title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Computational Red Teaming in a Sudoku Solving Context: Neural Network
Based Skill Representation and Acquisition | In this paper we provide an insight into the skill representation, where
skill representation is seen as an essential part of the skill assessment stage
in the Computational Red Teaming process. Skill representation is demonstrated
in the context of Sudoku puzzle, for which the real human skills used in Sudoku
solving, along with their acquisition, are represented computationally in a
cognitively plausible manner, by using feed-forward neural networks with
back-propagation, and supervised learning. The neural network based skills are
then coupled with a hard-coded constraint propagation computational Sudoku
solver, in which the solving sequence is kept hard-coded, and the skills are
represented through neural networks. The paper demonstrates that the modified
solver can achieve different levels of proficiency, depending on the amount of
skills acquired through the neural networks. Results are encouraging for
developing more complex skill and skill acquisition models usable in general
frameworks related to the skill assessment aspect of Computational Red Teaming.
|
Multi-Observation Regression | Recent work introduced loss functions which measure the error of a prediction
based on multiple simultaneous observations or outcomes. In this paper, we
explore the theoretical and practical questions that arise when using such
multi-observation losses for regression on data sets of $(x,y)$ pairs. When a
loss depends on only one observation, the average empirical loss decomposes by
applying the loss to each pair, but for the multi-observation case, empirical
loss is not even well-defined, and the possibility of statistical guarantees is
unclear without several $(x,y)$ pairs with exactly the same $x$ value. We
propose four algorithms formalizing the concept of empirical risk minimization
for this problem, two of which have statistical guarantees in settings allowing
both slow and fast convergence rates, but which are out-performed empirically
by the other two. Empirical results demonstrate practicality of these
algorithms in low-dimensional settings, while lower bounds demonstrate
intrinsic difficulty in higher dimensions. Finally, we demonstrate the
potential benefit of the algorithms over natural baselines that use traditional
single-observation losses via both lower bounds and simulations.
|
Link Prediction Based on Graph Neural Networks | Link prediction is a key problem for network-structured data. Link prediction
heuristics use some score functions, such as common neighbors and Katz index,
to measure the likelihood of links. They have obtained wide practical uses due
to their simplicity, interpretability, and for some of them, scalability.
However, every heuristic has a strong assumption on when two nodes are likely
to link, which limits their effectiveness on networks where these assumptions
fail. In this regard, a more reasonable way should be learning a suitable
heuristic from a given network instead of using predefined ones. By extracting
a local subgraph around each target link, we aim to learn a function mapping
the subgraph patterns to link existence, thus automatically learning a
`heuristic' that suits the current network. In this paper, we study this
heuristic learning paradigm for link prediction. First, we develop a novel
$\gamma$-decaying heuristic theory. The theory unifies a wide range of
heuristics in a single framework, and proves that all these heuristics can be
well approximated from local subgraphs. Our results show that local subgraphs
reserve rich information related to link existence. Second, based on the
$\gamma$-decaying theory, we propose a new algorithm to learn heuristics from
local subgraphs using a graph neural network (GNN). Its experimental results
show unprecedented performance, working consistently well on a wide range of
problems.
|
Convolutional Neural Network Achieves Human-level Accuracy in Music
Genre Classification | Music genre classification is one example of content-based analysis of music
signals. Traditionally, human-engineered features were used to automatize this
task and 61% accuracy has been achieved in the 10-genre classification.
However, it's still below the 70% accuracy that humans could achieve in the
same task. Here, we propose a new method that combines knowledge of human
perception study in music genre classification and the neurophysiology of the
auditory system. The method works by training a simple convolutional neural
network (CNN) to classify a short segment of the music signal. Then, the genre
of a music is determined by splitting it into short segments and then combining
CNN's predictions from all short segments. After training, this method achieves
human-level (70%) accuracy and the filters learned in the CNN resemble the
spectrotemporal receptive field (STRF) in the auditory system.
|
Robust GANs against Dishonest Adversaries | Robustness of deep learning models is a property that has recently gained
increasing attention. We explore a notion of robustness for generative
adversarial models that is pertinent to their internal interactive structure,
and show that, perhaps surprisingly, the GAN in its original form is not
robust. Our notion of robustness relies on a perturbed discriminator, or noisy,
adversarial interference with its feedback. We explore, theoretically and
empirically, the effect of model and training properties on this robustness. In
particular, we show theoretical conditions for robustness that are supported by
empirical evidence. We also test the effect of regularization. Our results
suggest variations of GANs that are indeed more robust to noisy attacks and
have more stable training behavior, requiring less regularization in general.
Inspired by our theoretical results, we further extend our framework to obtain
a class of models related to WGAN, with good empirical performance. Overall,
our results suggest a new perspective on understanding and designing GAN models
from the viewpoint of their internal robustness.
|
Understanding and Enhancing the Transferability of Adversarial Examples | State-of-the-art deep neural networks are known to be vulnerable to
adversarial examples, formed by applying small but malicious perturbations to
the original inputs. Moreover, the perturbations can \textit{transfer across
models}: adversarial examples generated for a specific model will often mislead
other unseen models. Consequently the adversary can leverage it to attack
deployed systems without any query, which severely hinder the application of
deep learning, especially in the areas where security is crucial. In this work,
we systematically study how two classes of factors that might influence the
transferability of adversarial examples. One is about model-specific factors,
including network architecture, model capacity and test accuracy. The other is
the local smoothness of loss function for constructing adversarial examples.
Based on these understanding, a simple but effective strategy is proposed to
enhance transferability. We call it variance-reduced attack, since it utilizes
the variance-reduced gradient to generate adversarial example. The
effectiveness is confirmed by a variety of experiments on both CIFAR-10 and
ImageNet datasets.
|
Robust Actor-Critic Contextual Bandit for Mobile Health (mHealth)
Interventions | We consider the actor-critic contextual bandit for the mobile health
(mHealth) intervention. State-of-the-art decision-making algorithms generally
ignore the outliers in the dataset. In this paper, we propose a novel robust
contextual bandit method for the mHealth. It can achieve the conflicting goal
of reducing the influence of outliers while seeking for a similar solution
compared with the state-of-the-art contextual bandit methods on the datasets
without outliers. Such performance relies on two technologies: (1) the
capped-$\ell_{2}$ norm; (2) a reliable method to set the thresholding
hyper-parameter, which is inspired by one of the most fundamental techniques in
the statistics. Although the model is non-convex and non-differentiable, we
propose an effective reweighted algorithm and provide solid theoretical
analyses. We prove that the proposed algorithm can find sufficiently decreasing
points after each iteration and finally converges after a finite number of
iterations. Extensive experiment results on two datasets demonstrate that our
method can achieve almost identical results compared with state-of-the-art
contextual bandit methods on the dataset without outliers, and significantly
outperform those state-of-the-art methods on the badly noised dataset with
outliers in a variety of parameter settings.
|
Network-Clustered Multi-Modal Bug Localization | Developers often spend much effort and resources to debug a program. To help
the developers debug, numerous information retrieval (IR)-based and
spectrum-based bug localization techniques have been devised. IR-based
techniques process textual information in bug reports, while spectrum-based
techniques process program spectra (i.e., a record of which program elements
are executed for each test case). While both techniques ultimately generate a
ranked list of program elements that likely contain a bug, they only consider
one source of information--either bug reports or program spectra--which is not
optimal. In light of this deficiency, this paper presents a new approach dubbed
Network-clustered Multi-modal Bug Localization (NetML), which utilizes
multi-modal information from both bug reports and program spectra to localize
bugs. NetML facilitates an effective bug localization by carrying out a joint
optimization of bug localization error and clustering of both bug reports and
program elements (i.e., methods). The clustering is achieved through the
incorporation of network Lasso regularization, which incentivizes the model
parameters of similar bug reports and similar program elements to be close
together. To estimate the model parameters of both bug reports and methods,
NetML employs an adaptive learning procedure based on Newton method that
updates the parameters on a per-feature basis. Extensive experiments on 355
real bugs from seven software systems have been conducted to benchmark NetML
against various state-of-the-art localization methods. The results show that
NetML surpasses the best-performing baseline by 31.82%, 22.35%, 19.72%, and
19.24%, in terms of the number of bugs successfully localized when a developer
inspects the top 1, 5, and 10 methods and Mean Average Precision (MAP),
respectively.
|
Online learning with kernel losses | We present a generalization of the adversarial linear bandits framework,
where the underlying losses are kernel functions (with an associated
reproducing kernel Hilbert space) rather than linear functions. We study a
version of the exponential weights algorithm and bound its regret in this
setting. Under conditions on the eigendecay of the kernel we provide a sharp
characterization of the regret for this algorithm. When we have polynomial
eigendecay $\mu_j \le \mathcal{O}(j^{-\beta})$, we find that the regret is
bounded by $\mathcal{R}_n \le \mathcal{O}(n^{\beta/(2(\beta-1))})$; while under
the assumption of exponential eigendecay $\mu_j \le \mathcal{O}(e^{-\beta j
})$, we get an even tighter bound on the regret $\mathcal{R}_n \le
\mathcal{O}(n^{1/2}\log(n)^{1/2})$. We also study the full information setting
when the underlying losses are kernel functions and present an adapted
exponential weights algorithm and a conditional gradient descent algorithm.
|
Accelerating Asynchronous Algorithms for Convex Optimization by Momentum
Compensation | Asynchronous algorithms have attracted much attention recently due to the
crucial demands on solving large-scale optimization problems. However, the
accelerated versions of asynchronous algorithms are rarely studied. In this
paper, we propose the "momentum compensation" technique to accelerate
asynchronous algorithms for convex problems. Specifically, we first accelerate
the plain Asynchronous Gradient Descent, which achieves a faster
$O(1/\sqrt{\epsilon})$ (v.s. $O(1/\epsilon)$) convergence rate for non-strongly
convex functions, and $O(\sqrt{\kappa}\log(1/\epsilon))$ (v.s. $O(\kappa
\log(1/\epsilon))$) for strongly convex functions to reach an $\epsilon$-
approximate minimizer with the condition number $\kappa$. We further apply the
technique to accelerate modern stochastic asynchronous algorithms such as
Asynchronous Stochastic Coordinate Descent and Asynchronous Stochastic Gradient
Descent. Both of the resultant practical algorithms are faster than existing
ones by order. To the best of our knowledge, we are the first to consider
accelerated algorithms that allow updating by delayed gradients and are the
first to propose truly accelerated asynchronous algorithms. Finally, the
experimental results on a shared memory system show that acceleration can lead
to significant performance gains on ill-conditioned problems.
|
Train Feedfoward Neural Network with Layer-wise Adaptive Rate via
Approximating Back-matching Propagation | Stochastic gradient descent (SGD) has achieved great success in training deep
neural network, where the gradient is computed through back-propagation.
However, the back-propagated values of different layers vary dramatically. This
inconsistence of gradient magnitude across different layers renders
optimization of deep neural network with a single learning rate problematic. We
introduce the back-matching propagation which computes the backward values on
the layer's parameter and the input by matching backward values on the layer's
output. This leads to solving a bunch of least-squares problems, which requires
high computational cost. We then reduce the back-matching propagation with
approximations and propose an algorithm that turns to be the regular SGD with a
layer-wise adaptive learning rate strategy. This allows an easy implementation
of our algorithm in current machine learning frameworks equipped with
auto-differentiation. We apply our algorithm in training modern deep neural
networks and achieve favorable results over SGD.
|
Real-Time Bidding with Multi-Agent Reinforcement Learning in Display
Advertising | Real-time advertising allows advertisers to bid for each impression for a
visiting user. To optimize specific goals such as maximizing revenue and return
on investment (ROI) led by ad placements, advertisers not only need to estimate
the relevance between the ads and user's interests, but most importantly
require a strategic response with respect to other advertisers bidding in the
market. In this paper, we formulate bidding optimization with multi-agent
reinforcement learning. To deal with a large number of advertisers, we propose
a clustering method and assign each cluster with a strategic bidding agent. A
practical Distributed Coordinated Multi-Agent Bidding (DCMAB) has been proposed
and implemented to balance the tradeoff between the competition and cooperation
among advertisers. The empirical study on our industry-scaled real-world data
has demonstrated the effectiveness of our methods. Our results show
cluster-based bidding would largely outperform single-agent and bandit
approaches, and the coordinated bidding achieves better overall objectives than
purely self-interested bidding agents.
|
Learning Representations for Neural Network-Based Classification Using
the Information Bottleneck Principle | In this theory paper, we investigate training deep neural networks (DNNs) for
classification via minimizing the information bottleneck (IB) functional. We
show that the resulting optimization problem suffers from two severe issues:
First, for deterministic DNNs, either the IB functional is infinite for almost
all values of network parameters, making the optimization problem ill-posed, or
it is piecewise constant, hence not admitting gradient-based optimization
methods. Second, the invariance of the IB functional under bijections prevents
it from capturing properties of the learned representation that are desirable
for classification, such as robustness and simplicity. We argue that these
issues are partly resolved for stochastic DNNs, DNNs that include a (hard or
soft) decision rule, or by replacing the IB functional with related, but more
well-behaved cost functions. We conclude that recent successes reported about
training DNNs using the IB framework must be attributed to such solutions. As a
side effect, our results indicate limitations of the IB framework for the
analysis of DNNs. We also note that rather than trying to repair the inherent
problems in the IB functional, a better approach may be to design regularizers
on latent representation enforcing the desired properties directly.
|
L1-Norm Batch Normalization for Efficient Training of Deep Neural
Networks | Batch Normalization (BN) has been proven to be quite effective at
accelerating and improving the training of deep neural networks (DNNs).
However, BN brings additional computation, consumes more memory and generally
slows down the training process by a large margin, which aggravates the
training effort. Furthermore, the nonlinear square and root operations in BN
also impede the low bit-width quantization techniques, which draws much
attention in deep learning hardware community. In this work, we propose an
L1-norm BN (L1BN) with only linear operations in both the forward and the
backward propagations during training. L1BN is shown to be approximately
equivalent to the original L2-norm BN (L2BN) by multiplying a scaling factor.
Experiments on various convolutional neural networks (CNNs) and generative
adversarial networks (GANs) reveal that L1BN maintains almost the same
accuracies and convergence rates compared to L2BN but with higher computational
efficiency. On FPGA platform, the proposed signum and absolute operations in
L1BN can achieve 1.5$\times$ speedup and save 50\% power consumption, compared
with the original costly square and root operations, respectively. This
hardware-friendly normalization method not only surpasses L2BN in speed, but
also simplify the hardware design of ASIC accelerators with higher energy
efficiency. Last but not the least, L1BN promises a fully quantized training of
DNNs, which is crucial to future adaptive terminal devices.
|
Gaussian meta-embeddings for efficient scoring of a heavy-tailed PLDA
model | Embeddings in machine learning are low-dimensional representations of complex
input patterns, with the property that simple geometric operations like
Euclidean distances and dot products can be used for classification and
comparison tasks. The proposed meta-embeddings are special embeddings that live
in more general inner product spaces. They are designed to propagate
uncertainty to the final output in speaker recognition and similar
applications. The familiar Gaussian PLDA model (GPLDA) can be re-formulated as
an extractor for Gaussian meta-embeddings (GMEs), such that likelihood ratio
scores are given by Hilbert space inner products between Gaussian likelihood
functions. GMEs extracted by the GPLDA model have fixed precisions and do not
propagate uncertainty. We show that a generalization to heavy-tailed PLDA gives
GMEs with variable precisions, which do propagate uncertainty. Experiments on
NIST SRE 2010 and 2016 show that the proposed method applied to i-vectors
without length normalization is up to 20% more accurate than GPLDA applied to
length-normalized ivectors.
|
Time-sensitive Customer Churn Prediction based on PU Learning | With the fast development of Internet companies throughout the world,
customer churn has become a serious concern. To better help the companies
retain their customers, it is important to build a customer churn prediction
model to identify the customers who are most likely to churn ahead of time. In
this paper, we propose a Time-sensitive Customer Churn Prediction (TCCP)
framework based on Positive and Unlabeled (PU) learning technique.
Specifically, we obtain the recent data by shortening the observation period,
and start to train model as long as enough positive samples are collected,
ignoring the absence of the negative examples. We conduct thoroughly
experiments on real industry data from Alipay.com. The experimental results
demonstrate that TCCP outperforms the rule-based models and the traditional
supervised learning models.
|
Bioinformatics and Medicine in the Era of Deep Learning | Many of the current scientific advances in the life sciences have their
origin in the intensive use of data for knowledge discovery. In no area this is
so clear as in bioinformatics, led by technological breakthroughs in data
acquisition technologies. It has been argued that bioinformatics could quickly
become the field of research generating the largest data repositories, beating
other data-intensive areas such as high-energy physics or astroinformatics.
Over the last decade, deep learning has become a disruptive advance in machine
learning, giving new live to the long-standing connectionist paradigm in
artificial intelligence. Deep learning methods are ideally suited to
large-scale data and, therefore, they should be ideally suited to knowledge
discovery in bioinformatics and biomedicine at large. In this brief paper, we
review key aspects of the application of deep learning in bioinformatics and
medicine, drawing from the themes covered by the contributions to an ESANN 2018
special session devoted to this topic.
|
Matching Convolutional Neural Networks without Priors about Data | We propose an extension of Convolutional Neural Networks (CNNs) to
graph-structured data, including strided convolutions and data augmentation on
graphs.
Our method matches the accuracy of state-of-the-art CNNs when applied on
images, without any prior about their 2D regular structure.
On fMRI data, we obtain a significant gain in accuracy compared with existing
graph-based alternatives.
|
Coarse to fine non-rigid registration: a chain of scale-specific neural
networks for multimodal image alignment with application to remote sensing | We tackle here the problem of multimodal image non-rigid registration, which
is of prime importance in remote sensing and medical imaging. The difficulties
encountered by classical registration approaches include feature design and
slow optimization by gradient descent. By analyzing these methods, we note the
significance of the notion of scale. We design easy-to-train,
fully-convolutional neural networks able to learn scale-specific features. Once
chained appropriately, they perform global registration in linear time, getting
rid of gradient descent schemes by predicting directly the deformation.We show
their performance in terms of quality and speed through various tasks of remote
sensing multimodal image alignment. In particular, we are able to register
correctly cadastral maps of buildings as well as road polylines onto RGB
images, and outperform current keypoint matching methods.
|
Adversarial Active Learning for Deep Networks: a Margin Based Approach | We propose a new active learning strategy designed for deep neural networks.
The goal is to minimize the number of data annotation queried from an oracle
during training. Previous active learning strategies scalable for deep networks
were mostly based on uncertain sample selection. In this work, we focus on
examples lying close to the decision boundary. Based on theoretical works on
margin theory for active learning, we know that such examples may help to
considerably decrease the number of annotations. While measuring the exact
distance to the decision boundaries is intractable, we propose to rely on
adversarial examples. We do not consider anymore them as a threat instead we
exploit the information they provide on the distribution of the input space in
order to approximate the distance to decision boundaries. We demonstrate
empirically that adversarial active queries yield faster convergence of CNNs
trained on MNIST, the Shoe-Bag and the Quick-Draw datasets.
|
Solving Inverse Computational Imaging Problems using Deep Pixel-level
Prior | Signal reconstruction is a challenging aspect of computational imaging as it
often involves solving ill-posed inverse problems. Recently, deep feed-forward
neural networks have led to state-of-the-art results in solving various inverse
imaging problems. However, being task specific, these networks have to be
learned for each inverse problem. On the other hand, a more flexible approach
would be to learn a deep generative model once and then use it as a signal
prior for solving various inverse problems. We show that among the various
state of the art deep generative models, autoregressive models are especially
suitable for our purpose for the following reasons. First, they explicitly
model the pixel level dependencies and hence are capable of reconstructing
low-level details such as texture patterns and edges better. Second, they
provide an explicit expression for the image prior which can then be used for
MAP based inference along with the forward model. Third, they can model long
range dependencies in images which make them ideal for handling global
multiplexing as encountered in various compressive imaging systems. We
demonstrate the efficacy of our proposed approach in solving three
computational imaging problems: Single Pixel Camera (SPC), LiSens and FlatCam.
For both real and simulated cases, we obtain better reconstructions than the
state-of-the-art methods in terms of perceptual and quantitative metrics.
|
Query-Free Attacks on Industry-Grade Face Recognition Systems under
Resource Constraints | To launch black-box attacks against a Deep Neural Network (DNN) based Face
Recognition (FR) system, one needs to build \textit{substitute} models to
simulate the target model, so the adversarial examples discovered from
substitute models could also mislead the target model. Such
\textit{transferability} is achieved in recent studies through querying the
target model to obtain data for training the substitute models. A real-world
target, likes the FR system of law enforcement, however, is less accessible to
the adversary. To attack such a system, a substitute model with similar quality
as the target model is needed to identify their common defects. This is hard
since the adversary often does not have the enough resources to train such a
powerful model (hundreds of millions of images and rooms of GPUs are needed to
train a commercial FR system).
We found in our research, however, that a resource-constrained adversary
could still effectively approximate the target model's capability to recognize
\textit{specific} individuals, by training \textit{biased} substitute models on
additional images of those victims whose identities the attacker want to cover
or impersonate. This is made possible by a new property we discovered, called
\textit{Nearly Local Linearity} (NLL), which models the observation that an
ideal DNN model produces the image representations (embeddings) whose distances
among themselves truthfully describe the human perception of the differences
among the input images. By simulating this property around the victim's images,
we significantly improve the transferability of black-box impersonation attacks
by nearly 50\%. Particularly, we successfully attacked a commercial system
trained over 20 million images, using 4 million images and 1/5 of the training
time but achieving 62\% transferability in an impersonation attack and 89\% in
a dodging attack.
|
Learning to recognize touch gestures: recurrent vs. convolutional
features and dynamic sampling | We propose a fully automatic method for learning gestures on big touch
devices in a potentially multi-user context. The goal is to learn general
models capable of adapting to different gestures, user styles and hardware
variations (e.g. device sizes, sampling frequencies and regularities).
Based on deep neural networks, our method features a novel dynamic sampling
and temporal normalization component, transforming variable length gestures
into fixed length representations while preserving finger/surface contact
transitions, that is, the topology of the signal. This sequential
representation is then processed with a convolutional model capable, unlike
recurrent networks, of learning hierarchical representations with different
levels of abstraction.
To demonstrate the interest of the proposed method, we introduce a new touch
gestures dataset with 6591 gestures performed by 27 people, which is, up to our
knowledge, the first of its kind: a publicly available multi-touch gesture
dataset for interaction.
We also tested our method on a standard dataset of symbolic touch gesture
recognition, the MMG dataset, outperforming the state of the art and reporting
close to perfect performance.
|
Attention-Based Guided Structured Sparsity of Deep Neural Networks | Network pruning is aimed at imposing sparsity in a neural network
architecture by increasing the portion of zero-valued weights for reducing its
size regarding energy-efficiency consideration and increasing evaluation speed.
In most of the conducted research efforts, the sparsity is enforced for network
pruning without any attention to the internal network characteristics such as
unbalanced outputs of the neurons or more specifically the distribution of the
weights and outputs of the neurons. That may cause severe accuracy drop due to
uncontrolled sparsity. In this work, we propose an attention mechanism that
simultaneously controls the sparsity intensity and supervised network pruning
by keeping important information bottlenecks of the network to be active. On
CIFAR-10, the proposed method outperforms the best baseline method by 6% and
reduced the accuracy drop by 2.6x at the same level of sparsity.
|
Instance Optimal Decoding and the Restricted Isometry Property | In this paper, we address the question of information preservation in
ill-posed, non-linear inverse problems, assuming that the measured data is
close to a low-dimensional model set. We provide necessary and sufficient
conditions for the existence of a so-called instance optimal decoder, i.e.,
that is robust to noise and modelling error. Inspired by existing results in
compressive sensing, our analysis is based on a (Lower) Restricted Isometry
Property (LRIP), formulated in a non-linear fashion. We also provide sufficient
conditions for non-uniform recovery with random measurement operators, with a
new formulation of the LRIP. We finish by describing typical strategies to
prove the LRIP in both linear and non-linear cases, and illustrate our results
by studying the invertibility of a one-layer neural net with random weights.
|
High-Dimensional Vector Semantics | In this paper we explore the "vector semantics" problem from the perspective
of "almost orthogonal" property of high-dimensional random vectors. We show
that this intriguing property can be used to "memorize" random vectors by
simply adding them, and we provide an efficient probabilistic solution to the
set membership problem. Also, we discuss several applications to word context
vector embeddings, document sentences similarity, and spam filtering.
|
Extraction of V2V Encountering Scenarios from Naturalistic Driving
Database | It is necessary to thoroughly evaluate the effectiveness and safety of
Connected Vehicles (CVs) algorithm before their release and deployment. Current
evaluation approach mainly relies on simulation platform with the
single-vehicle driving model. The main drawback of it is the lack of network
realism. To overcome this problem, we extract naturalistic V2V encounters data
from the database, and then separate the primary vehicle encounter category by
clustering. A fast mining algorithm is proposed that can be applied to parallel
query for further process acceleration. 4,500 encounters are mined from a 275
GB database collected in the Safety Pilot Model Program in Ann Arbor Michigan,
USA. K-means and Dynamic Time Warping (DTW) are used in clustering. Results
show this method can quickly mine and cluster primary driving scenarios from a
large database. Our results separate the car-following, intersection and
by-passing, which are the primary category of the vehicle encounter. We
anticipate the work in the essay can become a general method to effectively
extract vehicle encounters from any existing database that contains vehicular
GPS information. What's more, the naturalistic data of different vehicle
encounters can be applied in Connected Vehicles evaluation.
|
VR-SGD: A Simple Stochastic Variance Reduction Method for Machine
Learning | In this paper, we propose a simple variant of the original SVRG, called
variance reduced stochastic gradient descent (VR-SGD). Unlike the choices of
snapshot and starting points in SVRG and its proximal variant, Prox-SVRG, the
two vectors of VR-SGD are set to the average and last iterate of the previous
epoch, respectively. The settings allow us to use much larger learning rates,
and also make our convergence analysis more challenging. We also design two
different update rules for smooth and non-smooth objective functions,
respectively, which means that VR-SGD can tackle non-smooth and/or non-strongly
convex problems directly without any reduction techniques. Moreover, we analyze
the convergence properties of VR-SGD for strongly convex problems, which show
that VR-SGD attains linear convergence. Different from its counterparts that
have no convergence guarantees for non-strongly convex problems, we also
provide the convergence guarantees of VR-SGD for this case, and empirically
verify that VR-SGD with varying learning rates achieves similar performance to
its momentum accelerated variant that has the optimal convergence rate
$\mathcal{O}(1/T^2)$. Finally, we apply VR-SGD to solve various machine
learning problems, such as convex and non-convex empirical risk minimization,
and leading eigenvalue computation. Experimental results show that VR-SGD
converges significantly faster than SVRG and Prox-SVRG, and usually outperforms
state-of-the-art accelerated methods, e.g., Katyusha.
|
Guaranteed Sufficient Decrease for Stochastic Variance Reduced Gradient
Optimization | In this paper, we propose a novel sufficient decrease technique for
stochastic variance reduced gradient descent methods such as SVRG and SAGA. In
order to make sufficient decrease for stochastic optimization, we design a new
sufficient decrease criterion, which yields sufficient decrease versions of
stochastic variance reduction algorithms such as SVRG-SD and SAGA-SD as a
byproduct. We introduce a coefficient to scale current iterate and to satisfy
the sufficient decrease property, which takes the decisions to shrink, expand
or even move in the opposite direction, and then give two specific update rules
of the coefficient for Lasso and ridge regression. Moreover, we analyze the
convergence properties of our algorithms for strongly convex problems, which
show that our algorithms attain linear convergence rates. We also provide the
convergence guarantees of our algorithms for non-strongly convex problems. Our
experimental results further verify that our algorithms achieve significantly
better performance than their counterparts.
|
Demystifying Parallel and Distributed Deep Learning: An In-Depth
Concurrency Analysis | Deep Neural Networks (DNNs) are becoming an important tool in modern
computing applications. Accelerating their training is a major challenge and
techniques range from distributed algorithms to low-level circuit design. In
this survey, we describe the problem from a theoretical perspective, followed
by approaches for its parallelization. We present trends in DNN architectures
and the resulting implications on parallelization strategies. We then review
and model the different types of concurrency in DNNs: from the single operator,
through parallelism in network inference and training, to distributed deep
learning. We discuss asynchronous stochastic optimization, distributed system
architectures, communication schemes, and neural architecture search. Based on
those approaches, we extrapolate potential directions for parallelism in deep
learning.
|
Convolutional Neural Networks for Toxic Comment Classification | Flood of information is produced in a daily basis through the global Internet
usage arising from the on-line interactive communications among users. While
this situation contributes significantly to the quality of human life,
unfortunately it involves enormous dangers, since on-line texts with high
toxicity can cause personal attacks, on-line harassment and bullying behaviors.
This has triggered both industrial and research community in the last few years
while there are several tries to identify an efficient model for on-line toxic
comment prediction. However, these steps are still in their infancy and new
approaches and frameworks are required. On parallel, the data explosion that
appears constantly, makes the construction of new machine learning
computational tools for managing this information, an imperative need.
Thankfully advances in hardware, cloud computing and big data management allow
the development of Deep Learning approaches appearing very promising
performance so far. For text classification in particular the use of
Convolutional Neural Networks (CNN) have recently been proposed approaching
text analytics in a modern manner emphasizing in the structure of words in a
document. In this work, we employ this approach to discover toxic comments in a
large pool of documents provided by a current Kaggle's competition regarding
Wikipedia's talk page edits. To justify this decision we choose to compare CNNs
against the traditional bag-of-words approach for text analysis combined with a
selection of algorithms proven to be very effective in text classification. The
reported results provide enough evidence that CNN enhance toxic comment
classification reinforcing research interest towards this direction.
|
Breaking the $1/\sqrt{n}$ Barrier: Faster Rates for Permutation-based
Models in Polynomial Time | Many applications, including rank aggregation and crowd-labeling, can be
modeled in terms of a bivariate isotonic matrix with unknown permutations
acting on its rows and columns. We consider the problem of estimating such a
matrix based on noisy observations of a subset of its entries, and design and
analyze a polynomial-time algorithm that improves upon the state of the art. In
particular, our results imply that any such $n \times n$ matrix can be
estimated efficiently in the normalized Frobenius norm at rate
$\widetilde{\mathcal O}(n^{-3/4})$, thus narrowing the gap between
$\widetilde{\mathcal O}(n^{-1})$ and $\widetilde{\mathcal O}(n^{-1/2})$, which
were hitherto the rates of the most statistically and computationally efficient
methods, respectively.
|
The Emergence of Spectral Universality in Deep Networks | Recent work has shown that tight concentration of the entire spectrum of
singular values of a deep network's input-output Jacobian around one at
initialization can speed up learning by orders of magnitude. Therefore, to
guide important design choices, it is important to build a full theoretical
understanding of the spectra of Jacobians at initialization. To this end, we
leverage powerful tools from free probability theory to provide a detailed
analytic understanding of how a deep network's Jacobian spectrum depends on
various hyperparameters including the nonlinearity, the weight and bias
distributions, and the depth. For a variety of nonlinearities, our work reveals
the emergence of new universal limiting spectral distributions that remain
concentrated around one even as the depth goes to infinity.
|
Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs | The loss functions of deep neural networks are complex and their geometric
properties are not well understood. We show that the optima of these complex
loss functions are in fact connected by simple curves over which training and
test accuracy are nearly constant. We introduce a training procedure to
discover these high-accuracy pathways between modes. Inspired by this new
geometric insight, we also propose a new ensembling method entitled Fast
Geometric Ensembling (FGE). Using FGE we can train high-performing ensembles in
the time required to train a single model. We achieve improved performance
compared to the recent state-of-the-art Snapshot Ensembles, on CIFAR-10,
CIFAR-100, and ImageNet.
|
The Mirage of Action-Dependent Baselines in Reinforcement Learning | Policy gradient methods are a widely used class of model-free reinforcement
learning algorithms where a state-dependent baseline is used to reduce gradient
estimator variance. Several recent papers extend the baseline to depend on both
the state and action and suggest that this significantly reduces variance and
improves sample efficiency without introducing bias into the gradient
estimates. To better understand this development, we decompose the variance of
the policy gradient estimator and numerically show that learned
state-action-dependent baselines do not in fact reduce variance over a
state-dependent baseline in commonly tested benchmark domains. We confirm this
unexpected result by reviewing the open-source code accompanying these prior
papers, and show that subtle implementation decisions cause deviations from the
methods presented in the papers and explain the source of the previously
observed empirical gains. Furthermore, the variance decomposition highlights
areas for improvement, which we demonstrate by illustrating a simple change to
the typical value function parameterization that can significantly improve
performance.
|
A Mathematical Framework for Deep Learning in Elastic Source Imaging | An inverse elastic source problem with sparse measurements is of concern. A
generic mathematical framework is proposed which incorporates a low-
dimensional manifold regularization in the conventional source reconstruction
algorithms thereby enhancing their performance with sparse datasets. It is
rigorously established that the proposed framework is equivalent to the
so-called \emph{deep convolutional framelet expansion} in machine learning
literature for inverse problems. Apposite numerical examples are furnished to
substantiate the efficacy of the proposed framework.
|
A Differential Privacy Mechanism Design Under Matrix-Valued Query | Traditionally, differential privacy mechanism design has been tailored for a
scalar-valued query function. Although many mechanisms such as the Laplace and
Gaussian mechanisms can be extended to a matrix-valued query function by adding
i.i.d. noise to each element of the matrix, this method is often sub-optimal as
it forfeits an opportunity to exploit the structural characteristics typically
associated with matrix analysis. In this work, we consider the design of
differential privacy mechanism specifically for a matrix-valued query function.
The proposed solution is to utilize a matrix-variate noise, as opposed to the
traditional scalar-valued noise. Particularly, we propose a novel differential
privacy mechanism called the Matrix-Variate Gaussian (MVG) mechanism, which
adds a matrix-valued noise drawn from a matrix-variate Gaussian distribution.
We prove that the MVG mechanism preserves $(\epsilon,\delta)$-differential
privacy, and show that it allows the structural characteristics of the
matrix-valued query function to naturally be exploited. Furthermore, due to the
multi-dimensional nature of the MVG mechanism and the matrix-valued query, we
introduce the concept of directional noise, which can be utilized to mitigate
the impact the noise has on the utility of the query. Finally, we demonstrate
the performance of the MVG mechanism and the advantages of directional noise
using three matrix-valued queries on three privacy-sensitive datasets. We find
that the MVG mechanism notably outperforms four previous state-of-the-art
approaches, and provides comparable utility to the non-private baseline. Our
work thus presents a promising prospect for both future research and
implementation of differential privacy for matrix-valued query functions.
|
Latent-space Physics: Towards Learning the Temporal Evolution of Fluid
Flow | We propose a method for the data-driven inference of temporal evolutions of
physical functions with deep learning. More specifically, we target fluid
flows, i.e. Navier-Stokes problems, and we propose a novel LSTM-based approach
to predict the changes of pressure fields over time. The central challenge in
this context is the high dimensionality of Eulerian space-time data sets. We
demonstrate for the first time that dense 3D+time functions of physics system
can be predicted within the latent spaces of neural networks, and we arrive at
a neural-network based simulation algorithm with significant practical
speed-ups. We highlight the capabilities of our method with a series of complex
liquid simulations, and with a set of single-phase buoyancy simulations. With a
set of trained networks, our method is more than two orders of magnitudes
faster than a traditional pressure solver. Additionally, we present and discuss
a series of detailed evaluations for the different components of our algorithm.
|
Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data | Learning inter-domain mappings from unpaired data can improve performance in
structured prediction tasks, such as image segmentation, by reducing the need
for paired data. CycleGAN was recently proposed for this problem, but
critically assumes the underlying inter-domain mapping is approximately
deterministic and one-to-one. This assumption renders the model ineffective for
tasks requiring flexible, many-to-many mappings. We propose a new model, called
Augmented CycleGAN, which learns many-to-many mappings between domains. We
examine Augmented CycleGAN qualitatively and quantitatively on several image
datasets.
|
ADMM-based Networked Stochastic Variational Inference | Owing to the recent advances in "Big Data" modeling and prediction tasks,
variational Bayesian estimation has gained popularity due to their ability to
provide exact solutions to approximate posteriors. One key technique for
approximate inference is stochastic variational inference (SVI). SVI poses
variational inference as a stochastic optimization problem and solves it
iteratively using noisy gradient estimates. It aims to handle massive data for
predictive and classification tasks by applying complex Bayesian models that
have observed as well as latent variables. This paper aims to decentralize it
allowing parallel computation, secure learning and robustness benefits. We use
Alternating Direction Method of Multipliers in a top-down setting to develop a
distributed SVI algorithm such that independent learners running inference
algorithms only require sharing the estimated model parameters instead of their
private datasets. Our work extends the distributed SVI-ADMM algorithm that we
first propose, to an ADMM-based networked SVI algorithm in which not only are
the learners working distributively but they share information according to
rules of a graph by which they form a network. This kind of work lies under the
umbrella of `deep learning over networks' and we verify our algorithm for a
topic-modeling problem for corpus of Wikipedia articles. We illustrate the
results on latent Dirichlet allocation (LDA) topic model in large document
classification, compare performance with the centralized algorithm, and use
numerical experiments to corroborate the analytical results.
|
Tell Me Where to Look: Guided Attention Inference Network | Weakly supervised learning with only coarse labels can obtain visual
explanations of deep neural network such as attention maps by back-propagating
gradients. These attention maps are then available as priors for tasks such as
object localization and semantic segmentation. In one common framework we
address three shortcomings of previous approaches in modeling such attention
maps: We (1) first time make attention maps an explicit and natural component
of the end-to-end training, (2) provide self-guidance directly on these maps by
exploring supervision form the network itself to improve them, and (3)
seamlessly bridge the gap between using weak and extra supervision if
available. Despite its simplicity, experiments on the semantic segmentation
task demonstrate the effectiveness of our methods. We clearly surpass the
state-of-the-art on Pascal VOC 2012 val. and test set. Besides, the proposed
framework provides a way not only explaining the focus of the learner but also
feeding back with direct guidance towards specific tasks. Under mild
assumptions our method can also be understood as a plug-in to existing weakly
supervised learners to improve their generalization performance.
|
Semi-Supervised Learning Enabled by Multiscale Deep Neural Network
Inversion | Deep Neural Networks (DNNs) provide state-of-the-art solutions in several
difficult machine perceptual tasks. However, their performance relies on the
availability of a large set of labeled training data, which limits the breadth
of their applicability. Hence, there is a need for new {\em semi-supervised
learning} methods for DNNs that can leverage both (a small amount of) labeled
and unlabeled training data. In this paper, we develop a general loss function
enabling DNNs of any topology to be trained in a semi-supervised manner without
extra hyper-parameters. As opposed to current semi-supervised techniques based
on topology-specific or unstable approaches, ours is both robust and general.
We demonstrate that our approach reaches state-of-the-art performance on the
SVHN ($9.82\%$ test error, with $500$ labels and wide Resnet) and CIFAR10
(16.38% test error, with 8000 labels and sigmoid convolutional neural network)
data sets.
|
Mirrored Langevin Dynamics | We consider the problem of sampling from constrained distributions, which has
posed significant challenges to both non-asymptotic analysis and algorithmic
design. We propose a unified framework, which is inspired by the classical
mirror descent, to derive novel first-order sampling schemes. We prove that,
for a general target distribution with strongly convex potential, our framework
implies the existence of a first-order algorithm achieving
$\tilde{O}(\epsilon^{-2}d)$ convergence, suggesting that the state-of-the-art
$\tilde{O}(\epsilon^{-6}d^5)$ can be vastly improved. With the important Latent
Dirichlet Allocation (LDA) application in mind, we specialize our algorithm to
sample from Dirichlet posteriors, and derive the first non-asymptotic
$\tilde{O}(\epsilon^{-2}d^2)$ rate for first-order sampling. We further extend
our framework to the mini-batch setting and prove convergence rates when only
stochastic gradients are available. Finally, we report promising experimental
results for LDA on real datasets.
|
Behavioral Learning of Aircraft Landing Sequencing Using a Society of
Probabilistic Finite State Machines | Air Traffic Control (ATC) is a complex safety critical environment. A tower
controller would be making many decisions in real-time to sequence aircraft.
While some optimization tools exist to help the controller in some airports,
even in these situations, the real sequence of the aircraft adopted by the
controller is significantly different from the one proposed by the optimization
algorithm. This is due to the very dynamic nature of the environment. The
objective of this paper is to test the hypothesis that one can learn from the
sequence adopted by the controller some strategies that can act as heuristics
in decision support tools for aircraft sequencing. This aim is tested in this
paper by attempting to learn sequences generated from a well-known sequencing
method that is being used in the real world. The approach relies on a genetic
algorithm (GA) to learn these sequences using a society Probabilistic
Finite-state Machines (PFSMs). Each PFSM learns a different sub-space; thus,
decomposing the learning problem into a group of agents that need to work
together to learn the overall problem. Three sequence metrics (Levenshtein,
Hamming and Position distances) are compared as the fitness functions in GA. As
the results suggest, it is possible to learn the behavior of the
algorithm/heuristic that generated the original sequence from very limited
information.
|
Networking the Boids is More Robust Against Adversarial Learning | Swarm behavior using Boids-like models has been studied primarily using
close-proximity spatial sensory information (e.g. vision range). In this study,
we propose a novel approach in which the classic definition of
boids\textquoteright \ neighborhood that relies on sensory perception and
Euclidian space locality is replaced with graph-theoretic network-based
proximity mimicking communication and social networks. We demonstrate that
networking the boids leads to faster swarming and higher quality of the
formation. We further investigate the effect of adversarial learning, whereby
an observer attempts to reverse engineer the dynamics of the swarm through
observing its behavior. The results show that networking the swarm demonstrated
a more robust approach against adversarial learning than a local-proximity
neighborhood structure.
|
Cluster Naturalistic Driving Encounters Using Deep Unsupervised Learning | Learning knowledge from driving encounters could help self-driving cars make
appropriate decisions when driving in complex settings with nearby vehicles
engaged. This paper develops an unsupervised classifier to group naturalistic
driving encounters into distinguishable clusters by combining an auto-encoder
with k-means clustering (AE-kMC). The effectiveness of AE-kMC was validated
using the data of 10,000 naturalistic driving encounters which were collected
by the University of Michigan, Ann Arbor in the past five years. We compare our
developed method with the $k$-means clustering methods and experimental results
demonstrate that the AE-kMC method outperforms the original k-means clustering
method.
|
Var-CNN: A Data-Efficient Website Fingerprinting Attack Based on Deep
Learning | In recent years, there have been several works that use website
fingerprinting techniques to enable a local adversary to determine which
website a Tor user visits. While the current state-of-the-art attack, which
uses deep learning, outperforms prior art with medium to large amounts of data,
it attains marginal to no accuracy improvements when both use small amounts of
training data. In this work, we propose Var-CNN, a website fingerprinting
attack that leverages deep learning techniques along with novel insights
specific to packet sequence classification. In open-world settings with large
amounts of data, Var-CNN attains over $1\%$ higher true positive rate (TPR)
than state-of-the-art attacks while achieving $4\times$ lower false positive
rate (FPR). Var-CNN's improvements are especially notable in low-data
scenarios, where it reduces the FPR of prior art by $3.12\%$ while increasing
the TPR by $13\%$. Overall, insights used to develop Var-CNN can be applied to
future deep learning based attacks, and substantially reduce the amount of
training data needed to perform a successful website fingerprinting attack.
This shortens the time needed for data collection and lowers the likelihood of
having data staleness issues.
|
Investigating Human Priors for Playing Video Games | What makes humans so good at solving seemingly complex video games? Unlike
computers, humans bring in a great deal of prior knowledge about the world,
enabling efficient decision making. This paper investigates the role of human
priors for solving video games. Given a sample game, we conduct a series of
ablation studies to quantify the importance of various priors on human
performance. We do this by modifying the video game environment to
systematically mask different types of visual information that could be used by
humans as priors. We find that removal of some prior knowledge causes a drastic
degradation in the speed with which human players solve the game, e.g. from 2
minutes to over 20 minutes. Furthermore, our results indicate that general
priors, such as the importance of objects and visual consistency, are critical
for efficient game-play. Videos and the game manipulations are available at
https://rach0012.github.io/humanRL_website/
|
Parametrized Accelerated Methods Free of Condition Number | Analyses of accelerated (momentum-based) gradient descent usually assume
bounded condition number to obtain exponential convergence rates. However, in
many real problems, e.g., kernel methods or deep neural networks, the condition
number, even locally, can be unbounded, unknown or mis-estimated. This poses
problems in both implementing and analyzing accelerated algorithms. In this
paper, we address this issue by proposing parametrized accelerated methods by
considering the condition number as a free parameter. We provide spectral-level
analysis for several important accelerated algorithms, obtain explicit
expressions and improve worst case convergence rates. Moreover, we show that
those algorithm converge exponentially even when the condition number is
unknown or mis-estimated.
|
An EMG Gesture Recognition System with Flexible High-Density Sensors and
Brain-Inspired High-Dimensional Classifier | EMG-based gesture recognition shows promise for human-machine interaction.
Systems are often afflicted by signal and electrode variability which degrades
performance over time. We present an end-to-end system combating this
variability using a large-area, high-density sensor array and a robust
classification algorithm. EMG electrodes are fabricated on a flexible substrate
and interfaced to a custom wireless device for 64-channel signal acquisition
and streaming. We use brain-inspired high-dimensional (HD) computing for
processing EMG features in one-shot learning. The HD algorithm is tolerant to
noise and electrode misplacement and can quickly learn from few gestures
without gradient descent or back-propagation. We achieve an average
classification accuracy of 96.64% for five gestures, with only 7% degradation
when training and testing across different days. Our system maintains this
accuracy when trained with only three trials of gestures; it also demonstrates
comparable accuracy with the state-of-the-art when trained with one trial.
|
DeepSOFA: A Continuous Acuity Score for Critically Ill Patients using
Clinically Interpretable Deep Learning | Traditional methods for assessing illness severity and predicting in-hospital
mortality among critically ill patients require time-consuming, error-prone
calculations using static variable thresholds. These methods do not capitalize
on the emerging availability of streaming electronic health record data or
capture time-sensitive individual physiological patterns, a critical task in
the intensive care unit. We propose a novel acuity score framework (DeepSOFA)
that leverages temporal measurements and interpretable deep learning models to
assess illness severity at any point during an ICU stay. We compare DeepSOFA
with SOFA (Sequential Organ Failure Assessment) baseline models using the same
model inputs and find that at any point during an ICU admission, DeepSOFA
yields significantly more accurate predictions of in-hospital mortality. A
DeepSOFA model developed in a public database and validated in a single
institutional cohort had a mean AUC for the entire ICU stay of 0.90 (95% CI
0.90-0.91) compared with baseline SOFA models with mean AUC 0.79 (95% CI
0.79-0.80) and 0.85 (95% CI 0.85-0.86). Deep models are well-suited to identify
ICU patients in need of life-saving interventions prior to the occurrence of an
unexpected adverse event and inform shared decision-making processes among
patients, providers, and families regarding goals of care and optimal resource
utilization.
|
Deep Reinforcement Learning for Vision-Based Robotic Grasping: A
Simulated Comparative Evaluation of Off-Policy Methods | In this paper, we explore deep reinforcement learning algorithms for
vision-based robotic grasping. Model-free deep reinforcement learning (RL) has
been successfully applied to a range of challenging environments, but the
proliferation of algorithms makes it difficult to discern which particular
approach would be best suited for a rich, diverse task like grasping. To answer
this question, we propose a simulated benchmark for robotic grasping that
emphasizes off-policy learning and generalization to unseen objects. Off-policy
learning enables utilization of grasping data over a wide variety of objects,
and diversity is important to enable the method to generalize to new objects
that were not seen during training. We evaluate the benchmark tasks against a
variety of Q-function estimation methods, a method previously proposed for
robotic grasping with deep neural network models, and a novel approach based on
a combination of Monte Carlo return estimation and an off-policy correction.
Our results indicate that several simple methods provide a surprisingly strong
competitor to popular algorithms such as double Q-learning, and our analysis of
stability sheds light on the relative tradeoffs between the algorithms.
|
Solving for high dimensional committor functions using artificial neural
networks | In this note we propose a method based on artificial neural network to study
the transition between states governed by stochastic processes. In particular,
we aim for numerical schemes for the committor function, the central object of
transition path theory, which satisfies a high-dimensional Fokker-Planck
equation. By working with the variational formulation of such partial
differential equation and parameterizing the committor function in terms of a
neural network, approximations can be obtained via optimizing the neural
network weights using stochastic algorithms. The numerical examples show that
moderate accuracy can be achieved for high-dimensional problems.
|
Escoin: Efficient Sparse Convolutional Neural Network Inference on GPUs | Deep neural networks have achieved remarkable accuracy in many artificial
intelligence applications, e.g. computer vision, at the cost of a large number
of parameters and high computational complexity. Weight pruning can compress
DNN models by removing redundant parameters in the networks, but it brings
sparsity in the weight matrix, and therefore makes the computation inefficient
on GPUs. Although pruning can remove more than 80% of the weights, it actually
hurts inference performance (speed) when running models on GPUs.
Two major problems cause this unsatisfactory performance on GPUs. First,
lowering convolution onto matrix multiplication reduces data reuse
opportunities and wastes memory bandwidth. Second, the sparsity brought by
pruning makes the computation irregular, which leads to inefficiency when
running on massively parallel GPUs. To overcome these two limitations, we
propose Escort, an efficient sparse convolutional neural networks on GPUs.
Instead of using the lowering method, we choose to compute the sparse
convolutions directly. We then orchestrate the parallelism and locality for the
direct sparse convolution kernel, and apply customized optimization techniques
to further improve performance. Evaluation on NVIDIA GPUs show that Escort can
improve sparse convolution speed by 2.63x and 3.07x, and inference speed by
1.43x and 1.69x, compared to CUBLAS and CUSPARSE respectively.
|
Relational Neural Expectation Maximization: Unsupervised Discovery of
Objects and their Interactions | Common-sense physical reasoning is an essential ingredient for any
intelligent agent operating in the real-world. For example, it can be used to
simulate the environment, or to infer the state of parts of the world that are
currently unobserved. In order to match real-world conditions this causal
knowledge must be learned without access to supervised data. To address this
problem we present a novel method that learns to discover objects and model
their physical interactions from raw visual images in a purely
\emph{unsupervised} fashion. It incorporates prior knowledge about the
compositional nature of human perception to factor interactions between
object-pairs and learn efficiently. On videos of bouncing balls we show the
superior modelling capabilities of our method compared to other unsupervised
neural approaches that do not incorporate such prior knowledge. We demonstrate
its ability to handle occlusion and show that it can extrapolate learned
knowledge to scenes with different numbers of objects.
|
A Bayesian Model for Activities Recommendation and Event Structure
Optimization Using Visitors Tracking | In events that are composed by many activities, there is a problem that
involves retrieve and management the information of visitors that are visiting
the activities. This management is crucial to find some activities that are
drawing attention of visitors; identify an ideal positioning for activities;
which path is more frequented by visitors. In this work, these features are
studied using Complex Network theory. For the beginning, an artificial database
was generated to study the mentioned features. Secondly, this work shows a
method to optimize the event structure that is better than a random method and
a recommendation system that achieves ~95% of accuracy.
|
Tensor Decomposition for Compressing Recurrent Neural Network | In the machine learning fields, Recurrent Neural Network (RNN) has become a
popular architecture for sequential data modeling. However, behind the
impressive performance, RNNs require a large number of parameters for both
training and inference. In this paper, we are trying to reduce the number of
parameters and maintain the expressive power from RNN simultaneously. We
utilize several tensor decompositions method including CANDECOMP/PARAFAC (CP),
Tucker decomposition and Tensor Train (TT) to re-parameterize the Gated
Recurrent Unit (GRU) RNN. We evaluate all tensor-based RNNs performance on
sequence modeling tasks with a various number of parameters. Based on our
experiment results, TT-GRU achieved the best results in a various number of
parameters compared to other decomposition methods.
|
Precision medicine as a control problem: Using simulation and deep
reinforcement learning to discover adaptive, personalized multi-cytokine
therapy for sepsis | Sepsis is a life-threatening condition affecting one million people per year
in the US in which dysregulation of the body's own immune system causes damage
to its tissues, resulting in a 28 - 50% mortality rate. Clinical trials for
sepsis treatment over the last 20 years have failed to produce a single
currently FDA approved drug treatment. In this study, we attempt to discover an
effective cytokine mediation treatment strategy for sepsis using a previously
developed agent-based model that simulates the innate immune response to
infection: the Innate Immune Response agent-based model (IIRABM). Previous
attempts at reducing mortality with multi-cytokine mediation using the IIRABM
have failed to reduce mortality across all patient parameterizations and
motivated us to investigate whether adaptive, personalized multi-cytokine
mediation can control the trajectory of sepsis and lower patient mortality. We
used the IIRABM to compute a treatment policy in which systemic patient
measurements are used in a feedback loop to inform future treatment. Using deep
reinforcement learning, we identified a policy that achieves 0% mortality on
the patient parameterization on which it was trained. More importantly, this
policy also achieves 0.8% mortality over 500 randomly selected patient
parameterizations with baseline mortalities ranging from 1 - 99% (with an
average of 49%) spanning the entire clinically plausible parameter space of the
IIRABM. These results suggest that adaptive, personalized multi-cytokine
mediation therapy could be a promising approach for treating sepsis. We hope
that this work motivates researchers to consider such an approach as part of
future clinical trials. To the best of our knowledge, this work is the first to
consider adaptive, personalized multi-cytokine mediation therapy for sepsis,
and is the first to exploit deep reinforcement learning on a biological
simulation.
|
A High GOPs/Slice Time Series Classifier for Portable and Embedded
Biomedical Applications | Nowadays a diverse range of physiological data can be captured continuously
for various applications in particular wellbeing and healthcare. Such data
require efficient methods for classification and analysis. Deep learning
algorithms have shown remarkable potential regarding such analyses, however,
the use of these algorithms on low-power wearable devices is challenged by
resource constraints such as area and power consumption. Most of the available
on-chip deep learning processors contain complex and dense hardware
architectures in order to achieve the highest possible throughput. Such a trend
in hardware design may not be efficient in applications where on-node
computation is required and the focus is more on the area and power efficiency
as in the case of portable and embedded biomedical devices. This paper presents
an efficient time-series classifier capable of automatically detecting
effective features and classifying the input signals in real-time. In the
proposed classifier, throughput is traded off with hardware complexity and cost
using resource sharing techniques. A Convolutional Neural Network (CNN) is
employed to extract input features and then a Long-Short-Term-Memory (LSTM)
architecture with ternary weight precision classifies the input signals
according to the extracted features. Hardware implementation on a Xilinx FPGA
confirm that the proposed hardware can accurately classify multiple complex
biomedical time series data with low area and power consumption and outperform
all previously presented state-of-the-art records. Most notably, our classifier
reaches 1.3$\times$ higher GOPs/Slice than similar state of the art FPGA-based
accelerators.
|
DiGrad: Multi-Task Reinforcement Learning with Shared Actions | Most reinforcement learning algorithms are inefficient for learning multiple
tasks in complex robotic systems, where different tasks share a set of actions.
In such environments a compound policy may be learnt with shared neural network
parameters, which performs multiple tasks concurrently. However such compound
policy may get biased towards a task or the gradients from different tasks
negate each other, making the learning unstable and sometimes less data
efficient. In this paper, we propose a new approach for simultaneous training
of multiple tasks sharing a set of common actions in continuous action spaces,
which we call as DiGrad (Differential Policy Gradient). The proposed framework
is based on differential policy gradients and can accommodate multi-task
learning in a single actor-critic network. We also propose a simple heuristic
in the differential policy gradient update to further improve the learning. The
proposed architecture was tested on 8 link planar manipulator and 27 degrees of
freedom(DoF) Humanoid for learning multi-goal reachability tasks for 3 and 2
end effectors respectively. We show that our approach supports efficient
multi-task learning in complex robotic systems, outperforming related methods
in continuous action spaces.
|
As you like it: Localization via paired comparisons | Suppose that we wish to estimate a vector $\mathbf{x}$ from a set of binary
paired comparisons of the form "$\mathbf{x}$ is closer to $\mathbf{p}$ than to
$\mathbf{q}$" for various choices of vectors $\mathbf{p}$ and $\mathbf{q}$. The
problem of estimating $\mathbf{x}$ from this type of observation arises in a
variety of contexts, including nonmetric multidimensional scaling, "unfolding,"
and ranking problems, often because it provides a powerful and flexible model
of preference. We describe theoretical bounds for how well we can expect to
estimate $\mathbf{x}$ under a randomized model for $\mathbf{p}$ and
$\mathbf{q}$. We also present results for the case where the comparisons are
noisy and subject to some degree of error. Additionally, we show that under a
randomized model for $\mathbf{p}$ and $\mathbf{q}$, a suitable number of binary
paired comparisons yield a stable embedding of the space of target vectors.
Finally, we also show that we can achieve significant gains by adaptively
changing the distribution for choosing $\mathbf{p}$ and $\mathbf{q}$.
|
Learning Discriminative Multilevel Structured Dictionaries for
Supervised Image Classification | Sparse representations using overcomplete dictionaries have proved to be a
powerful tool in many signal processing applications such as denoising,
super-resolution, inpainting, compression or classification. The sparsity of
the representation very much depends on how well the dictionary is adapted to
the data at hand. In this paper, we propose a method for learning structured
multilevel dictionaries with discriminative constraints to make them well
suited for the supervised pixelwise classification of images. A multilevel
tree-structured discriminative dictionary is learnt for each class, with a
learning objective concerning the reconstruction errors of the image patches
around the pixels over each class-representative dictionary. After the initial
assignment of the class labels to image pixels based on their sparse
representations over the learnt dictionaries, the final classification is
achieved by smoothing the label image with a graph cut method and an erosion
method. Applied to a common set of texture images, our supervised
classification method shows competitive results with the state of the art.
|
Predictive Uncertainty Estimation via Prior Networks | Estimating how uncertain an AI system is in its predictions is important to
improve the safety of such systems. Uncertainty in predictive can result from
uncertainty in model parameters, irreducible data uncertainty and uncertainty
due to distributional mismatch between the test and training data
distributions. Different actions might be taken depending on the source of the
uncertainty so it is important to be able to distinguish between them.
Recently, baseline tasks and metrics have been defined and several practical
methods to estimate uncertainty developed. These methods, however, attempt to
model uncertainty due to distributional mismatch either implicitly through
model uncertainty or as data uncertainty. This work proposes a new framework
for modeling predictive uncertainty called Prior Networks (PNs) which
explicitly models distributional uncertainty. PNs do this by parameterizing a
prior distribution over predictive distributions. This work focuses on
uncertainty for classification and evaluates PNs on the tasks of identifying
out-of-distribution (OOD) samples and detecting misclassification on the MNIST
dataset, where they are found to outperform previous methods. Experiments on
synthetic and MNIST and CIFAR-10 data show that unlike previous non-Bayesian
methods PNs are able to distinguish between data and distributional
uncertainty.
|
Stochastic Dynamic Programming Heuristics for Influence
Maximization-Revenue Optimization | The well-known Influence Maximization (IM) problem has been actively studied
by researchers over the past decade, with emphasis on marketing and social
networks. Existing research have obtained solutions to the IM problem by
obtaining the influence spread and utilizing the property of submodularity.
This paper is based on a novel approach to the IM problem geared towards
optimizing clicks and consequently revenue within anOnline Social Network
(OSN). Our approach diverts from existing approaches by adopting a novel,
decision-making perspective through implementing Stochastic Dynamic Programming
(SDP). Thus, we define a new problem Influence Maximization-Revenue
Optimization (IM-RO) and propose SDP as a method in which this problem can be
solved. The SDP method has lucrative gains for an advertiser in terms of
optimizing clicks and generating revenue however, one drawback to the method is
its associated "curse of dimensionality" particularly for problems involving a
large state space. Thus, we introduce the Lawrence Degree Heuristic (LDH),
Adaptive Hill-Climbing (AHC) and Multistage Particle Swarm Optimization (MPSO)
heuristics as methods which are orders of magnitude faster than the SDP method
whilst achieving near-optimal results. Through a comparative analysis on
various synthetic and real-world networks we present the AHC and LDH as
heuristics well suited to to the IM-RO problem in terms of their accuracy,
running times and scalability under ideal model parameters. In this paper we
present a compelling survey on the SDP method as a practical and lucrative
method for spreading information and optimizing revenue within the context of
OSNs.
|
Maximum likelihood estimation of a finite mixture of logistic regression
models in a continuous data stream | In marketing we are often confronted with a continuous stream of responses to
marketing messages. Such streaming data provide invaluable information
regarding message effectiveness and segmentation. However, streaming data are
hard to analyze using conventional methods: their high volume and the fact that
they are continuously augmented means that it takes considerable time to
analyze them. We propose a method for estimating a finite mixture of logistic
regression models which can be used to cluster customers based on a continuous
stream of responses. This method, which we coin oFMLR, allows segments to be
identified in data streams or extremely large static datasets. Contrary to
black box algorithms, oFMLR provides model estimates that are directly
interpretable. We first introduce oFMLR, explaining in passing general topics
such as online estimation and the EM algorithm, making this paper a high level
overview of possible methods of dealing with large data streams in marketing
practice. Next, we discuss model convergence, identifiability, and relations to
alternative, Bayesian, methods; we also identify more general issues that arise
from dealing with continuously augmented data sets. Finally, we introduce the
oFMLR [R] package and evaluate the method by numerical simulation and by
analyzing a large customer clickstream dataset.
|
Memory-based Parameter Adaptation | Deep neural networks have excelled on a wide range of problems, from vision
to language and game playing. Neural networks very gradually incorporate
information into weights as they process data, requiring very low learning
rates. If the training distribution shifts, the network is slow to adapt, and
when it does adapt, it typically performs badly on the training distribution
before the shift. Our method, Memory-based Parameter Adaptation, stores
examples in memory and then uses a context-based lookup to directly modify the
weights of a neural network. Much higher learning rates can be used for this
local adaptation, reneging the need for many iterations over similar data
before good predictions can be made. As our method is memory-based, it
alleviates several shortcomings of neural networks, such as catastrophic
forgetting, fast, stable acquisition of new knowledge, learning with an
imbalanced class labels, and fast learning during evaluation. We demonstrate
this on a range of supervised tasks: large-scale image classification and
language modelling.
|
Computational Theories of Curiosity-Driven Learning | What are the functions of curiosity? What are the mechanisms of
curiosity-driven learning? We approach these questions about the living using
concepts and tools from machine learning and developmental robotics. We argue
that curiosity-driven learning enables organisms to make discoveries to solve
complex problems with rare or deceptive rewards. By fostering exploration and
discovery of a diversity of behavioural skills, and ignoring these rewards,
curiosity can be efficient to bootstrap learning when there is no information,
or deceptive information, about local improvement towards these problems. We
also explain the key role of curiosity for efficient learning of world models.
We review both normative and heuristic computational frameworks used to
understand the mechanisms of curiosity in humans, conceptualizing the child as
a sense-making organism. These frameworks enable us to discuss the
bi-directional causal links between curiosity and learning, and to provide new
hypotheses about the fundamental role of curiosity in self-organizing
developmental structures through curriculum learning. We present various
developmental robotics experiments that study these mechanisms in action, both
supporting these hypotheses to understand better curiosity in humans and
opening new research avenues in machine learning and artificial intelligence.
Finally, we discuss challenges for the design of experimental paradigms for
studying curiosity in psychology and cognitive neuroscience.
Keywords: Curiosity, intrinsic motivation, lifelong learning, predictions,
world model, rewards, free-energy principle, learning progress, machine
learning, AI, developmental robotics, development, curriculum learning,
self-organization.
|
Using Deep Learning for Segmentation and Counting within Microscopy Data | Cell counting is a ubiquitous, yet tedious task that would greatly benefit
from automation. From basic biological questions to clinical trials, cell
counts provide key quantitative feedback that drive research. Unfortunately,
cell counting is most commonly a manual task and can be time-intensive. The
task is made even more difficult due to overlapping cells, existence of
multiple focal planes, and poor imaging quality, among other factors. Here, we
describe a convolutional neural network approach, using a recently described
feature pyramid network combined with a VGG-style neural network, for
segmenting and subsequent counting of cells in a given microscopy image.
|
Automatic topography of high-dimensional data sets by non-parametric
Density Peak clustering | Data analysis in high-dimensional spaces aims at obtaining a synthetic
description of a data set, revealing its main structure and its salient
features. We here introduce an approach providing this description in the form
of a topography of the data, namely a human-readable chart of the probability
density from which the data are harvested. The approach is based on an
unsupervised extension of Density Peak clustering and a non-parametric density
estimator that measures the probability density in the manifold containing the
data. This allows finding automatically the number and the height of the peaks
of the probability density, and the depth of the "valleys" separating them.
Importantly, the density estimator provides a measure of the error, which
allows distinguishing genuine density peaks from density fluctuations due to
finite sampling. The approach thus provides robust and visual information about
the density peaks' height, their statistical reliability, and their
hierarchical organization, offering a conceptually powerful extension of the
standard clustering partitions. We show that this framework is particularly
useful in the analysis of complex data sets.
|
A Variational Inequality Perspective on Generative Adversarial Networks | Generative adversarial networks (GANs) form a generative modeling approach
known for producing appealing samples, but they are notably difficult to train.
One common way to tackle this issue has been to propose new formulations of the
GAN objective. Yet, surprisingly few studies have looked at optimization
methods designed for this adversarial training. In this work, we cast GAN
optimization problems in the general variational inequality framework. Tapping
into the mathematical programming literature, we counter some common
misconceptions about the difficulties of saddle point optimization and propose
to extend techniques designed for variational inequalities to the training of
GANs. We apply averaging, extrapolation and a computationally cheaper variant
that we call extrapolation from the past to the stochastic gradient method
(SGD) and Adam.
|
Exactly Robust Kernel Principal Component Analysis | Robust principal component analysis (RPCA) can recover low-rank matrices when
they are corrupted by sparse noises. In practice, many matrices are, however,
of high-rank and hence cannot be recovered by RPCA. We propose a novel method
called robust kernel principal component analysis (RKPCA) to decompose a
partially corrupted matrix as a sparse matrix plus a high or full-rank matrix
with low latent dimensionality. RKPCA can be applied to many problems such as
noise removal and subspace clustering and is still the only unsupervised
nonlinear method robust to sparse noises. Our theoretical analysis shows that,
with high probability, RKPCA can provide high recovery accuracy. The
optimization of RKPCA involves nonconvex and indifferentiable problems. We
propose two nonconvex optimization algorithms for RKPCA. They are alternating
direction method of multipliers with backtracking line search and proximal
linearized minimization with adaptive step size. Comparative studies in noise
removal and robust subspace clustering corroborate the effectiveness and
superiority of RKPCA.
|
Learning by Playing - Solving Sparse Reward Tasks from Scratch | We propose Scheduled Auxiliary Control (SAC-X), a new learning paradigm in
the context of Reinforcement Learning (RL). SAC-X enables learning of complex
behaviors - from scratch - in the presence of multiple sparse reward signals.
To this end, the agent is equipped with a set of general auxiliary tasks, that
it attempts to learn simultaneously via off-policy RL. The key idea behind our
method is that active (learned) scheduling and execution of auxiliary policies
allows the agent to efficiently explore its environment - enabling it to excel
at sparse reward RL. Our experiments in several challenging robotic
manipulation settings demonstrate the power of our approach.
|
Near-Optimal Sample Complexity Bounds for Maximum Likelihood Estimation
of Multivariate Log-concave Densities | We study the problem of learning multivariate log-concave densities with
respect to a global loss function. We obtain the first upper bound on the
sample complexity of the maximum likelihood estimator (MLE) for a log-concave
density on $\mathbb{R}^d$, for all $d \geq 4$. Prior to this work, no finite
sample upper bound was known for this estimator in more than $3$ dimensions.
In more detail, we prove that for any $d \geq 1$ and $\epsilon>0$, given
$\tilde{O}_d((1/\epsilon)^{(d+3)/2})$ samples drawn from an unknown log-concave
density $f_0$ on $\mathbb{R}^d$, the MLE outputs a hypothesis $h$ that with
high probability is $\epsilon$-close to $f_0$, in squared Hellinger loss. A
sample complexity lower bound of $\Omega_d((1/\epsilon)^{(d+1)/2})$ was
previously known for any learning algorithm that achieves this guarantee. We
thus establish that the sample complexity of the log-concave MLE is
near-optimal, up to an $\tilde{O}(1/\epsilon)$ factor.
|
Modeling Activity Tracker Data Using Deep Boltzmann Machines | Commercial activity trackers are set to become an essential tool in health
research, due to increasing availability in the general population. The
corresponding vast amounts of mostly unlabeled data pose a challenge to
statistical modeling approaches. To investigate the feasibility of deep
learning approaches for unsupervised learning with such data, we examine weekly
usage patterns of Fitbit activity trackers with deep Boltzmann machines (DBMs).
This method is particularly suitable for modeling complex joint distributions
via latent variables. We also chose this specific procedure because it is a
generative approach, i.e., artificial samples can be generated to explore the
learned structure. We describe how the data can be preprocessed to be
compatible with binary DBMs. The results reveal two distinct usage patterns in
which one group frequently uses trackers on Mondays and Tuesdays, whereas the
other uses trackers during the entire week. This exemplary result shows that
DBMs are feasible and can be useful for modeling activity tracker data.
|
Model-Ensemble Trust-Region Policy Optimization | Model-free reinforcement learning (RL) methods are succeeding in a growing
number of tasks, aided by recent advances in deep learning. However, they tend
to suffer from high sample complexity, which hinders their use in real-world
domains. Alternatively, model-based reinforcement learning promises to reduce
sample complexity, but tends to require careful tuning and to date have
succeeded mainly in restrictive domains where simple models are sufficient for
learning. In this paper, we analyze the behavior of vanilla model-based
reinforcement learning methods when deep neural networks are used to learn both
the model and the policy, and show that the learned policy tends to exploit
regions where insufficient data is available for the model to be learned,
causing instability in training. To overcome this issue, we propose to use an
ensemble of models to maintain the model uncertainty and regularize the
learning process. We further show that the use of likelihood ratio derivatives
yields much more stable learning than backpropagation through time. Altogether,
our approach Model-Ensemble Trust-Region Policy Optimization (ME-TRPO)
significantly reduces the sample complexity compared to model-free deep RL
methods on challenging continuous control benchmark tasks.
|
The Alpha-Beta-Symetric Divergence and their Positive Definite Kernel | In this article we study the field of Hilbertian metrics and positive definit
(pd) kernels on probability measures, they have a real interest in kernel
methods. Firstly we will make a study based on the Alpha-Beta-divergence to
have a Hilbercan metric by proposing an improvement of this divergence by
constructing it so that its is symmetrical the Alpha-Beta-Symmetric-divergence
(ABS-divergence) and also do some studies on these properties but also propose
the kernels associated with this divergence. Secondly we will do mumerical
studies incorporating all proposed metrics/kernels into support vector machine
(SVM). Finally we presented a algorithm for image classification by using our
divergence.
|
INSPECTRE: Privately Estimating the Unseen | We develop differentially private methods for estimating various
distributional properties. Given a sample from a discrete distribution $p$,
some functional $f$, and accuracy and privacy parameters $\alpha$ and
$\varepsilon$, the goal is to estimate $f(p)$ up to accuracy $\alpha$, while
maintaining $\varepsilon$-differential privacy of the sample.
We prove almost-tight bounds on the sample size required for this problem for
several functionals of interest, including support size, support coverage, and
entropy. We show that the cost of privacy is negligible in a variety of
settings, both theoretically and experimentally. Our methods are based on a
sensitivity analysis of several state-of-the-art methods for estimating these
properties with sublinear sample complexities.
|
Deep Reinforcement Learning for Join Order Enumeration | Join order selection plays a significant role in query performance. However,
modern query optimizers typically employ static join enumeration algorithms
that do not receive any feedback about the quality of the resulting plan.
Hence, optimizers often repeatedly choose the same bad plan, as they do not
have a mechanism for "learning from their mistakes". In this paper, we argue
that existing deep reinforcement learning techniques can be applied to address
this challenge. These techniques, powered by artificial neural networks, can
automatically improve decision making by incorporating feedback from their
successes and failures. Towards this goal, we present ReJOIN, a
proof-of-concept join enumerator, and present preliminary results indicating
that ReJOIN can match or outperform the PostgreSQL optimizer in terms of plan
quality and join enumeration efficiency.
|
A Neural Multi-sequence Alignment TeCHnique (NeuMATCH) | The alignment of heterogeneous sequential data (video to text) is an
important and challenging problem. Standard techniques for this task, including
Dynamic Time Warping (DTW) and Conditional Random Fields (CRFs), suffer from
inherent drawbacks. Mainly, the Markov assumption implies that, given the
immediate past, future alignment decisions are independent of further history.
The separation between similarity computation and alignment decision also
prevents end-to-end training. In this paper, we propose an end-to-end neural
architecture where alignment actions are implemented as moving data between
stacks of Long Short-term Memory (LSTM) blocks. This flexible architecture
supports a large variety of alignment tasks, including one-to-one, one-to-many,
skipping unmatched elements, and (with extensions) non-monotonic alignment.
Extensive experiments on semi-synthetic and real datasets show that our
algorithm outperforms state-of-the-art baselines.
|
Constrained Classification and Ranking via Quantiles | In most machine learning applications, classification accuracy is not the
primary metric of interest. Binary classifiers which face class imbalance are
often evaluated by the $F_\beta$ score, area under the precision-recall curve,
Precision at K, and more. The maximization of many of these metrics can be
expressed as a constrained optimization problem, where the constraint is a
function of the classifier's predictions.
In this paper we propose a novel framework for learning with constraints that
can be expressed as a predicted positive rate (or negative rate) on a subset of
the training data. We explicitly model the threshold at which a classifier must
operate to satisfy the constraint, yielding a surrogate loss function which
avoids the complexity of constrained optimization. The method is model-agnostic
and only marginally more expensive than minimization of the unconstrained loss.
Experiments on a variety of benchmarks show competitive performance relative to
existing baselines.
|
NETT: Solving Inverse Problems with Deep Neural Networks | Recovering a function or high-dimensional parameter vector from indirect
measurements is a central task in various scientific areas. Several methods for
solving such inverse problems are well developed and well understood. Recently,
novel algorithms using deep learning and neural networks for inverse problems
appeared. While still in their infancy, these techniques show astonishing
performance for applications like low-dose CT or various sparse data problems.
However, there are few theoretical results for deep learning in inverse
problems. In this paper, we establish a complete convergence analysis for the
proposed NETT (Network Tikhonov) approach to inverse problems. NETT considers
data consistent solutions having small value of a regularizer defined by a
trained neural network. We derive well-posedness results and quantitative error
estimates, and propose a possible strategy for training the regularizer. Our
theoretical results and framework are different from any previous work using
neural networks for solving inverse problems. A possible data driven
regularizer is proposed. Numerical results are presented for a tomographic
sparse data problem, which demonstrate good performance of NETT even for
unknowns of different type from the training data. To derive the convergence
and convergence rates results we introduce a new framework based on the
absolute Bregman distance generalizing the standard Bregman distance from the
convex to the non-convex case.
|
Neural Networks Should Be Wide Enough to Learn Disconnected Decision
Regions | In the recent literature the important role of depth in deep learning has
been emphasized. In this paper we argue that sufficient width of a feedforward
network is equally important by answering the simple question under which
conditions the decision regions of a neural network are connected. It turns out
that for a class of activation functions including leaky ReLU, neural networks
having a pyramidal structure, that is no layer has more hidden units than the
input dimension, produce necessarily connected decision regions. This implies
that a sufficiently wide hidden layer is necessary to guarantee that the
network can produce disconnected decision regions. We discuss the implications
of this result for the construction of neural networks, in particular the
relation to the problem of adversarial manipulation of classifiers.
|
Model-Based Value Estimation for Efficient Model-Free Reinforcement
Learning | Recent model-free reinforcement learning algorithms have proposed
incorporating learned dynamics models as a source of additional data with the
intention of reducing sample complexity. Such methods hold the promise of
incorporating imagined data coupled with a notion of model uncertainty to
accelerate the learning of continuous control tasks. Unfortunately, they rely
on heuristics that limit usage of the dynamics model. We present model-based
value expansion, which controls for uncertainty in the model by only allowing
imagination to fixed depth. By enabling wider use of learned dynamics models
within a model-free reinforcement learning algorithm, we improve value
estimation, which, in turn, reduces the sample complexity of learning.
|
Predicting Recall Probability to Adaptively Prioritize Study | Students have a limited time to study and are typically ineffective at
allocating study time. Machine-directed study strategies that identify which
items need reinforcement and dictate the spacing of repetition have been shown
to help students optimize mastery (Mozer & Lindsey 2017). The large volume of
research on this matter is typically conducted in constructed experimental
settings with fixed instruction, content, and scheduling; in contrast, we aim
to develop methods that can address any demographic, subject matter, or study
schedule. We show two methods that model item-specific recall probability for
use in a discrepancy-reduction instruction strategy. The first model predicts
item recall probability using a multiple logistic regression (MLR) model based
on previous answer correctness and temporal spacing of study. Prompted by
literature suggesting that forgetting is better modeled by the power law than
an exponential decay (Wickelgren 1974), we compare the MLR approach with a
Recurrent Power Law (RPL) model which adaptively fits a forgetting curve. We
then discuss the performance of these models against study datasets comprised
of millions of answers and show that the RPL approach is more accurate and
flexible than the MLR model. Finally, we give an overview of promising future
approaches to knowledge modeling.
|
Approximate Inference for Constructing Astronomical Catalogs from Images | We present a new, fully generative model for constructing astronomical
catalogs from optical telescope image sets. Each pixel intensity is treated as
a random variable with parameters that depend on the latent properties of stars
and galaxies. These latent properties are themselves modeled as random. We
compare two procedures for posterior inference. One procedure is based on
Markov chain Monte Carlo (MCMC) while the other is based on variational
inference (VI). The MCMC procedure excels at quantifying uncertainty, while the
VI procedure is 1000 times faster. On a supercomputer, the VI procedure
efficiently uses 665,000 CPU cores to construct an astronomical catalog from 50
terabytes of images in 14.6 minutes, demonstrating the scaling characteristics
necessary to construct catalogs for upcoming astronomical surveys.
|
SQL-Rank: A Listwise Approach to Collaborative Ranking | In this paper, we propose a listwise approach for constructing user-specific
rankings in recommendation systems in a collaborative fashion. We contrast the
listwise approach to previous pointwise and pairwise approaches, which are
based on treating either each rating or each pairwise comparison as an
independent instance respectively. By extending the work of (Cao et al. 2007),
we cast listwise collaborative ranking as maximum likelihood under a
permutation model which applies probability mass to permutations based on a low
rank latent score matrix. We present a novel algorithm called SQL-Rank, which
can accommodate ties and missing data and can run in linear time. We develop a
theoretical framework for analyzing listwise ranking methods based on a novel
representation theory for the permutation model. Applying this framework to
collaborative ranking, we derive asymptotic statistical rates as the number of
users and items grow together. We conclude by demonstrating that our SQL-Rank
method often outperforms current state-of-the-art algorithms for implicit
feedback such as Weighted-MF and BPR and achieve favorable results when
compared to explicit feedback algorithms such as matrix factorization and
collaborative ranking.
|
Separators and Adjustment Sets in Causal Graphs: Complete Criteria and
an Algorithmic Framework | Principled reasoning about the identifiability of causal effects from
non-experimental data is an important application of graphical causal models.
This paper focuses on effects that are identifiable by covariate adjustment, a
commonly used estimation approach. We present an algorithmic framework for
efficiently testing, constructing, and enumerating $m$-separators in ancestral
graphs (AGs), a class of graphical causal models that can represent uncertainty
about the presence of latent confounders. Furthermore, we prove a reduction
from causal effect identification by covariate adjustment to $m$-separation in
a subgraph for directed acyclic graphs (DAGs) and maximal ancestral graphs
(MAGs). Jointly, these results yield constructive criteria that characterize
all adjustment sets as well as all minimal and minimum adjustment sets for
identification of a desired causal effect with multivariate exposures and
outcomes in the presence of latent confounding. Our results extend several
existing solutions for special cases of these problems. Our efficient
algorithms allowed us to empirically quantify the identifiability gap between
covariate adjustment and the do-calculus in random DAGs and MAGs, covering a
wide range of scenarios. Implementations of our algorithms are provided in the
R package dagitty.
|
Learning Longer-term Dependencies in RNNs with Auxiliary Losses | Despite recent advances in training recurrent neural networks (RNNs),
capturing long-term dependencies in sequences remains a fundamental challenge.
Most approaches use backpropagation through time (BPTT), which is difficult to
scale to very long sequences. This paper proposes a simple method that improves
the ability to capture long term dependencies in RNNs by adding an unsupervised
auxiliary loss to the original objective. This auxiliary loss forces RNNs to
either reconstruct previous events or predict next events in a sequence, making
truncated backpropagation feasible for long sequences and also improving full
BPTT. We evaluate our method on a variety of settings, including pixel-by-pixel
image classification with sequence lengths up to 16\,000, and a real document
classification benchmark. Our results highlight good performance and resource
efficiency of this approach over competitive baselines, including other
recurrent models and a comparable sized Transformer. Further analyses reveal
beneficial effects of the auxiliary loss on optimization and regularization, as
well as extreme cases where there is little to no backpropagation.
|
Deep Learning for Causal Inference | In this paper, we propose deep learning techniques for econometrics,
specifically for causal inference and for estimating individual as well as
average treatment effects. The contribution of this paper is twofold: 1. For
generalized neighbor matching to estimate individual and average treatment
effects, we analyze the use of autoencoders for dimensionality reduction while
maintaining the local neighborhood structure among the data points in the
embedding space. This deep learning based technique is shown to perform better
than simple k nearest neighbor matching for estimating treatment effects,
especially when the data points have several features/covariates but reside in
a low dimensional manifold in high dimensional space. We also observe better
performance than manifold learning methods for neighbor matching. 2. Propensity
score matching is one specific and popular way to perform matching in order to
estimate average and individual treatment effects. We propose the use of deep
neural networks (DNNs) for propensity score matching, and present a network
called PropensityNet for this. This is a generalization of the logistic
regression technique traditionally used to estimate propensity scores and we
show empirically that DNNs perform better than logistic regression at
propensity score matching. Code for both methods will be made available shortly
on Github at: https://github.com/vikas84bf
|
Autoencoding topology | The problem of learning a manifold structure on a dataset is framed in terms
of a generative model, to which we use ideas behind autoencoders (namely
adversarial/Wasserstein autoencoders) to fit deep neural networks. From a
machine learning perspective, the resulting structure, an atlas of a manifold,
may be viewed as a combination of dimensionality reduction and "fuzzy"
clustering.
|
Modeling reverse thinking for machine learning | Human inertial thinking schemes can be formed through learning, which are
then applied to quickly solve similar problems later. However, when problems
are significantly different, inertial thinking generally presents the solutions
that are definitely imperfect. In such cases, people will apply creative
thinking, such as reverse thinking, to solve problems. Similarly, machine
learning methods also form inertial thinking schemes through learning the
knowledge from a large amount of data. However, when the testing data are
vastly difference, the formed inertial thinking schemes will inevitably
generate errors. This kind of inertial thinking is called illusion inertial
thinking. Because all machine learning methods do not consider illusion
inertial thinking, in this paper we propose a new method that uses reverse
thinking to correct illusion inertial thinking, which increases the
generalization ability of machine learning methods. Experimental results on
benchmark datasets are used to validate the proposed method.
|
Towards Cooperation in Sequential Prisoner's Dilemmas: a Deep Multiagent
Reinforcement Learning Approach | The Iterated Prisoner's Dilemma has guided research on social dilemmas for
decades. However, it distinguishes between only two atomic actions: cooperate
and defect. In real-world prisoner's dilemmas, these choices are temporally
extended and different strategies may correspond to sequences of actions,
reflecting grades of cooperation. We introduce a Sequential Prisoner's Dilemma
(SPD) game to better capture the aforementioned characteristics. In this work,
we propose a deep multiagent reinforcement learning approach that investigates
the evolution of mutual cooperation in SPD games. Our approach consists of two
phases. The first phase is offline: it synthesizes policies with different
cooperation degrees and then trains a cooperation degree detection network. The
second phase is online: an agent adaptively selects its policy based on the
detected degree of opponent cooperation. The effectiveness of our approach is
demonstrated in two representative SPD 2D games: the Apple-Pear game and the
Fruit Gathering game. Experimental results show that our strategy can avoid
being exploited by exploitative opponents and achieve cooperation with
cooperative opponents.
|
Learning with Correntropy-induced Losses for Regression with Mixture of
Symmetric Stable Noise | In recent years, correntropy and its applications in machine learning have
been drawing continuous attention owing to its merits in dealing with
non-Gaussian noise and outliers. However, theoretical understanding of
correntropy, especially in the statistical learning context, is still limited.
In this study, within the statistical learning framework, we investigate
correntropy based regression in the presence of non-Gaussian noise or outliers.
Motivated by the practical way of generating non-Gaussian noise or outliers, we
introduce mixture of symmetric stable noise, which include Gaussian noise,
Cauchy noise, and their mixture as special cases, to model non-Gaussian noise
or outliers. We demonstrate that under the mixture of symmetric stable noise
assumption, correntropy based regression can learn the conditional mean
function or the conditional median function well without resorting to the
finite-variance or even the finite first-order moment condition on the noise.
In particular, for the above two cases, we establish asymptotic optimal
learning rates for correntropy based regression estimators that are
asymptotically of type $\mathcal{O}(n^{-1})$. These results justify the
effectiveness of the correntropy based regression estimators in dealing with
outliers as well as non-Gaussian noise. We believe that the present study
completes our understanding towards correntropy based regression from a
statistical learning viewpoint, and may also shed some light on robust
statistical learning for regression.
|
Learning Sparse Structured Ensembles with SG-MCMC and Network Pruning | An ensemble of neural networks is known to be more robust and accurate than
an individual network, however usually with linearly-increased cost in both
training and testing. In this work, we propose a two-stage method to learn
Sparse Structured Ensembles (SSEs) for neural networks. In the first stage, we
run SG-MCMC with group sparse priors to draw an ensemble of samples from the
posterior distribution of network parameters. In the second stage, we apply
weight-pruning to each sampled network and then perform retraining over the
remained connections. In this way of learning SSEs with SG-MCMC and pruning, we
not only achieve high prediction accuracy since SG-MCMC enhances exploration of
the model-parameter space, but also reduce memory and computation cost
significantly in both training and testing of NN ensembles. This is thoroughly
evaluated in the experiments of learning SSE ensembles of both FNNs and LSTMs.
For example, in LSTM based language modeling (LM), we obtain 21% relative
reduction in LM perplexity by learning a SSE of 4 large LSTM models, which has
only 30% of model parameters and 70% of computations in total, as compared to
the baseline large LSTM LM. To the best of our knowledge, this work represents
the first methodology and empirical study of integrating SG-MCMC, group sparse
prior and network pruning together for learning NN ensembles.
|
Smoothed analysis for low-rank solutions to semidefinite programs in
quadratic penalty form | Semidefinite programs (SDP) are important in learning and combinatorial
optimization with numerous applications. In pursuit of low-rank solutions and
low complexity algorithms, we consider the Burer--Monteiro factorization
approach for solving SDPs. We show that all approximate local optima are global
optima for the penalty formulation of appropriately rank-constrained SDPs as
long as the number of constraints scales sub-quadratically with the desired
rank of the optimal solution. Our result is based on a simple penalty function
formulation of the rank-constrained SDP along with a smoothed analysis to avoid
worst-case cost matrices. We particularize our results to two applications,
namely, Max-Cut and matrix completion.
|
The Anisotropic Noise in Stochastic Gradient Descent: Its Behavior of
Escaping from Sharp Minima and Regularization Effects | Understanding the behavior of stochastic gradient descent (SGD) in the
context of deep neural networks has raised lots of concerns recently. Along
this line, we study a general form of gradient based optimization dynamics with
unbiased noise, which unifies SGD and standard Langevin dynamics. Through
investigating this general optimization dynamics, we analyze the behavior of
SGD on escaping from minima and its regularization effects. A novel indicator
is derived to characterize the efficiency of escaping from minima through
measuring the alignment of noise covariance and the curvature of loss function.
Based on this indicator, two conditions are established to show which type of
noise structure is superior to isotropic noise in term of escaping efficiency.
We further show that the anisotropic noise in SGD satisfies the two conditions,
and thus helps to escape from sharp and poor minima effectively, towards more
stable and flat minima that typically generalize well. We systematically design
various experiments to verify the benefits of the anisotropic noise, compared
with full gradient descent plus isotropic diffusion (i.e. Langevin dynamics).
|
Learning Flexible and Reusable Locomotion Primitives for a Microrobot | The design of gaits for robot locomotion can be a daunting process which
requires significant expert knowledge and engineering. This process is even
more challenging for robots that do not have an accurate physical model, such
as compliant or micro-scale robots. Data-driven gait optimization provides an
automated alternative to analytical gait design. In this paper, we propose a
novel approach to efficiently learn a wide range of locomotion tasks with
walking robots. This approach formalizes locomotion as a contextual policy
search task to collect data, and subsequently uses that data to learn
multi-objective locomotion primitives that can be used for planning. As a
proof-of-concept we consider a simulated hexapod modeled after a recently
developed microrobot, and we thoroughly evaluate the performance of this
microrobot on different tasks and gaits. Our results validate the proposed
controller and learning scheme on single and multi-objective locomotion tasks.
Moreover, the experimental simulations show that without any prior knowledge
about the robot used (e.g., dynamics model), our approach is capable of
learning locomotion primitives within 250 trials and subsequently using them to
successfully navigate through a maze.
|
Scalar Quantization as Sparse Least Square Optimization | Quantization can be used to form new vectors/matrices with shared values
close to the original. In recent years, the popularity of scalar quantization
for value-sharing applications has been soaring as it has been found huge
utilities in reducing the complexity of neural networks. Existing
clustering-based quantization techniques, while being well-developed, have
multiple drawbacks including the dependency of the random seed, empty or
out-of-the-range clusters, and high time complexity for a large number of
clusters. To overcome these problems, in this paper, the problem of scalar
quantization is examined from a new perspective, namely sparse least square
optimization. Specifically, inspired by the property of sparse least square
regression, several quantization algorithms based on $l_1$ least square are
proposed. In addition, similar schemes with $l_1 + l_2$ and $l_0$
regularization are proposed. Furthermore, to compute quantization results with
a given amount of values/clusters, this paper designed an iterative method and
a clustering-based method, and both of them are built on sparse least square.
The paper shows that the latter method is mathematically equivalent to an
improved version of k-means clustering-based quantization algorithm, although
the two algorithms originated from different intuitions. The algorithms
proposed were tested with three types of data and their computational
performances, including information loss, time consumption, and the
distribution of the values of the sparse vectors, were compared and analyzed.
The paper offers a new perspective to probe the area of quantization, and the
algorithms proposed can outperform existing methods especially under some
bit-width reduction scenarios, when the required post-quantization resolution
(number of values) is not significantly lower than the original number.
|
prDeep: Robust Phase Retrieval with a Flexible Deep Network | Phase retrieval algorithms have become an important component in many modern
computational imaging systems. For instance, in the context of ptychography and
speckle correlation imaging, they enable imaging past the diffraction limit and
through scattering media, respectively. Unfortunately, traditional phase
retrieval algorithms struggle in the presence of noise. Progress has been made
recently on more robust algorithms using signal priors, but at the expense of
limiting the range of supported measurement models (e.g., to Gaussian or coded
diffraction patterns). In this work we leverage the regularization-by-denoising
framework and a convolutional neural network denoiser to create prDeep, a new
phase retrieval algorithm that is both robust and broadly applicable. We test
and validate prDeep in simulation to demonstrate that it is robust to noise and
can handle a variety of system models.
A MatConvNet implementation of prDeep is available at
https://github.com/ricedsp/prDeep.
|
Interval-based Prediction Uncertainty Bound Computation in Learning with
Missing Values | The problem of machine learning with missing values is common in many areas.
A simple approach is to first construct a dataset without missing values simply
by discarding instances with missing entries or by imputing a fixed value for
each missing entry, and then train a prediction model with the new dataset. A
drawback of this naive approach is that the uncertainty in the missing entries
is not properly incorporated in the prediction. In order to evaluate prediction
uncertainty, the multiple imputation (MI) approach has been studied, but the
performance of MI is sensitive to the choice of the probabilistic model of the
true values in the missing entries, and the computational cost of MI is high
because multiple models must be trained. In this paper, we propose an
alternative approach called the Interval-based Prediction Uncertainty Bounding
(IPUB) method. The IPUB method represents the uncertainties due to missing
entries as intervals, and efficiently computes the lower and upper bounds of
the prediction results when all possible training sets constructed by imputing
arbitrary values in the intervals are considered. The IPUB method can be
applied to a wide class of convex learning algorithms including penalized
least-squares regression, support vector machine (SVM), and logistic
regression. We demonstrate the advantages of the IPUB method by comparing it
with an existing method in numerical experiment with benchmark datasets.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.