOF) rigid body and a 7 DOF fixed-based manipulator,
indicate that the network estimations are very close to the true size of SV and
is more than 1500 times faster than a state of the art SV estimation algorithm.
|
How Does Batch Normalization Help Optimization? | Batch Normalization (BatchNorm) is a widely adopted technique that enables
faster and more stable training of deep neural networks (DNNs). Despite its
pervasiveness, the exact reasons for BatchNorm's effectiveness are still poorly
understood. The popular belief is that this effectiveness stems from
controlling the change of the layers' input distributions during training to
reduce the so-called "internal covariate shift". In this work, we demonstrate
that such distributional stability of layer inputs has little to do with the
success of BatchNorm. Instead, we uncover a more fundamental impact of
BatchNorm on the training process: it makes the optimization landscape
significantly smoother. This smoothness induces a more predictive and stable
behavior of the gradients, allowing for faster training.
|
Deep Learning under Privileged Information Using Heteroscedastic Dropout | Unlike machines, humans learn through rapid, abstract model-building. The
role of a teacher is not simply to hammer home right or wrong answers, but
rather to provide intuitive comments, comparisons, and explanations to a pupil.
This is what the Learning Under Privileged Information (LUPI) paradigm
endeavors to model by utilizing extra knowledge only available during training.
We propose a new LUPI algorithm specifically designed for Convolutional Neural
Networks (CNNs) and Recurrent Neural Networks (RNNs). We propose to use a
heteroscedastic dropout (i.e. dropout with a varying variance) and make the
variance of the dropout a function of privileged information. Intuitively, this
corresponds to using the privileged information to control the uncertainty of
the model output. We perform experiments using CNNs and RNNs for the tasks of
image classification and machine translation. Our method significantly
increases the sample efficiency during learning, resulting in higher accuracy
with a large margin when the number of training examples is limited. We also
theoretically justify the gains in sample efficiency by providing a
generalization error bound decreasing with $O(\frac{1}{n})$, where $n$ is the
number of training examples, in an oracle case.
|
K-Beam Minimax: Efficient Optimization for Deep Adversarial Learning | Minimax optimization plays a key role in adversarial training of machine
learning algorithms, such as learning generative models, domain adaptation,
privacy preservation, and robust learning. In this paper, we demonstrate the
failure of alternating gradient descent in minimax optimization problems due to
the discontinuity of solutions of the inner maximization. To address this, we
propose a new epsilon-subgradient descent algorithm that addresses this problem
by simultaneously tracking K candidate solutions. Practically, the algorithm
can find solutions that previous saddle-point algorithms cannot find, with only
a sublinear increase of complexity in K. We analyze the conditions under which
the algorithm converges to the true solution in detail. A significant
improvement in stability and convergence speed of the algorithm is observed in
simple representative problems, GAN training, and domain-adaptation problems.
|
High Dimensional Robust Sparse Regression | We provide a novel -- and to the best of our knowledge, the first --
algorithm for high dimensional sparse regression with constant fraction of
corruptions in explanatory and/or response variables. Our algorithm recovers
the true sparse parameters with sub-linear sample complexity, in the presence
of a constant fraction of arbitrary corruptions. Our main contribution is a
robust variant of Iterative Hard Thresholding. Using this, we provide accurate
estimators: when the covariance matrix in sparse regression is identity, our
error guarantee is near information-theoretically optimal. We then deal with
robust sparse regression with unknown structured covariance matrix. We propose
a filtering algorithm which consists of a novel randomized outlier removal
technique for robust sparse mean estimation that may be of interest in its own
right: the filtering algorithm is flexible enough to deal with unknown
covariance. Also, it is orderwise more efficient computationally than the
ellipsoid algorithm. Using sub-linear sample complexity, our algorithm achieves
the best known (and first) error guarantee. We demonstrate the effectiveness on
large-scale sparse regression problems with arbitrary corruptions.
|
A Unified Particle-Optimization Framework for Scalable Bayesian Sampling | There has been recent interest in developing scalable Bayesian sampling
methods such as stochastic gradient MCMC (SG-MCMC) and Stein variational
gradient descent (SVGD) for big-data analysis. A standard SG-MCMC algorithm
simulates samples from a discrete-time Markov chain to approximate a target
distribution, thus samples could be highly correlated, an undesired property
for SG-MCMC. In contrary, SVGD directly optimizes a set of particles to
approximate a target distribution, and thus is able to obtain good
approximations with relatively much fewer samples. In this paper, we propose a
principle particle-optimization framework based on Wasserstein gradient flows
to unify SG-MCMC and SVGD, and to allow new algorithms to be developed. Our
framework interprets SG-MCMC as particle optimization on the space of
probability measures, revealing a strong connection between SG-MCMC and SVGD.
The key component of our framework is several particle-approximate techniques
to efficiently solve the original partial differential equations on the space
of probability measures. Extensive experiments on both synthetic data and deep
neural networks demonstrate the effectiveness and efficiency of our framework
for scalable Bayesian sampling.
|
Variational Inverse Control with Events: A General Framework for
Data-Driven Reward Definition | The design of a reward function often poses a major practical challenge to
real-world applications of reinforcement learning. Approaches such as inverse
reinforcement learning attempt to overcome this challenge, but require expert
demonstrations, which can be difficult or expensive to obtain in practice. We
propose variational inverse control with events (VICE), which generalizes
inverse reinforcement learning methods to cases where full demonstrations are
not needed, such as when only samples of desired goal states are available. Our
method is grounded in an alternative perspective on control and reinforcement
learning, where an agent's goal is to maximize the probability that one or more
events will happen at some point in the future, rather than maximizing
cumulative rewards. We demonstrate the effectiveness of our methods on
continuous control tasks, with a focus on high-dimensional observations like
images where rewards are hard or even impossible to specify.
|
Biologically Motivated Algorithms for Propagating Local Target
Representations | Finding biologically plausible alternatives to back-propagation of errors is
a fundamentally important challenge in artificial neural network research. In
this paper, we propose a learning algorithm called error-driven Local
Representation Alignment (LRA-E), which has strong connections to predictive
coding, a theory that offers a mechanistic way of describing neurocomputational
machinery. In addition, we propose an improved variant of Difference Target
Propagation, another procedure that comes from the same family of algorithms as
LRA-E. We compare our procedures to several other biologically-motivated
algorithms, including two feedback alignment algorithms and Equilibrium
Propagation. In two benchmarks, we find that both of our proposed algorithms
yield stable performance and strong generalization compared to other competing
back-propagation alternatives when training deeper, highly nonlinear networks,
with LRA-E performing the best overall.
|
Supervised Policy Update for Deep Reinforcement Learning | We propose a new sample-efficient methodology, called Supervised Policy
Update (SPU), for deep reinforcement learning. Starting with data generated by
the current policy, SPU formulates and solves a constrained optimization
problem in the non-parameterized proximal policy space. Using supervised
regression, it then converts the optimal non-parameterized policy to a
parameterized policy, from which it draws new samples. The methodology is
general in that it applies to both discrete and continuous action spaces, and
can handle a wide variety of proximity constraints for the non-parameterized
optimization problem. We show how the Natural Policy Gradient and Trust Region
Policy Optimization (NPG/TRPO) problems, and the Proximal Policy Optimization
(PPO) problem can be addressed by this methodology. The SPU implementation is
much simpler than TRPO. In terms of sample efficiency, our extensive
experiments show SPU outperforms TRPO in Mujoco simulated robotic tasks and
outperforms PPO in Atari video game tasks.
|
Active and Adaptive Sequential learning | A framework is introduced for actively and adaptively solving a sequence of
machine learning problems, which are changing in bounded manner from one time
step to the next. An algorithm is developed that actively queries the labels of
the most informative samples from an unlabeled data pool, and that adapts to
the change by utilizing the information acquired in the previous steps. Our
analysis shows that the proposed active learning algorithm based on stochastic
gradient descent achieves a near-optimal excess risk performance for maximum
likelihood estimation. Furthermore, an estimator of the change in the learning
problems using the active learning samples is constructed, which provides an
adaptive sample size selection rule that guarantees the excess risk is bounded
for sufficiently large number of time steps. Experiments with synthetic and
real data are presented to validate our algorithm and theoretical results.
|
Depth and nonlinearity induce implicit exploration for RL | The question of how to explore, i.e., take actions with uncertain outcomes to
learn about possible future rewards, is a key question in reinforcement
learning (RL). Here, we show a surprising result: We show that Q-learning with
nonlinear Q-function and no explicit exploration (i.e., a purely greedy policy)
can learn several standard benchmark tasks, including mountain car, equally
well as, or better than, the most commonly-used $\epsilon$-greedy exploration.
We carefully examine this result and show that both the depth of the Q-network
and the type of nonlinearity are important to induce such deterministic
exploration.
|
A Novel Multi-clustering Method for Hierarchical Clusterings, Based on
Boosting | Bagging and boosting are proved to be the best methods of building multiple
classifiers in classification combination problems. In the area of "flat
clustering" problems, it is also recognized that multi-clustering methods based
on boosting provide clusterings of an improved quality. In this paper, we
introduce a novel multi-clustering method for "hierarchical clusterings" based
on boosting theory, which creates a more stable hierarchical clustering of a
dataset. The proposed algorithm includes a boosting iteration in which a
bootstrap of samples is created by weighted random sampling of elements from
the original dataset. A hierarchical clustering algorithm is then applied to
selected subsample to build a dendrogram which describes the hierarchy.
Finally, dissimilarity description matrices of multiple dendrogram results are
combined to a consensus one, using a hierarchical-clustering-combination
approach. Experiments on real popular datasets show that boosted method
provides superior quality solutions compared to standard hierarchical
clustering methods.
|
Random mesh projectors for inverse problems | We propose a new learning-based approach to solve ill-posed inverse problems
in imaging. We address the case where ground truth training samples are rare
and the problem is severely ill-posed - both because of the underlying physics
and because we can only get few measurements. This setting is common in
geophysical imaging and remote sensing. We show that in this case the common
approach to directly learn the mapping from the measured data to the
reconstruction becomes unstable. Instead, we propose to first learn an ensemble
of simpler mappings from the data to projections of the unknown image into
random piecewise-constant subspaces. We then combine the projections to form a
final reconstruction by solving a deconvolution-like problem. We show
experimentally that the proposed method is more robust to measurement noise and
corruptions not seen during training than a directly learned inverse.
|
Learn to Combine Modalities in Multimodal Deep Learning | Combining complementary information from multiple modalities is intuitively
appealing for improving the performance of learning-based approaches. However,
it is challenging to fully leverage different modalities due to practical
challenges such as varying levels of noise and conflicts between modalities.
Existing methods do not adopt a joint approach to capturing synergies between
the modalities while simultaneously filtering noise and resolving conflicts on
a per sample basis. In this work we propose a novel deep neural network based
technique that multiplicatively combines information from different source
modalities. Thus the model training process automatically focuses on
information from more reliable modalities while reducing emphasis on the less
reliable modalities. Furthermore, we propose an extension that multiplicatively
combines not only the single-source modalities, but a set of mixtured source
modalities to better capture cross-modal signal correlations. We demonstrate
the effectiveness of our proposed technique by presenting empirical results on
three multimodal classification tasks from different domains. The results show
consistent accuracy improvements on all three tasks.
|
Multi-turn Dialogue Response Generation in an Adversarial Learning
Framework | We propose an adversarial learning approach for generating multi-turn
dialogue responses. Our proposed framework, hredGAN, is based on conditional
generative adversarial networks (GANs). The GAN's generator is a modified
hierarchical recurrent encoder-decoder network (HRED) and the discriminator is
a word-level bidirectional RNN that shares context and word embeddings with the
generator. During inference, noise samples conditioned on the dialogue history
are used to perturb the generator's latent space to generate several possible
responses. The final response is the one ranked best by the discriminator. The
hredGAN shows improved performance over existing methods: (1) it generalizes
better than networks trained using only the log-likelihood criterion, and (2)
it generates longer, more informative and more diverse responses with high
utterance and topic relevance even with limited training data. This improvement
is demonstrated on the Movie triples and Ubuntu dialogue datasets using both
automatic and human evaluations.
|
Optimal Testing in the Experiment-rich Regime | Motivated by the widespread adoption of large-scale A/B testing in industry,
we propose a new experimentation framework for the setting where potential
experiments are abundant (i.e., many hypotheses are available to test), and
observations are costly; we refer to this as the experiment-rich regime. Such
scenarios require the experimenter to internalize the opportunity cost of
assigning a sample to a particular experiment. We fully characterize the
optimal policy and give an algorithm to compute it. Furthermore, we develop a
simple heuristic that also provides intuition for the optimal policy. We use
simulations based on real data to compare both the optimal algorithm and the
heuristic to other natural alternative experimental design frameworks. In
particular, we discuss the paradox of power: high-powered classical tests can
lead to highly inefficient sampling in the experiment-rich regime.
|
Collaborative Learning for Deep Neural Networks | We introduce collaborative learning in which multiple classifier heads of the
same network are simultaneously trained on the same training data to improve
generalization and robustness to label noise with no extra inference cost. It
acquires the strengths from auxiliary training, multi-task learning and
knowledge distillation. There are two important mechanisms involved in
collaborative learning. First, the consensus of multiple views from different
classifier heads on the same example provides supplementary information as well
as regularization to each classifier, thereby improving generalization. Second,
intermediate-level representation (ILR) sharing with backpropagation rescaling
aggregates the gradient flows from all heads, which not only reduces training
computational complexity, but also facilitates supervision to the shared
layers. The empirical results on CIFAR and ImageNet datasets demonstrate that
deep neural networks learned as a group in a collaborative way significantly
reduce the generalization error and increase the robustness to label noise.
|
Fast Incremental von Neumann Graph Entropy Computation: Theory,
Algorithm, and Applications | The von Neumann graph entropy (VNGE) facilitates measurement of information
divergence and distance between graphs in a graph sequence. It has been
successfully applied to various learning tasks driven by network-based data.
While effective, VNGE is computationally demanding as it requires the full
eigenspectrum of the graph Laplacian matrix. In this paper, we propose a new
computational framework, Fast Incremental von Neumann Graph EntRopy (FINGER),
which approaches VNGE with a performance guarantee. FINGER reduces the cubic
complexity of VNGE to linear complexity in the number of nodes and edges, and
thus enables online computation based on incremental graph changes. We also
show asymptotic equivalence of FINGER to the exact VNGE, and derive its
approximation error bounds. Based on FINGER, we propose efficient algorithms
for computing Jensen-Shannon distance between graphs. Our experimental results
on different random graph models demonstrate the computational efficiency and
the asymptotic equivalence of FINGER. In addition, we apply FINGER to two
real-world applications and one synthesized anomaly detection dataset, and
corroborate its superior performance over seven baseline graph similarity
methods.
|
To Trust Or Not To Trust A Classifier | Knowing when a classifier's prediction can be trusted is useful in many
applications and critical for safely using AI. While the bulk of the effort in
machine learning research has been towards improving classifier performance,
understanding when a classifier's predictions should and should not be trusted
has received far less attention. The standard approach is to use the
classifier's discriminant or confidence score; however, we show there exists an
alternative that is more effective in many situations. We propose a new score,
called the trust score, which measures the agreement between the classifier and
a modified nearest-neighbor classifier on the testing example. We show
empirically that high (low) trust scores produce surprisingly high precision at
identifying correctly (incorrectly) classified examples, consistently
outperforming the classifier's confidence score as well as many other
baselines. Further, under some mild distributional assumptions, we show that if
the trust score for an example is high (low), the classifier will likely agree
(disagree) with the Bayes-optimal classifier. Our guarantees consist of
non-asymptotic rates of statistical consistency under various nonparametric
settings and build on recent developments in topological data analysis.
|
Tight Regret Bounds for Bayesian Optimization in One Dimension | We consider the problem of Bayesian optimization (BO) in one dimension, under
a Gaussian process prior and Gaussian sampling noise. We provide a theoretical
analysis showing that, under fairly mild technical assumptions on the kernel,
the best possible cumulative regret up to time $T$ behaves as
$\Omega(\sqrt{T})$ and $O(\sqrt{T\log T})$. This gives a tight characterization
up to a $\sqrt{\log T}$ factor, and includes the first non-trivial lower bound
for noisy BO. Our assumptions are satisfied, for example, by the squared
exponential and Mat\'ern-$\nu$ kernels, with the latter requiring $\nu > 2$.
Our results certify the near-optimality of existing bounds (Srinivas {\em et
al.}, 2009) for the SE kernel, while proving them to be strictly suboptimal for
the Mat\'ern kernel with $\nu > 2$.
|
Infinite Arms Bandit: Optimality via Confidence Bounds | Berry et al. (1997) initiated the development of the infinite arms bandit
problem. They derived a regret lower bound of all allocation strategies for
Bernoulli rewards with uniform priors, and proposed strategies based on success
runs. Bonald and Prouti\`{e}re (2013) proposed a two-target algorithm that
achieves the regret lower bound, and extended optimality to Bernoulli rewards
with general priors. We present here a confidence bound target (CBT) algorithm
that achieves optimality for rewards that are bounded above. For each arm we
construct a confidence bound and compare it against each other and a target
value to determine if the arm should be sampled further. The target value
depends on the assumed priors of the arm means. In the absence of information
on the prior, the target value is determined empirically. Numerical studies
here show that CBT is versatile and outperforms its competitors.
|
Grow and Prune Compact, Fast, and Accurate LSTMs | Long short-term memory (LSTM) has been widely used for sequential data
modeling. Researchers have increased LSTM depth by stacking LSTM cells to
improve performance. This incurs model redundancy, increases run-time delay,
and makes the LSTMs more prone to overfitting. To address these problems, we
propose a hidden-layer LSTM (H-LSTM) that adds hidden layers to LSTM's original
one level non-linear control gates. H-LSTM increases accuracy while employing
fewer external stacked layers, thus reducing the number of parameters and
run-time latency significantly. We employ grow-and-prune (GP) training to
iteratively adjust the hidden layers through gradient-based growth and
magnitude-based pruning of connections. This learns both the weights and the
compact architecture of H-LSTM control gates. We have GP-trained H-LSTMs for
image captioning and speech recognition applications. For the NeuralTalk
architecture on the MSCOCO dataset, our three models reduce the number of
parameters by 38.7x [floating-point operations (FLOPs) by 45.5x], run-time
latency by 4.5x, and improve the CIDEr score by 2.6. For the DeepSpeech2
architecture on the AN4 dataset, our two models reduce the number of parameters
by 19.4x (FLOPs by 23.5x), run-time latency by 15.7%, and the word error rate
from 12.9% to 8.7%. Thus, GP-trained H-LSTMs can be seen to be compact, fast,
and accurate.
|
Automated proof synthesis for propositional logic with deep neural
networks | This work explores the application of deep learning, a machine learning
technique that uses deep neural networks (DNN) in its core, to an automated
theorem proving (ATP) problem. To this end, we construct a statistical model
which quantifies the likelihood that a proof is indeed a correct one of a given
proposition. Based on this model, we give a proof-synthesis procedure that
searches for a proof in the order of the likelihood. This procedure uses an
estimator of the likelihood of an inference rule being applied at each step of
a proof. As an implementation of the estimator, we propose a
proposition-to-proof architecture, which is a DNN tailored to the automated
proof synthesis problem. To empirically demonstrate its usefulness, we apply
our model to synthesize proofs of propositional logic. We train the
proposition-to-proof model using a training dataset of proposition-proof pairs.
The evaluation against a benchmark set shows the very high accuracy and an
improvement to the recent work of neural proof synthesis.
|
Stochastic Zeroth-order Optimization via Variance Reduction method | Derivative-free optimization has become an important technique used in
machine learning for optimizing black-box models. To conduct updates without
explicitly computing gradient, most current approaches iteratively sample a
random search direction from Gaussian distribution and compute the estimated
gradient along that direction. However, due to the variance in the search
direction, the convergence rates and query complexities of existing methods
suffer from a factor of $d$, where $d$ is the problem dimension. In this paper,
we introduce a novel Stochastic Zeroth-order method with Variance Reduction
under Gaussian smoothing (SZVR-G) and establish the complexity for optimizing
non-convex problems. With variance reduction on both sample space and search
space, the complexity of our algorithm is sublinear to $d$ and is strictly
better than current approaches, in both smooth and non-smooth cases. Moreover,
we extend the proposed method to the mini-batch version. Our experimental
results demonstrate the superior performance of the proposed method over
existing derivative-free optimization techniques. Furthermore, we successfully
apply our method to conduct a universal black-box attack to deep neural
networks and present some interesting results.
|
Learning multiple non-mutually-exclusive tasks for improved
classification of inherently ordered labels | Medical image classification involves thresholding of labels that represent
malignancy risk levels. Usually, a task defines a single threshold, and when
developing computer-aided diagnosis tools, a single network is trained per such
threshold, e.g. as screening out healthy (very low risk) patients to leave
possibly sick ones for further analysis (low threshold), or trying to find
malignant cases among those marked as non-risk by the radiologist ("second
reading", high threshold). We propose a way to rephrase the classification
problem in a manner that yields several problems (corresponding to different
thresholds) to be solved simultaneously. This allows the use of Multiple Task
Learning (MTL) methods, significantly improving the performance of the original
classifier, by facilitating effective extraction of information from existing
data.
|
An Information-Theoretic Analysis for Thompson Sampling with Many
Actions | Information-theoretic Bayesian regret bounds of Russo and Van Roy capture the
dependence of regret on prior uncertainty. However, this dependence is through
entropy, which can become arbitrarily large as the number of actions increases.
We establish new bounds that depend instead on a notion of rate-distortion.
Among other things, this allows us to recover through information-theoretic
arguments a near-optimal bound for the linear bandit. We also offer a bound for
the logistic bandit that dramatically improves on the best previously
available, though this bound depends on an information-theoretic statistic that
we have only been able to quantify via computation.
|
ADAGIO: Interactive Experimentation with Adversarial Attack and Defense
for Audio | Adversarial machine learning research has recently demonstrated the
feasibility to confuse automatic speech recognition (ASR) models by introducing
acoustically imperceptible perturbations to audio samples. To help researchers
and practitioners gain better understanding of the impact of such attacks, and
to provide them with tools to help them more easily evaluate and craft strong
defenses for their models, we present ADAGIO, the first tool designed to allow
interactive experimentation with adversarial attacks and defenses on an ASR
model in real time, both visually and aurally. ADAGIO incorporates AMR and MP3
audio compression techniques as defenses, which users can interactively apply
to attacked audio samples. We show that these techniques, which are based on
psychoacoustic principles, effectively eliminate targeted attacks, reducing the
attack success rate from 92.5% to 0%. We will demonstrate ADAGIO and invite the
audience to try it on the Mozilla Common Voice dataset.
|
Foresee: Attentive Future Projections of Chaotic Road Environments with
Online Training | In this paper, we train a recurrent neural network to learn dynamics of a
chaotic road environment and to project the future of the environment on an
image. Future projection can be used to anticipate an unseen environment for
example, in autonomous driving. Road environment is highly dynamic and complex
due to the interaction among traffic participants such as vehicles and
pedestrians. Even in this complex environment, a human driver is efficacious to
safely drive on chaotic roads irrespective of the number of traffic
participants. The proliferation of deep learning research has shown the
efficacy of neural networks in learning this human behavior. In the same
direction, we investigate recurrent neural networks to understand the chaotic
road environment which is shared by pedestrians, vehicles (cars, trucks,
bicycles etc.), and sometimes animals as well. We propose \emph{Foresee}, a
unidirectional gated recurrent units (GRUs) network with attention to project
future of the environment in the form of images. We have collected several
videos on Delhi roads consisting of various traffic participants, background
and infrastructure differences (like 3D pedestrian crossing) at various times
on various days. We train \emph{Foresee} in an unsupervised way and we use
online training to project frames up to $0.5$ seconds in advance. We show that
our proposed model performs better than state of the art methods (prednet and
Enc. Dec. LSTM) and finally, we show that our trained model generalizes to a
public dataset for future projections.
|
Predictive Performance Modeling for Distributed Computing using
Black-Box Monitoring and Machine Learning | In many domains, the previous decade was characterized by increasing data
volumes and growing complexity of computational workloads, creating new demands
for highly data-parallel computing in distributed systems. Effective operation
of these systems is challenging when facing uncertainties about the performance
of jobs and tasks under varying resource configurations, e.g., for scheduling
and resource allocation. We survey predictive performance modeling (PPM)
approaches to estimate performance metrics such as execution duration, required
memory or wait times of future jobs and tasks based on past performance
observations. We focus on non-intrusive methods, i.e., methods that can be
applied to any workload without modification, since the workload is usually a
black-box from the perspective of the systems managing the computational
infrastructure. We classify and compare sources of performance variation,
predicted performance metrics, required training data, use cases, and the
underlying prediction techniques. We conclude by identifying several open
problems and pressing research needs in the field.
|
Differential Properties of Sinkhorn Approximation for Learning with
Wasserstein Distance | Applications of optimal transport have recently gained remarkable attention
thanks to the computational advantages of entropic regularization. However, in
most situations the Sinkhorn approximation of the Wasserstein distance is
replaced by a regularized version that is less accurate but easy to
differentiate. In this work we characterize the differential properties of the
original Sinkhorn distance, proving that it enjoys the same smoothness as its
regularized version and we explicitly provide an efficient algorithm to compute
its gradient. We show that this result benefits both theory and applications:
on one hand, high order smoothness confers statistical guarantees to learning
with Wasserstein approximations. On the other hand, the gradient formula allows
us to efficiently solve learning and optimization problems in practice.
Promising preliminary experiments complement our analysis.
|
Propagating Confidences through CNNs for Sparse Data Regression | In most computer vision applications, convolutional neural networks (CNNs)
operate on dense image data generated by ordinary cameras. Designing CNNs for
sparse and irregularly spaced input data is still an open problem with numerous
applications in autonomous driving, robotics, and surveillance. To tackle this
challenging problem, we introduce an algebraically-constrained convolution
layer for CNNs with sparse input and demonstrate its capabilities for the scene
depth completion task. We propose novel strategies for determining the
confidence from the convolution operation and propagating it to consecutive
layers. Furthermore, we propose an objective function that simultaneously
minimizes the data error while maximizing the output confidence. Comprehensive
experiments are performed on the KITTI depth benchmark and the results clearly
demonstrate that the proposed approach achieves superior performance while
requiring three times fewer parameters than the state-of-the-art methods.
Moreover, our approach produces a continuous pixel-wise confidence map enabling
information fusion, state inference, and decision support.
|
On the Spectrum of Random Features Maps of High Dimensional Data | Random feature maps are ubiquitous in modern statistical machine learning,
where they generalize random projections by means of powerful, yet often
difficult to analyze nonlinear operators. In this paper, we leverage the
"concentration" phenomenon induced by random matrix theory to perform a
spectral analysis on the Gram matrix of these random feature maps, here for
Gaussian mixture models of simultaneously large dimension and size. Our results
are instrumental to a deeper understanding on the interplay of the nonlinearity
and the statistics of the data, thereby allowing for a better tuning of random
feature-based techniques.
|
The Dynamics of Learning: A Random Matrix Approach | Understanding the learning dynamics of neural networks is one of the key
issues for the improvement of optimization algorithms as well as for the
theoretical comprehension of why deep neural nets work so well today. In this
paper, we introduce a random matrix-based framework to analyze the learning
dynamics of a single-layer linear network on a binary classification problem,
for data of simultaneously large dimension and size, trained by gradient
descent. Our results provide rich insights into common questions in neural
nets, such as overfitting, early stopping and the initialization of training,
thereby opening the door for future studies of more elaborate structures and
models appearing in today's neural networks.
|
Anonymous Walk Embeddings | The task of representing entire graphs has seen a surge of prominent results,
mainly due to learning convolutional neural networks (CNNs) on graph-structured
data. While CNNs demonstrate state-of-the-art performance in graph
classification task, such methods are supervised and therefore steer away from
the original problem of network representation in task-agnostic manner. Here,
we coherently propose an approach for embedding entire graphs and show that our
feature representations with SVM classifier increase classification accuracy of
CNN algorithms and traditional graph kernels. For this we describe a recently
discovered graph object, anonymous walk, on which we design task-independent
algorithms for learning graph representations in explicit and distributed way.
Overall, our work represents a new scalable unsupervised learning of
state-of-the-art representations of entire graphs.
|
Fast L1-Minimization Algorithm for Sparse Approximation Based on an
Improved LPNN-LCA framework | The aim of sparse approximation is to estimate a sparse signal according to
the measurement matrix and an observation vector. It is widely used in data
analytics, image processing, and communication, etc. Up to now, a lot of
research has been done in this area, and many off-the-shelf algorithms have
been proposed. However, most of them cannot offer a real-time solution. To some
extent, this shortcoming limits its application prospects. To address this
issue, we devise a novel sparse approximation algorithm based on Lagrange
programming neural network (LPNN), locally competitive algorithm (LCA), and
projection theorem. LPNN and LCA are both analog neural network which can help
us get a real-time solution. The non-differentiable objective function can be
solved by the concept of LCA. Utilizing the projection theorem, we further
modify the dynamics and proposed a new system with global asymptotic stability.
Simulation results show that the proposed sparse approximation method has the
real-time solutions with satisfactory MSEs.
|
Long Short-Term Memory Networks for CSI300 Volatility Prediction with
Baidu Search Volume | Intense volatility in financial markets affect humans worldwide. Therefore,
relatively accurate prediction of volatility is critical. We suggest that
massive data sources resulting from human interaction with the Internet may
offer a new perspective on the behavior of market participants in periods of
large market movements. First we select 28 key words, which are related to
finance as indicators of the public mood and macroeconomic factors. Then those
28 words of the daily search volume based on Baidu index are collected
manually, from June 1, 2006 to October 29, 2017. We apply a Long Short-Term
Memory neural network to forecast CSI300 volatility using those search volume
data. Compared to the benchmark GARCH model, our forecast is more accurate,
which demonstrates the effectiveness of the LSTM neural network in volatility
forecasting.
|
Short-term Load Forecasting with Deep Residual Networks | We present in this paper a model for forecasting short-term power loads based
on deep residual networks. The proposed model is able to integrate domain
knowledge and researchers' understanding of the task by virtue of different
neural network building blocks. Specifically, a modified deep residual network
is formulated to improve the forecast results. Further, a two-stage ensemble
strategy is used to enhance the generalization capability of the proposed
model. We also apply the proposed model to probabilistic load forecasting using
Monte Carlo dropout. Three public datasets are used to prove the effectiveness
of the proposed model. Multiple test cases and comparison with existing models
show that the proposed model is able to provide accurate load forecasting
results and has high generalization capability.
|
Algebraic Expression of Subjective Spatial and Temporal Patterns | Universal learning machine is a theory trying to study machine learning from
mathematical point of view. The outside world is reflected inside an universal
learning machine according to pattern of incoming data. This is subjective
pattern of learning machine. In [2,4], we discussed subjective spatial pattern,
and established a powerful tool -- X-form, which is an algebraic expression for
subjective spatial pattern. However, as the initial stage of study, there we
only discussed spatial pattern. Here, we will discuss spatial and temporal
patterns, and algebraic expression for them.
|
Automatic Large-Scale Data Acquisition via Crowdsourcing for Crosswalk
Classification: A Deep Learning Approach | Correctly identifying crosswalks is an essential task for the driving
activity and mobility autonomy. Many crosswalk classification, detection and
localization systems have been proposed in the literature over the years. These
systems use different perspectives to tackle the crosswalk classification
problem: satellite imagery, cockpit view (from the top of a car or behind the
windshield), and pedestrian perspective. Most of the works in the literature
are designed and evaluated using small and local datasets, i.e. datasets that
present low diversity. Scaling to large datasets imposes a challenge for the
annotation procedure. Moreover, there is still need for cross-database
experiments in the literature because it is usually hard to collect the data in
the same place and conditions of the final application. In this paper, we
present a crosswalk classification system based on deep learning. For that,
crowdsourcing platforms, such as OpenStreetMap and Google Street View, are
exploited to enable automatic training via automatic acquisition and annotation
of a large-scale database. Additionally, this work proposes a comparison study
of models trained using fully-automatic data acquisition and annotation against
models that were partially annotated. Cross-database experiments were also
included in the experimentation to show that the proposed methods enable use
with real world applications. Our results show that the model trained on the
fully-automatic database achieved high overall accuracy (94.12%), and that a
statistically significant improvement (to 96.30%) can be achieved by manually
annotating a specific part of the database. Finally, the results of the
cross-database experiments show that both models are robust to the many
variations of image and scenarios, presenting a consistent behavior.
|
MolGAN: An implicit generative model for small molecular graphs | Deep generative models for graph-structured data offer a new angle on the
problem of chemical synthesis: by optimizing differentiable models that
directly generate molecular graphs, it is possible to side-step expensive
search procedures in the discrete and vast space of chemical structures. We
introduce MolGAN, an implicit, likelihood-free generative model for small
molecular graphs that circumvents the need for expensive graph matching
procedures or node ordering heuristics of previous likelihood-based methods.
Our method adapts generative adversarial networks (GANs) to operate directly on
graph-structured data. We combine our approach with a reinforcement learning
objective to encourage the generation of molecules with specific desired
chemical properties. In experiments on the QM9 chemical database, we
demonstrate that our model is capable of generating close to 100% valid
compounds. MolGAN compares favorably both to recent proposals that use
string-based (SMILES) representations of molecules and to a likelihood-based
method that directly generates graphs, albeit being susceptible to mode
collapse.
|
l0-norm Based Centers Selection for Training Fault Tolerant RBF Networks
and Selecting Centers | The aim of this paper is to train an RBF neural network and select centers
under concurrent faults. It is well known that fault tolerance is a very
attractive property for neural networks. And center selection is an important
procedure during the training process of an RBF neural network. In this paper,
we devise two novel algorithms to address these two issues simultaneously. Both
of them are based on the ADMM framework. In the first method, the minimax
concave penalty (MCP) function is introduced to select centers. In the second
method, an l0-norm term is directly used, and the hard threshold (HT) is
utilized to address the l0-norm term. Under several mild conditions, we can
prove that both methods can globally converge to a unique limit point.
Simulation results show that, under concurrent fault, the proposed algorithms
are superior to many existing methods.
|
Reference-free Calibration in Sensor Networks | Sensor calibration is one of the fundamental challenges in large-scale IoT
networks. In this article, we address the challenge of reference-free
calibration of a densely deployed sensor network. Conventionally, to calibrate
an in-place sensor network (or sensor array), a reference is arbitrarily chosen
with or without prior information on sensor performance. However, an arbitrary
selection of a reference could prove fatal, if an erroneous sensor is
inadvertently chosen. To avert single point of dependence, and to improve
estimator performance, we propose unbiased reference-free algorithms. Although,
our focus is on reference-free solutions, the proposed framework, allows the
incorporation of additional references, if available. We show with the help of
simulations that the proposed solutions achieve the derived statistical lower
bounds asymptotically. In addition, the proposed algorithms show improvements
on real-life datasets, as compared to prevalent algorithms.
|
Why Is My Classifier Discriminatory? | Recent attempts to achieve fairness in predictive models focus on the balance
between fairness and accuracy. In sensitive applications such as healthcare or
criminal justice, this trade-off is often undesirable as any increase in
prediction error could have devastating consequences. In this work, we argue
that the fairness of predictions should be evaluated in context of the data,
and that unfairness induced by inadequate samples sizes or unmeasured
predictive variables should be addressed through data collection, rather than
by constraining the model. We decompose cost-based metrics of discrimination
into bias, variance, and noise, and propose actions aimed at estimating and
reducing each term. Finally, we perform case-studies on prediction of income,
mortality, and review ratings, confirming the value of this analysis. We find
that data collection is often a means to reduce discrimination without
sacrificing accuracy.
|
Robustifying Models Against Adversarial Attacks by Langevin Dynamics | Adversarial attacks on deep learning models have compromised their
performance considerably. As remedies, a lot of defense methods were proposed,
which however, have been circumvented by newer attacking strategies. In the
midst of this ensuing arms race, the problem of robustness against adversarial
attacks still remains unsolved. This paper proposes a novel, simple yet
effective defense strategy where adversarial samples are relaxed onto the
underlying manifold of the (unknown) target class distribution. Specifically,
our algorithm drives off-manifold adversarial samples towards high density
regions of the data generating distribution of the target class by the
Metroplis-adjusted Langevin algorithm (MALA) with perceptual boundary taken
into account. Although the motivation is similar to projection methods, e.g.,
Defense-GAN, our algorithm, called MALA for DEfense (MALADE), is equipped with
significant dispersion - projection is distributed broadly, and therefore any
whitebox attack cannot accurately align the input so that the MALADE moves it
to a targeted untrained spot where the model predicts a wrong label. In our
experiments, MALADE exhibited state-of-the-art performance against various
elaborate attacking strategies.
|
Towards Adversarial Configurations for Software Product Lines | Ensuring that all supposedly valid configurations of a software product line
(SPL) lead to well-formed and acceptable products is challenging since it is
most of the time impractical to enumerate and test all individual products of
an SPL. Machine learning classifiers have been recently used to predict the
acceptability of products associated with unseen configurations. For some
configurations, a tiny change in their feature values can make them pass from
acceptable to non-acceptable regarding users' requirements and vice-versa. In
this paper, we introduce the idea of leveraging these specific configurations
and their positions in the feature space to improve the classifier and
therefore the engineering of an SPL. Starting from a variability model, we
propose to use Adversarial Machine Learning techniques to create new,
adversarial configurations out of already known configurations by modifying
their feature values. Using an industrial video generator we show how
adversarial configurations can improve not only the classifier, but also the
variability model, the variability implementation, and the testing oracle.
|
Privacy Aware Offloading of Deep Neural Networks | Deep neural networks require large amounts of resources which makes them hard
to use on resource constrained devices such as Internet-of-things devices.
Offloading the computations to the cloud can circumvent these constraints but
introduces a privacy risk since the operator of the cloud is not necessarily
trustworthy. We propose a technique that obfuscates the data before sending it
to the remote computation node. The obfuscated data is unintelligible for a
human eavesdropper but can still be classified with a high accuracy by a neural
network trained on unobfuscated images.
|
Predicting County Level Corn Yields Using Deep Long Short Term Memory
Models | Corn yield prediction is beneficial as it provides valuable information about
production and prices prior the harvest. Publicly available high-quality corn
yield prediction can help address emergent information asymmetry problems and
in doing so improve price efficiency in futures markets. This paper is the
first to employ Long Short-Term Memory (LSTM), a special form of Recurrent
Neural Network (RNN) method to predict corn yields. A cross sectional time
series of county-level corn yield and hourly weather data made the sample space
large enough to use deep learning technics. LSTM is efficient in time series
prediction with complex inner relations, which makes it suitable for this task.
The empirical results from county level data in Iowa show promising predictive
power relative to existing survey based methods.
|
Sobolev Descent | We study a simplification of GAN training: the problem of transporting
particles from a source to a target distribution. Starting from the Sobolev GAN
critic, part of the gradient regularized GAN family, we show a strong relation
with Optimal Transport (OT). Specifically with the less popular dynamic
formulation of OT that finds a path of distributions from source to target
minimizing a ``kinetic energy''. We introduce Sobolev descent that constructs
similar paths by following gradient flows of a critic function in a kernel
space or parametrized by a neural network. In the kernel version, we show
convergence to the target distribution in the MMD sense. We show in theory and
experiments that regularization has an important role in favoring smooth
transitions between distributions, avoiding large gradients from the critic.
This analysis in a simplified particle setting provides insight in paths to
equilibrium in GANs.
|
Towards Understanding the Role of Over-Parametrization in Generalization
of Neural Networks | Despite existing work on ensuring generalization of neural networks in terms
of scale sensitive complexity measures, such as norms, margin and sharpness,
these complexity measures do not offer an explanation of why neural networks
generalize better with over-parametrization. In this work we suggest a novel
complexity measure based on unit-wise capacities resulting in a tighter
generalization bound for two layer ReLU networks. Our capacity bound correlates
with the behavior of test error with increasing network sizes, and could
potentially explain the improvement in generalization with
over-parametrization. We further present a matching lower bound for the
Rademacher complexity that improves over previous capacity lower bounds for
neural networks.
|
MPDCompress - Matrix Permutation Decomposition Algorithm for Deep Neural
Network Compression | Deep neural networks (DNNs) have become the state-of-the-art technique for
machine learning tasks in various applications. However, due to their size and
the computational complexity, large DNNs are not readily deployable on edge
devices in real-time. To manage complexity and accelerate computation, network
compression techniques based on pruning and quantization have been proposed and
shown to be effective in reducing network size. However, such network
compression can result in irregular matrix structures that are mismatched with
modern hardware-accelerated platforms, such as graphics processing units (GPUs)
designed to perform the DNN matrix multiplications in a structured
(block-based) way. We propose MPDCompress, a DNN compression algorithm based on
matrix permutation decomposition via random mask generation. In-training
application of the masks molds the synaptic weight connection matrix to a
sub-graph separation format. Aided by the random permutations, a
hardware-desirable block matrix is generated, allowing for a more efficient
implementation and compression of the network. To show versatility, we
empirically verify MPDCompress on several network models, compression rates,
and image datasets. On the LeNet 300-100 model (MNIST dataset), Deep MNIST, and
CIFAR10, we achieve 10 X network compression with less than 1% accuracy loss
compared to non-compressed accuracy performance. On AlexNet for the full
ImageNet ILSVRC-2012 dataset, we achieve 8 X network compression with less than
1% accuracy loss, with top-5 and top-1 accuracies of 79.6% and 56.4%,
respectively. Finally, we observe that the algorithm can offer inference
speedups across various hardware platforms, with 4 X faster operation achieved
on several mobile GPUs.
|
Dynamic Advisor-Based Ensemble (dynABE): Case study in stock trend
prediction of critical metal companies | Stock trend prediction is a challenging task due to the market's noise, and
machine learning techniques have recently been successful in coping with this
challenge. In this research, we create a novel framework for stock prediction,
Dynamic Advisor-Based Ensemble (dynABE). dynABE explores domain-specific areas
based on the companies of interest, diversifies the feature set by creating
different "advisors" that each handles a different area, follows an effective
model ensemble procedure for each advisor, and combines the advisors together
in a second-level ensemble through an online update strategy we developed.
dynABE is able to adapt to price pattern changes of the market during the
active trading period robustly, without needing to retrain the entire model. We
test dynABE on three cobalt-related companies, and it achieves the best-case
misclassification error of 31.12% and an annualized absolute return of 359.55%
with zero maximum drawdown. dynABE also consistently outperforms the baseline
models of support vector machine, neural network, and random forest in all case
studies.
|
Deep Reinforcement Learning in a Handful of Trials using Probabilistic
Dynamics Models | Model-based reinforcement learning (RL) algorithms can attain excellent
sample efficiency, but often lag behind the best model-free algorithms in terms
of asymptotic performance. This is especially true with high-capacity
parametric function approximators, such as deep networks. In this paper, we
study how to bridge this gap, by employing uncertainty-aware dynamics models.
We propose a new algorithm called probabilistic ensembles with trajectory
sampling (PETS) that combines uncertainty-aware deep network dynamics models
with sampling-based uncertainty propagation. Our comparison to state-of-the-art
model-based and model-free deep RL algorithms shows that our approach matches
the asymptotic performance of model-free algorithms on several challenging
benchmark tasks, while requiring significantly fewer samples (e.g., 8 and 125
times fewer samples than Soft Actor Critic and Proximal Policy Optimization
respectively on the half-cheetah task).
|
On Consensus-Optimality Trade-offs in Collaborative Deep Learning | In distributed machine learning, where agents collaboratively learn from
diverse private data sets, there is a fundamental tension between consensus and
optimality. In this paper, we build on recent algorithmic progresses in
distributed deep learning to explore various consensus-optimality trade-offs
over a fixed communication topology. First, we propose the incremental
consensus-based distributed SGD (i-CDSGD) algorithm, which involves multiple
consensus steps (where each agent communicates information with its neighbors)
within each SGD iteration. Second, we propose the generalized consensus-based
distributed SGD (g-CDSGD) algorithm that enables us to navigate the full
spectrum from complete consensus (all agents agree) to complete disagreement
(each agent converges to individual model parameters). We analytically
establish convergence of the proposed algorithms for strongly convex and
nonconvex objective functions; we also analyze the momentum variants of the
algorithms for the strongly convex case. We support our algorithms via
numerical experiments, and demonstrate significant improvements over existing
methods for collaborative deep learning.
|
Robustness May Be at Odds with Accuracy | We show that there may exist an inherent tension between the goal of
adversarial robustness and that of standard generalization. Specifically,
training robust models may not only be more resource-consuming, but also lead
to a reduction of standard accuracy. We demonstrate that this trade-off between
the standard accuracy of a model and its robustness to adversarial
perturbations provably exists in a fairly simple and natural setting. These
findings also corroborate a similar phenomenon observed empirically in more
complex settings. Further, we argue that this phenomenon is a consequence of
robust classifiers learning fundamentally different feature representations
than standard classifiers. These differences, in particular, seem to result in
unexpected benefits: the representations learned by robust models tend to align
better with salient data characteristics and human perception.
|
What the Vec? Towards Probabilistically Grounded Embeddings | Word2Vec (W2V) and GloVe are popular, fast and efficient word embedding
algorithms. Their embeddings are widely used and perform well on a variety of
natural language processing tasks. Moreover, W2V has recently been adopted in
the field of graph embedding, where it underpins several leading algorithms.
However, despite their ubiquity and relatively simple model architecture, a
theoretical understanding of what the embedding parameters of W2V and GloVe
learn and why that is useful in downstream tasks has been lacking. We show that
different interactions between PMI vectors reflect semantic word relationships,
such as similarity and paraphrasing, that are encoded in low dimensional word
embeddings under a suitable projection, theoretically explaining why embeddings
of W2V and GloVe work. As a consequence, we also reveal an interesting
mathematical interconnection between the considered semantic relationships
themselves.
|
A Flexible Framework for Multi-Objective Bayesian Optimization using
Random Scalarizations | Many real world applications can be framed as multi-objective optimization
problems, where we wish to simultaneously optimize for multiple criteria.
Bayesian optimization techniques for the multi-objective setting are pertinent
when the evaluation of the functions in question are expensive. Traditional
methods for multi-objective optimization, both Bayesian and otherwise, are
aimed at recovering the Pareto front of these objectives. However, in certain
cases a practitioner might desire to identify Pareto optimal points only in a
subset of the Pareto front due to external considerations. In this work, we
propose a strategy based on random scalarizations of the objectives that
addresses this problem. Our approach is able to flexibly sample from desired
regions of the Pareto front and, computationally, is considerably cheaper than
most approaches for MOO. We also study a notion of regret in the
multi-objective setting and show that our strategy achieves sublinear regret.
We experiment with both synthetic and real-life problems, and demonstrate
superior performance of our proposed algorithm in terms of the flexibility and
regret.
|
Deep Segment Hash Learning for Music Generation | Music generation research has grown in popularity over the past decade,
thanks to the deep learning revolution that has redefined the landscape of
artificial intelligence. In this paper, we propose a novel approach to music
generation inspired by musical segment concatenation methods and hash learning
algorithms. Given a segment of music, we use a deep recurrent neural network
and ranking-based hash learning to assign a forward hash code to the segment to
retrieve candidate segments for continuation with matching backward hash codes.
The proposed method is thus called Deep Segment Hash Learning (DSHL). To the
best of our knowledge, DSHL is the first end-to-end segment hash learning
method for music generation, and the first to use pair-wise training with
segments of music. We demonstrate that this method is capable of generating
music which is both original and enjoyable, and that DSHL offers a promising
new direction for music generation research.
|
Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural
Networks | Deep neural networks (DNNs) provide excellent performance across a wide range
of classification tasks, but their training requires high computational
resources and is often outsourced to third parties. Recent work has shown that
outsourced training introduces the risk that a malicious trainer will return a
backdoored DNN that behaves normally on most inputs but causes targeted
misclassifications or degrades the accuracy of the network when a trigger known
only to the attacker is present. In this paper, we provide the first effective
defenses against backdoor attacks on DNNs. We implement three backdoor attacks
from prior work and use them to investigate two promising defenses, pruning and
fine-tuning. We show that neither, by itself, is sufficient to defend against
sophisticated attackers. We then evaluate fine-pruning, a combination of
pruning and fine-tuning, and show that it successfully weakens or even
eliminates the backdoors, i.e., in some cases reducing the attack success rate
to 0% with only a 0.4% drop in accuracy for clean (non-triggering) inputs. Our
work provides the first step toward defenses against backdoor attacks in deep
neural networks.
|
Convolutional Embedded Networks for Population Scale Clustering and
Bio-ancestry Inferencing | The study of genetic variants can help find correlating population groups to
identify cohorts that are predisposed to common diseases and explain
differences in disease susceptibility and how patients react to drugs. Machine
learning algorithms are increasingly being applied to identify interacting GVs
to understand their complex phenotypic traits. Since the performance of a
learning algorithm not only depends on the size and nature of the data but also
on the quality of underlying representation, deep neural networks can learn
non-linear mappings that allow transforming GVs data into more clustering and
classification friendly representations than manual feature selection. In this
paper, we proposed convolutional embedded networks in which we combine two DNN
architectures called convolutional embedded clustering and convolutional
autoencoder classifier for clustering individuals and predicting geographic
ethnicity based on GVs, respectively. We employed CAE-based representation
learning on 95 million GVs from the 1000 genomes and Simons genome diversity
projects. Quantitative and qualitative analyses with a focus on accuracy and
scalability show that our approach outperforms state-of-the-art approaches such
as VariantSpark and ADMIXTURE. In particular, CEC can cluster targeted
population groups in 22 hours with an adjusted rand index of 0.915, the
normalized mutual information of 0.92, and the clustering accuracy of 89%.
Contrarily, the CAE classifier can predict the geographic ethnicity of unknown
samples with an F1 and Mathews correlation coefficient(MCC) score of 0.9004 and
0.8245, respectively. To provide interpretations of the predictions, we
identify significant biomarkers using gradient boosted trees(GBT) and SHAP.
Overall, our approach is transparent and faster than the baseline methods, and
scalable for 5% to 100% of the full human genome.
|
How Important Is a Neuron? | The problem of attributing a deep network's prediction to its
\emph{input/base} features is well-studied. We introduce the notion of
\emph{conductance} to extend the notion of attribution to the understanding the
importance of \emph{hidden} units.
Informally, the conductance of a hidden unit of a deep network is the
\emph{flow} of attribution via this hidden unit. We use conductance to
understand the importance of a hidden unit to the prediction for a specific
input, or over a set of inputs. We evaluate the effectiveness of conductance in
multiple ways, including theoretical properties, ablation studies, and a
feature selection task. The empirical evaluations are done using the Inception
network over ImageNet data, and a sentiment analysis network over reviews. In
both cases, we demonstrate the effectiveness of conductance in identifying
interesting insights about the internal workings of these networks.
|
Novel Video Prediction for Large-scale Scene using Optical Flow | Making predictions of future frames is a critical challenge in autonomous
driving research. Most of the existing methods for video prediction attempt to
generate future frames in simple and fixed scenes. In this paper, we propose a
novel and effective optical flow conditioned method for the task of video
prediction with an application to complex urban scenes. In contrast with
previous work, the prediction model only requires video sequences and optical
flow sequences for training and testing. Our method uses the rich
spatial-temporal features in video sequences. The method takes advantage of the
motion information extracting from optical flow maps between neighbor images as
well as previous images. Empirical evaluations on the KITTI dataset and the
Cityscapes dataset demonstrate the effectiveness of our method.
|
Mining gold from implicit models to improve likelihood-free inference | Simulators often provide the best description of real-world phenomena.
However, they also lead to challenging inverse problems because the density
they implicitly define is often intractable. We present a new suite of
simulation-based inference techniques that go beyond the traditional
Approximate Bayesian Computation approach, which struggles in a
high-dimensional setting, and extend methods that use surrogate models based on
neural networks. We show that additional information, such as the joint
likelihood ratio and the joint score, can often be extracted from simulators
and used to augment the training data for these surrogate models. Finally, we
demonstrate that these new techniques are more sample efficient and provide
higher-fidelity inference than traditional methods.
|
Optimized Participation of Multiple Fusion Functions in Consensus
Creation: An Evolutionary Approach | Recent studies show that ensemble methods enhance the stability and
robustness of unsupervised learning. These approaches are successfully utilized
to construct multiple clustering and combine them into a one representative
consensus clustering of an improved quality. The quality of the consensus
clustering is directly depended on fusion functions used in combination. In
this article, the hierarchical clustering ensemble techniques are extended by
introducing a new evolutionary fusion function. In the proposed method,
multiple hierarchical clustering methods are generated via bagging. Thereafter,
the consensus clustering is obtained using the search capability of genetic
algorithm among different aggregated clustering methods made by different
fusion functions. Putting some popular data sets to empirical study, the
quality of the proposed method is compared with regular clustering ensembles.
Experimental results demonstrate the accuracy improvement of the aggregated
clustering results.
|
Root-cause Analysis for Time-series Anomalies via Spatiotemporal
Graphical Modeling in Distributed Complex Systems | Performance monitoring, anomaly detection, and root-cause analysis in complex
cyber-physical systems (CPSs) are often highly intractable due to widely
diverse operational modes, disparate data types, and complex fault propagation
mechanisms. This paper presents a new data-driven framework for root-cause
analysis, based on a spatiotemporal graphical modeling approach built on the
concept of symbolic dynamics for discovering and representing causal
interactions among sub-systems of complex CPSs. We formulate the root-cause
analysis problem as a minimization problem via the proposed inference based
metric and present two approximate approaches for root-cause analysis, namely
the sequential state switching ($S^3$, based on free energy concept of a
restricted Boltzmann machine, RBM) and artificial anomaly association ($A^3$, a
classification framework using deep neural networks, DNN). Synthetic data from
cases with failed pattern(s) and anomalous node(s) are simulated to validate
the proposed approaches. Real dataset based on Tennessee Eastman process (TEP)
is also used for comparison with other approaches. The results show that: (1)
$S^3$ and $A^3$ approaches can obtain high accuracy in root-cause analysis
under both pattern-based and node-based fault scenarios, in addition to
successfully handling multiple nominal operating modes, (2) the proposed
tool-chain is shown to be scalable while maintaining high accuracy, and (3) the
proposed framework is robust and adaptive in different fault conditions and
performs better in comparison with the state-of-the-art methods.
|
Evaluating Reinforcement Learning Algorithms in Observational Health
Settings | Much attention has been devoted recently to the development of machine
learning algorithms with the goal of improving treatment policies in
healthcare. Reinforcement learning (RL) is a sub-field within machine learning
that is concerned with learning how to make sequences of decisions so as to
optimize long-term effects. Already, RL algorithms have been proposed to
identify decision-making strategies for mechanical ventilation, sepsis
management and treatment of schizophrenia. However, before implementing
treatment policies learned by black-box algorithms in high-stakes clinical
decision problems, special care must be taken in the evaluation of these
policies.
In this document, our goal is to expose some of the subtleties associated
with evaluating RL algorithms in healthcare. We aim to provide a conceptual
starting point for clinical and computational researchers to ask the right
questions when designing and evaluating algorithms for new ways of treating
patients. In the following, we describe how choices about how to summarize a
history, variance of statistical estimators, and confounders in more ad-hoc
measures can result in unreliable, even misleading estimates of the quality of
a treatment policy. We also provide suggestions for mitigating these
effects---for while there is much promise for mining observational health data
to uncover better treatment policies, evaluation must be performed
thoughtfully.
|
Rotation Equivariance and Invariance in Convolutional Neural Networks | Performance of neural networks can be significantly improved by encoding
known invariance for particular tasks. Many image classification tasks, such as
those related to cellular imaging, exhibit invariance to rotation. We present a
novel scheme using the magnitude response of the 2D-discrete-Fourier transform
(2D-DFT) to encode rotational invariance in neural networks, along with a new,
efficient convolutional scheme for encoding rotational equivariance throughout
convolutional layers. We implemented this scheme for several image
classification tasks and demonstrated improved performance, in terms of
classification accuracy, time required to train the model, and robustness to
hyperparameter selection, over a standard CNN and another state-of-the-art
method.
|
Adversarial Attacks on Face Detectors using Neural Net based Constrained
Optimization | Adversarial attacks involve adding, small, often imperceptible, perturbations
to inputs with the goal of getting a machine learning model to misclassifying
them. While many different adversarial attack strategies have been proposed on
image classification models, object detection pipelines have been much harder
to break. In this paper, we propose a novel strategy to craft adversarial
examples by solving a constrained optimization problem using an adversarial
generator network. Our approach is fast and scalable, requiring only a forward
pass through our trained generator network to craft an adversarial sample.
Unlike in many attack strategies, we show that the same trained generator is
capable of attacking new images without explicitly optimizing on them. We
evaluate our attack on a trained Faster R-CNN face detector on the cropped
300-W face dataset where we manage to reduce the number of detected faces to
$0.5\%$ of all originally detected faces. In a different experiment, also on
300-W, we demonstrate the robustness of our attack to a JPEG compression based
defense typical JPEG compression level of $75\%$ reduces the effectiveness of
our attack from only $0.5\%$ of detected faces to a modest $5.0\%$.
|
Conformation Clustering of Long MD Protein Dynamics with an Adversarial
Autoencoder | Recent developments in specialized computer hardware have greatly accelerated
atomic level Molecular Dynamics (MD) simulations. A single GPU-attached cluster
is capable of producing microsecond-length trajectories in reasonable amounts
of time. Multiple protein states and a large number of microstates associated
with folding and with the function of the protein can be observed as
conformations sampled in the trajectories. Clustering those conformations,
however, is needed for identifying protein states, evaluating transition rates
and understanding protein behavior. In this paper, we propose a novel
data-driven generative conformation clustering method based on the adversarial
autoencoder (AAE) and provide the associated software implementation Cong. The
method was tested using a 208 microseconds MD simulation of the fast-folding
peptide Trp-Cage (20 residues) obtained from the D.E. Shaw Research Group. The
proposed clustering algorithm identifies many of the salient features of the
folding process by grouping a large number of conformations that share common
features not easily identifiable in the trajectory.
|
Greedy Attack and Gumbel Attack: Generating Adversarial Examples for
Discrete Data | We present a probabilistic framework for studying adversarial attacks on
discrete data. Based on this framework, we derive a perturbation-based method,
Greedy Attack, and a scalable learning-based method, Gumbel Attack, that
illustrate various tradeoffs in the design of attacks. We demonstrate the
effectiveness of these methods using both quantitative metrics and human
evaluation on various state-of-the-art models for text classification,
including a word-based CNN, a character-based CNN and an LSTM. As as example of
our results, we show that the accuracy of character-based convolutional
networks drops to the level of random selection by modifying only five
characters through Greedy Attack.
|
Multiaccuracy: Black-Box Post-Processing for Fairness in Classification | Prediction systems are successfully deployed in applications ranging from
disease diagnosis, to predicting credit worthiness, to image recognition. Even
when the overall accuracy is high, these systems may exhibit systematic biases
that harm specific subpopulations; such biases may arise inadvertently due to
underrepresentation in the data used to train a machine-learning model, or as
the result of intentional malicious discrimination. We develop a rigorous
framework of *multiaccuracy* auditing and post-processing to ensure accurate
predictions across *identifiable subgroups*.
Our algorithm, MULTIACCURACY-BOOST, works in any setting where we have
black-box access to a predictor and a relatively small set of labeled data for
auditing; importantly, this black-box framework allows for improved fairness
and accountability of predictions, even when the predictor is minimally
transparent. We prove that MULTIACCURACY-BOOST converges efficiently and show
that if the initial model is accurate on an identifiable subgroup, then the
post-processed model will be also. We experimentally demonstrate the
effectiveness of the approach to improve the accuracy among minority subgroups
in diverse applications (image classification, finance, population health).
Interestingly, MULTIACCURACY-BOOST can improve subpopulation accuracy (e.g. for
"black women") even when the sensitive features (e.g. "race", "gender") are not
given to the algorithm explicitly.
|
A Divide-and-Conquer Approach to Geometric Sampling for Active Learning | Active learning (AL) repeatedly trains the classifier with the minimum
labeling budget to improve the current classification model. The training
process is usually supervised by an uncertainty evaluation strategy. However,
the uncertainty evaluation always suffers from performance degeneration when
the initial labeled set has insufficient labels. To completely eliminate the
dependence on the uncertainty evaluation sampling in AL, this paper proposes a
divide-and-conquer idea that directly transfers the AL sampling as the
geometric sampling over the clusters. By dividing the points of the clusters
into cluster boundary and core points, we theoretically discuss their margin
distance and {hypothesis relationship}. With the advantages of cluster boundary
points in the above two properties, we propose a Geometric Active Learning
(GAL) algorithm by knight's tour. Experimental studies of the two reported
experimental tasks including cluster boundary detection and AL classification
show that the proposed GAL method significantly outperforms the
state-of-the-art baselines.
|
Metric on Nonlinear Dynamical Systems with Perron-Frobenius Operators | The development of a metric for structural data is a long-term problem in
pattern recognition and machine learning. In this paper, we develop a general
metric for comparing nonlinear dynamical systems that is defined with
Perron-Frobenius operators in reproducing kernel Hilbert spaces. Our metric
includes the existing fundamental metrics for dynamical systems, which are
basically defined with principal angles between some appropriately-chosen
subspaces, as its special cases. We also describe the estimation of our metric
from finite data. We empirically illustrate our metric with an example of
rotation dynamics in a unit disk in a complex plane, and evaluate the
performance with real-world time-series data.
|
On representation power of neural network-based graph embedding and
beyond | We consider the representation power of siamese-style similarity functions
used in neural network-based graph embedding. The inner product similarity
(IPS) with feature vectors computed via neural networks is commonly used for
representing the strength of association between two nodes. However, only a
little work has been done on the representation capability of IPS. A very
recent work shed light on the nature of IPS and reveals that IPS has the
capability of approximating any positive definite (PD) similarities. However, a
simple example demonstrates the fundamental limitation of IPS to approximate
non-PD similarities. We then propose a novel model named Shifted IPS (SIPS)
that approximates any Conditionally PD (CPD) similarities arbitrary well. CPD
is a generalization of PD with many examples such as negative Poincar\'e
distance and negative Wasserstein distance, thus SIPS has a potential impact to
significantly improve the applicability of graph embedding without taking great
care in configuring the similarity function. Our numerical experiments
demonstrate the SIPS's superiority over IPS. In theory, we further extend SIPS
beyond CPD by considering the inner product in Minkowski space so that it
approximates more general similarities.
|
DialogWAE: Multimodal Response Generation with Conditional Wasserstein
Auto-Encoder | Variational autoencoders~(VAEs) have shown a promise in data-driven
conversation modeling. However, most VAE conversation models match the
approximate posterior distribution over the latent variables to a simple prior
such as standard normal distribution, thereby restricting the generated
responses to a relatively simple (e.g., unimodal) scope. In this paper, we
propose DialogWAE, a conditional Wasserstein autoencoder~(WAE) specially
designed for dialogue modeling. Unlike VAEs that impose a simple distribution
over the latent variables, DialogWAE models the distribution of data by
training a GAN within the latent variable space. Specifically, our model
samples from the prior and posterior distributions over the latent variables by
transforming context-dependent random noise using neural networks and minimizes
the Wasserstein distance between the two distributions. We further develop a
Gaussian mixture prior network to enrich the latent space. Experiments on two
popular datasets show that DialogWAE outperforms the state-of-the-art
approaches in generating more coherent, informative and diverse responses.
|
Classification of volcanic ash particles using a convolutional neural
network and probability | Analyses of volcanic ash are typically performed either by qualitatively
classifying ash particles by eye or by quantitatively parameterizing its shape
and texture. While complex shapes can be classified through qualitative
analyses, the results are subjective due to the difficulty of categorizing
complex shapes into a single class. Although quantitative analyses are
objective, selection of shape parameters is required. Here, we applied a
convolutional neural network (CNN) for the classification of volcanic ash.
First, we defined four basal particle shapes (blocky, vesicular, elongated,
rounded) generated by different eruption mechanisms (e.g., brittle
fragmentation), and then trained the CNN using particles composed of only one
basal shape. The CNN could recognize the basal shapes with over 90% accuracy.
Using the trained network, we classified ash particles composed of multiple
basal shapes based on the output of the network, which can be interpreted as a
mixing ratio of the four basal shapes. Clustering of samples by the averaged
probabilities and the intensity is consistent with the eruption type. The
mixing ratio output by the CNN can be used to quantitatively classify complex
shapes in nature without categorizing forcibly and without the need for shape
parameters, which may lead to a new taxonomy.
|
Deep-Energy: Unsupervised Training of Deep Neural Networks | The success of deep learning has been due, in no small part, to the
availability of large annotated datasets. Thus, a major bottleneck in current
learning pipelines is the time-consuming human annotation of data. In scenarios
where such input-output pairs cannot be collected, simulation is often used
instead, leading to a domain-shift between synthesized and real-world data.
This work offers an unsupervised alternative that relies on the availability of
task-specific energy functions, replacing the generic supervised loss. Such
energy functions are assumed to lead to the desired label as their minimizer
given the input. The proposed approach, termed "Deep Energy", trains a Deep
Neural Network (DNN) to approximate this minimization for any chosen input.
Once trained, a simple and fast feed-forward computation provides the inferred
label. This approach allows us to perform unsupervised training of DNNs with
real-world inputs only, and without the need for manually-annotated labels, nor
synthetically created data. "Deep Energy" is demonstrated in this paper on
three different tasks -- seeded segmentation, image matting and single image
dehazing -- exposing its generality and wide applicability. Our experiments
show that the solution provided by the network is often much better in quality
than the one obtained by a direct minimization of the energy function,
suggesting an added regularization property in our scheme.
|
Reinforced Continual Learning | Most artificial intelligence models have limiting ability to solve new tasks
faster, without forgetting previously acquired knowledge. The recently emerging
paradigm of continual learning aims to solve this issue, in which the model
learns various tasks in a sequential fashion. In this work, a novel approach
for continual learning is proposed, which searches for the best neural
architecture for each coming task via sophisticatedly designed reinforcement
learning strategies. We name it as Reinforced Continual Learning. Our method
not only has good performance on preventing catastrophic forgetting but also
fits new tasks well. The experiments on sequential classification tasks for
variants of MNIST and CIFAR-100 datasets demonstrate that the proposed approach
outperforms existing continual learning alternatives for deep networks.
|
Learning Tree Distributions by Hidden Markov Models | Hidden tree Markov models allow learning distributions for tree structured
data while being interpretable as nondeterministic automata. We provide a
concise summary of the main approaches in literature, focusing in particular on
the causality assumptions introduced by the choice of a specific tree visit
direction. We will then sketch a novel non-parametric generalization of the
bottom-up hidden tree Markov model with its interpretation as a
nondeterministic tree automaton with infinite states.
|
Sample-Efficient Deep Reinforcement Learning via Episodic Backward
Update | We propose Episodic Backward Update (EBU) - a novel deep reinforcement
learning algorithm with a direct value propagation. In contrast to the
conventional use of the experience replay with uniform random sampling, our
agent samples a whole episode and successively propagates the value of a state
to its previous states. Our computationally efficient recursive algorithm
allows sparse and delayed rewards to propagate directly through all transitions
of the sampled episode. We theoretically prove the convergence of the EBU
method and experimentally demonstrate its performance in both deterministic and
stochastic environments. Especially in 49 games of Atari 2600 domain, EBU
achieves the same mean and median human normalized performance of DQN by using
only 5% and 10% of samples, respectively.
|
Imbalanced Ensemble Classifier for learning from imbalanced business
school data set | Private business schools in India face a common problem of selecting quality
students for their MBA programs to achieve the desired placement percentage.
Generally, such data sets are biased towards one class, i.e., imbalanced in
nature. And learning from the imbalanced dataset is a difficult proposition.
This paper proposes an imbalanced ensemble classifier which can handle the
imbalanced nature of the dataset and achieves higher accuracy in case of the
feature selection (selection of important characteristics of students) cum
classification problem (prediction of placements based on the students'
characteristics) for Indian business school dataset. The optimal value of an
important model parameter is found. Numerical evidence is also provided using
Indian business school dataset to assess the outstanding performance of the
proposed classifier.
|
Agents and Devices: A Relative Definition of Agency | According to Dennett, the same system may be described using a `physical'
(mechanical) explanatory stance, or using an `intentional' (belief- and
goal-based) explanatory stance. Humans tend to find the physical stance more
helpful for certain systems, such as planets orbiting a star, and the
intentional stance for others, such as living animals. We define a formal
counterpart of physical and intentional stances within computational theory: a
description of a system as either a device, or an agent, with the key
difference being that `devices' are directly described in terms of an
input-output mapping, while `agents' are described in terms of the function
they optimise. Bayes' rule can then be applied to calculate the subjective
probability of a system being a device or an agent, based only on its
behaviour. We illustrate this using the trajectories of an object in a toy
grid-world domain.
|
KG^2: Learning to Reason Science Exam Questions with Contextual
Knowledge Graph Embeddings | The AI2 Reasoning Challenge (ARC), a new benchmark dataset for question
answering (QA) has been recently released. ARC only contains natural science
questions authored for human exams, which are hard to answer and require
advanced logic reasoning. On the ARC Challenge Set, existing state-of-the-art
QA systems fail to significantly outperform random baseline, reflecting the
difficult nature of this task. In this paper, we propose a novel framework for
answering science exam questions, which mimics human solving process in an
open-book exam. To address the reasoning challenge, we construct contextual
knowledge graphs respectively for the question itself and supporting sentences.
Our model learns to reason with neural embeddings of both knowledge graphs.
Experiments on the ARC Challenge Set show that our model outperforms the
previous state-of-the-art QA systems.
|
HOPF: Higher Order Propagation Framework for Deep Collective
Classification | Given a graph where every node has certain attributes associated with it and
some nodes have labels associated with them, Collective Classification (CC) is
the task of assigning labels to every unlabeled node using information from the
node as well as its neighbors. It is often the case that a node is not only
influenced by its immediate neighbors but also by higher order neighbors,
multiple hops away. Recent state-of-the-art models for CC learn end-to-end
differentiable variations of Weisfeiler-Lehman (WL) kernels to aggregate
multi-hop neighborhood information. In this work, we propose a Higher Order
Propagation Framework, HOPF, which provides an iterative inference mechanism
for these powerful differentiable kernels. Such a combination of classical
iterative inference mechanism with recent differentiable kernels allows the
framework to learn graph convolutional filters that simultaneously exploit the
attribute and label information available in the neighborhood. Further, these
iterative differentiable kernels can scale to larger hops beyond the memory
limitations of existing differentiable kernels. We also show that existing WL
kernel-based models suffer from the problem of Node Information Morphing where
the information of the node is morphed or overwhelmed by the information of its
neighbors when considering multiple hops. To address this, we propose a
specific instantiation of HOPF, called the NIP models, which preserves the node
information at every propagation step. The iterative formulation of NIP models
further helps in incorporating distant hop information concisely as summaries
of the inferred labels. We do an extensive evaluation across 11 datasets from
different domains. We show that existing CC models do not provide consistent
performance across datasets, while the proposed NIP model with iterative
inference is more robust.
|
On GANs and GMMs | A longstanding problem in machine learning is to find unsupervised methods
that can learn the statistical structure of high dimensional signals. In recent
years, GANs have gained much attention as a possible solution to the problem,
and in particular have shown the ability to generate remarkably realistic high
resolution sampled images. At the same time, many authors have pointed out that
GANs may fail to model the full distribution ("mode collapse") and that using
the learned models for anything other than generating samples may be very
difficult. In this paper, we examine the utility of GANs in learning
statistical models of images by comparing them to perhaps the simplest
statistical model, the Gaussian Mixture Model. First, we present a simple
method to evaluate generative models based on relative proportions of samples
that fall into predetermined bins. Unlike previous automatic methods for
evaluating models, our method does not rely on an additional neural network nor
does it require approximating intractable computations. Second, we compare the
performance of GANs to GMMs trained on the same datasets. While GMMs have
previously been shown to be successful in modeling small patches of images, we
show how to train them on full sized images despite the high dimensionality.
Our results show that GMMs can generate realistic samples (although less sharp
than those of GANs) but also capture the full distribution, which GANs fail to
do. Furthermore, GMMs allow efficient inference and explicit representation of
the underlying statistical structure. Finally, we discuss how GMMs can be used
to generate sharp images.
|
Distributed Estimation of Gaussian Correlations | We study a distributed estimation problem in which two remotely located
parties, Alice and Bob, observe an unlimited number of i.i.d. samples
corresponding to two different parts of a random vector. Alice can send $k$
bits on average to Bob, who in turn wants to estimate the cross-correlation
matrix between the two parts of the vector. In the case where the parties
observe jointly Gaussian scalar random variables with an unknown correlation
$\rho$, we obtain two constructive and simple unbiased estimators attaining a
variance of $(1-\rho^2)/(2k\ln 2)$, which coincides with a known but
non-constructive random coding result of Zhang and Berger. We extend our
approach to the vector Gaussian case, which has not been treated before, and
construct an estimator that is uniformly better than the scalar estimator
applied separately to each of the correlations. We then show that the Gaussian
performance can essentially be attained even when the distribution is
completely unknown. This in particular implies that in the general problem of
distributed correlation estimation, the variance can decay at least as $O(1/k)$
with the number of transmitted bits. This behavior, however, is not tight: we
give an example of a rich family of distributions for which local samples
reveal essentially nothing about the correlations, and where a slightly
modified estimator attains a variance of $2^{-\Omega(k)}$.
|
Sequential Attacks on Agents for Long-Term Adversarial Goals | Reinforcement learning (RL) has advanced greatly in the past few years with
the employment of effective deep neural networks (DNNs) on the policy networks.
With the great effectiveness came serious vulnerability issues with DNNs that
small adversarial perturbations on the input can change the output of the
network. Several works have pointed out that learned agents with a DNN policy
network can be manipulated against achieving the original task through a
sequence of small perturbations on the input states. In this paper, we
demonstrate furthermore that it is also possible to impose an arbitrary
adversarial reward on the victim policy network through a sequence of attacks.
Our method involves the latest adversarial attack technique, Adversarial
Transformer Network (ATN), that learns to generate the attack and is easy to
integrate into the policy network. As a result of our attack, the victim agent
is misguided to optimise for the adversarial reward over time. Our results
expose serious security threats for RL applications in safety-critical systems
including drones, medical analysis, and self-driving cars.
|
Efficacy of regularized multi-task learning based on SVM models | This paper investigates the efficacy of a regularized multi-task learning
(MTL) framework based on SVM (M-SVM) to answer whether MTL always provides
reliable results and how MTL outperforms independent learning. We first find
that M-SVM is Bayes risk consistent in the limit of large sample size. This
implies that despite the task dissimilarities, M-SVM always produces a reliable
decision rule for each task in terms of misclassification error when the data
size is large enough. Furthermore, we find that the task-interaction vanishes
as the data size goes to infinity, and the convergence rates of M-SVM and its
single-task counterpart have the same upper bound. The former suggests that
M-SVM cannot improve the limit classifier's performance; based on the latter,
we conjecture that the optimal convergence rate is not improved when the task
number is fixed. As a novel insight of MTL, our theoretical and experimental
results achieved an excellent agreement that the benefit of the MTL methods
lies in the improvement of the pre-convergence-rate factor (PCR, to be denoted
in Section III) rather than the convergence rate. Moreover, this improvement of
PCR factors is more significant when the data size is small.
|
Cyberattack Detection using Deep Generative Models with Variational
Inference | Recent years have witnessed a rise in the frequency and intensity of
cyberattacks targeted at critical infrastructure systems. This study designs a
versatile, data-driven cyberattack detection platform for infrastructure
systems cybersecurity, with a special demonstration in water sector. A deep
generative model with variational inference autonomously learns normal system
behavior and detects attacks as they occur. The model can process the natural
data in its raw form and automatically discover and learn its representations,
hence augmenting system knowledge discovery and reducing the need for laborious
human engineering and domain expertise. The proposed model is applied to a
simulated cyberattack detection problem involving a drinking water distribution
system subject to programmable logic controller hacks, malicious actuator
activation, and deception attacks. The model is only provided with observations
of the system, such as pump pressure and tank water level reads, and is blind
to the internal structures and workings of the water distribution system. The
simulated attacks are manifested in the model's generated reproduction
probability plot, indicating its ability to discern the attacks. There is,
however, need for improvements in reducing false alarms, especially by
optimizing detection thresholds. Altogether, the results indicate ability of
the model in distinguishing attacks and their repercussions from normal system
operation in water distribution systems, and the promise it holds for
cyberattack detection in other domains.
|
Scaling provable adversarial defenses | Recent work has developed methods for learning deep network classifiers that
are provably robust to norm-bounded adversarial perturbation; however, these
methods are currently only possible for relatively small feedforward networks.
In this paper, in an effort to scale these approaches to substantially larger
models, we extend previous work in three main directions. First, we present a
technique for extending these training procedures to much more general
networks, with skip connections (such as ResNets) and general nonlinearities;
the approach is fully modular, and can be implemented automatically (analogous
to automatic differentiation). Second, in the specific case of $\ell_\infty$
adversarial perturbations and networks with ReLU nonlinearities, we adopt a
nonlinear random projection for training, which scales linearly in the number
of hidden units (previous approaches scaled quadratically). Third, we show how
to further improve robust error through cascade models. On both MNIST and CIFAR
data sets, we train classifiers that improve substantially on the state of the
art in provable robust adversarial error bounds: from 5.8% to 3.1% on MNIST
(with $\ell_\infty$ perturbations of $\epsilon=0.1$), and from 80% to 36.4% on
CIFAR (with $\ell_\infty$ perturbations of $\epsilon=2/255$). Code for all
experiments in the paper is available at
https://github.com/locuslab/convex_adversarial/.
|
Fusion Graph Convolutional Networks | Semi-supervised node classification in attributed graphs, i.e., graphs with
node features, involves learning to classify unlabeled nodes given a partially
labeled graph. Label predictions are made by jointly modeling the node and its'
neighborhood features. State-of-the-art models for node classification on such
attributed graphs use differentiable recursive functions that enable
aggregation and filtering of neighborhood information from multiple hops. In
this work, we analyze the representation capacity of these models to regulate
information from multiple hops independently. From our analysis, we conclude
that these models despite being powerful, have limited representation capacity
to capture multi-hop neighborhood information effectively. Further, we also
propose a mathematically motivated, yet simple extension to existing graph
convolutional networks (GCNs) which has improved representation capacity. We
extensively evaluate the proposed model, F-GCN on eight popular datasets from
different domains. F-GCN outperforms the state-of-the-art models for
semi-supervised learning on six datasets while being extremely competitive on
the other two.
|
Analysis of Fast Structured Dictionary Learning | Sparsity-based models and techniques have been exploited in many signal
processing and imaging applications. Data-driven methods based on dictionary
and sparsifying transform learning enable learning rich image features from
data, and can outperform analytical models. In particular, alternating
optimization algorithms have been popular for learning such models. In this
work, we focus on alternating minimization for a specific structured unitary
sparsifying operator learning problem, and provide a convergence analysis.
While the algorithm converges to the critical points of the problem generally,
our analysis establishes under mild assumptions, the local linear convergence
of the algorithm to the underlying sparsifying model of the data. Analysis and
numerical simulations show that our assumptions hold for standard probabilistic
data models. In practice, the algorithm is robust to initialization.
|
Long-time predictive modeling of nonlinear dynamical systems using
neural networks | We study the use of feedforward neural networks (FNN) to develop models of
nonlinear dynamical systems from data. Emphasis is placed on predictions at
long times, with limited data availability. Inspired by global stability
analysis, and the observation of the strong correlation between the local error
and the maximum singular value of the Jacobian of the ANN, we introduce
Jacobian regularization in the loss function. This regularization suppresses
the sensitivity of the prediction to the local error and is shown to improve
accuracy and robustness. Comparison between the proposed approach and sparse
polynomial regression is presented in numerical examples ranging from simple
ODE systems to nonlinear PDE systems including vortex shedding behind a
cylinder, and instability-driven buoyant mixing flow. Furthermore, limitations
of feedforward neural networks are highlighted, especially when the training
data does not include a low dimensional attractor. Strategies of data
augmentation are presented as remedies to address these issues to a certain
extent.
|
Channel Gating Neural Networks | This paper introduces channel gating, a dynamic, fine-grained, and
hardware-efficient pruning scheme to reduce the computation cost for
convolutional neural networks (CNNs). Channel gating identifies regions in the
features that contribute less to the classification result, and skips the
computation on a subset of the input channels for these ineffective regions.
Unlike static network pruning, channel gating optimizes CNN inference at
run-time by exploiting input-specific characteristics, which allows
substantially reducing the compute cost with almost no accuracy loss. We
experimentally show that applying channel gating in state-of-the-art networks
achieves 2.7-8.0$\times$ reduction in floating-point operations (FLOPs) and
2.0-4.4$\times$ reduction in off-chip memory accesses with a minimal accuracy
loss on CIFAR-10. Combining our method with knowledge distillation reduces the
compute cost of ResNet-18 by 2.6$\times$ without accuracy drop on ImageNet. We
further demonstrate that channel gating can be realized in hardware
efficiently. Our approach exhibits sparsity patterns that are well-suited to
dense systolic arrays with minimal additional hardware. We have designed an
accelerator for channel gating networks, which can be implemented using either
FPGAs or ASICs. Running a quantized ResNet-18 model for ImageNet, our
accelerator achieves an encouraging speedup of 2.4$\times$ on average, with a
theoretical FLOP reduction of 2.8$\times$.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.