title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Geometrical Insights for Implicit Generative Modeling | Learning algorithms for implicit generative models can optimize a variety of
criteria that measure how the data distribution differs from the implicit model
distribution, including the Wasserstein distance, the Energy distance, and the
Maximum Mean Discrepancy criterion. A careful look at the geometries induced by
these distances on the space of probability measures reveals interesting
differences. In particular, we can establish surprising approximate global
convergence guarantees for the $1$-Wasserstein distance,even when the
parametric generator has a nonconvex parametrization.
|
DropMax: Adaptive Variational Softmax | We propose DropMax, a stochastic version of softmax classifier which at each
iteration drops non-target classes according to dropout probabilities
adaptively decided for each instance. Specifically, we overlay binary masking
variables over class output probabilities, which are input-adaptively learned
via variational inference. This stochastic regularization has an effect of
building an ensemble classifier out of exponentially many classifiers with
different decision boundaries. Moreover, the learning of dropout rates for
non-target classes on each instance allows the classifier to focus more on
classification against the most confusing classes. We validate our model on
multiple public datasets for classification, on which it obtains significantly
improved accuracy over the regular softmax classifier and other baselines.
Further analysis of the learned dropout probabilities shows that our model
indeed selects confusing classes more often when it performs classification.
|
Estimating the Success of Unsupervised Image to Image Translation | While in supervised learning, the validation error is an unbiased estimator
of the generalization (test) error and complexity-based generalization bounds
are abundant, no such bounds exist for learning a mapping in an unsupervised
way. As a result, when training GANs and specifically when using GANs for
learning to map between domains in a completely unsupervised way, one is forced
to select the hyperparameters and the stopping epoch by subjectively examining
multiple options. We propose a novel bound for predicting the success of
unsupervised cross domain mapping methods, which is motivated by the recently
proposed Simplicity Principle. The bound can be applied both in expectation,
for comparing hyperparameters and for selecting a stopping criterion, or per
sample, in order to predict the success of a specific cross-domain translation.
The utility of the bound is demonstrated in an extensive set of experiments
employing multiple recent algorithms. Our code is available at
https://github.com/sagiebenaim/gan_bound .
|
Non-convex Optimization for Machine Learning | A vast majority of machine learning algorithms train their models and perform
inference by solving optimization problems. In order to capture the learning
and prediction problems accurately, structural constraints such as sparsity or
low rank are frequently imposed or else the objective itself is designed to be
a non-convex function. This is especially true of algorithms that operate in
high-dimensional spaces or that train non-linear models such as tensor models
and deep networks.
The freedom to express the learning problem as a non-convex optimization
problem gives immense modeling power to the algorithm designer, but often such
problems are NP-hard to solve. A popular workaround to this has been to relax
non-convex problems to convex ones and use traditional methods to solve the
(convex) relaxed optimization problems. However this approach may be lossy and
nevertheless presents significant challenges for large scale optimization.
On the other hand, direct approaches to non-convex optimization have met with
resounding success in several domains and remain the methods of choice for the
practitioner, as they frequently outperform relaxation-based techniques -
popular heuristics include projected gradient descent and alternating
minimization. However, these are often poorly understood in terms of their
convergence and other properties.
This monograph presents a selection of recent advances that bridge a
long-standing gap in our understanding of these heuristics. The monograph will
lead the reader through several widely used non-convex optimization techniques,
as well as applications thereof. The goal of this monograph is to both,
introduce the rich literature in this area, as well as equip the reader with
the tools and techniques needed to analyze these simple procedures for
non-convex problems.
|
Note on Attacking Object Detectors with Adversarial Stickers | Deep learning has proven to be a powerful tool for computer vision and has
seen widespread adoption for numerous tasks. However, deep learning algorithms
are known to be vulnerable to adversarial examples. These adversarial inputs
are created such that, when provided to a deep learning algorithm, they are
very likely to be mislabeled. This can be problematic when deep learning is
used to assist in safety critical decisions. Recent research has shown that
classifiers can be attacked by physical adversarial examples under various
physical conditions. Given the fact that state-of-the-art objection detection
algorithms are harder to be fooled by the same set of adversarial examples,
here we show that these detectors can also be attacked by physical adversarial
examples. In this note, we briefly show both static and dynamic test results.
We design an algorithm that produces physical adversarial inputs, which can
fool the YOLO object detector and can also attack Faster-RCNN with relatively
high success rate based on transferability. Furthermore, our algorithm can
compress the size of the adversarial inputs to stickers that, when attached to
the targeted object, result in the detector either mislabeling or not detecting
the object a high percentage of the time. This note provides a small set of
results. Our upcoming paper will contain a thorough evaluation on other object
detectors, and will present the algorithm.
|
Multiview Deep Learning for Predicting Twitter Users' Location | The problem of predicting the location of users on large social networks like
Twitter has emerged from real-life applications such as social unrest detection
and online marketing. Twitter user geolocation is a difficult and active
research topic with a vast literature. Most of the proposed methods follow
either a content-based or a network-based approach. The former exploits
user-generated content while the latter utilizes the connection or interaction
between Twitter users. In this paper, we introduce a novel method combining the
strength of both approaches. Concretely, we propose a multi-entry neural
network architecture named MENET leveraging the advances in deep learning and
multiview learning. The generalizability of MENET enables the integration of
multiple data representations. In the context of Twitter user geolocation, we
realize MENET with textual, network, and metadata features. Considering the
natural distribution of Twitter users across the concerned geographical area,
we subdivide the surface of the earth into multi-scale cells and train MENET
with the labels of the cells. We show that our method outperforms the state of
the art by a large margin on three benchmark datasets.
|
Profit Driven Decision Trees for Churn Prediction | Customer retention campaigns increasingly rely on predictive models to detect
potential churners in a vast customer base. From the perspective of machine
learning, the task of predicting customer churn can be presented as a binary
classification problem. Using data on historic behavior, classification
algorithms are built with the purpose of accurately predicting the probability
of a customer defecting. The predictive churn models are then commonly selected
based on accuracy related performance measures such as the area under the ROC
curve (AUC). However, these models are often not well aligned with the core
business requirement of profit maximization, in the sense that, the models fail
to take into account not only misclassification costs, but also the benefits
originating from a correct classification. Therefore, the aim is to construct
churn prediction models that are profitable and preferably interpretable too.
The recently developed expected maximum profit measure for customer churn
(EMPC) has been proposed in order to select the most profitable churn model. We
present a new classifier that integrates the EMPC metric directly into the
model construction. Our technique, called ProfTree, uses an evolutionary
algorithm for learning profit driven decision trees. In a benchmark study with
real-life data sets from various telecommunication service providers, we show
that ProfTree achieves significant profit improvements compared to classic
accuracy driven tree-based methods.
|
A Deep Learning Interpretable Classifier for Diabetic Retinopathy
Disease Grading | Deep neural network models have been proven to be very successful in image
classification tasks, also for medical diagnosis, but their main concern is its
lack of interpretability. They use to work as intuition machines with high
statistical confidence but unable to give interpretable explanations about the
reported results. The vast amount of parameters of these models make difficult
to infer a rationale interpretation from them. In this paper we present a
diabetic retinopathy interpretable classifier able to classify retine images
into the different levels of disease severity and of explaining its results by
assigning a score for every point in the hidden and input space, evaluating its
contribution to the final classification in a linear way. The generated visual
maps can be interpreted by an expert in order to compare its own knowledge with
the interpretation given by the model.
|
Unifying Map and Landmark Based Representations for Visual Navigation | This works presents a formulation for visual navigation that unifies map
based spatial reasoning and path planning, with landmark based robust plan
execution in noisy environments. Our proposed formulation is learned from data
and is thus able to leverage statistical regularities of the world. This allows
it to efficiently navigate in novel environments given only a sparse set of
registered images as input for building representations for space. Our
formulation is based on three key ideas: a learned path planner that outputs
path plans to reach the goal, a feature synthesis engine that predicts features
for locations along the planned path, and a learned goal-driven closed loop
controller that can follow plans given these synthesized features. We test our
approach for goal-driven navigation in simulated real world environments and
report performance gains over competitive baseline approaches.
|
Combining Static and Dynamic Features for Multivariate Sequence
Classification | Model precision in a classification task is highly dependent on the feature
space that is used to train the model. Moreover, whether the features are
sequential or static will dictate which classification method can be applied as
most of the machine learning algorithms are designed to deal with either one or
another type of data. In real-life scenarios, however, it is often the case
that both static and dynamic features are present, or can be extracted from the
data. In this work, we demonstrate how generative models such as Hidden Markov
Models (HMM) and Long Short-Term Memory (LSTM) artificial neural networks can
be used to extract temporal information from the dynamic data. We explore how
the extracted information can be combined with the static features in order to
improve the classification performance. We evaluate the existing techniques and
suggest a hybrid approach, which outperforms other methods on several public
datasets.
|
Reachable Set Computation and Safety Verification for Neural Networks
with ReLU Activations | Neural networks have been widely used to solve complex real-world problems.
Due to the complicate, nonlinear, non-convex nature of neural networks, formal
safety guarantees for the output behaviors of neural networks will be crucial
for their applications in safety-critical systems.In this paper, the output
reachable set computation and safety verification problems for a class of
neural networks consisting of Rectified Linear Unit (ReLU) activation functions
are addressed. A layer-by-layer approach is developed to compute output
reachable set. The computation is formulated in the form of a set of
manipulations for a union of polyhedra, which can be efficiently applied with
the aid of polyhedron computation tools. Based on the output reachable set
computation results, the safety verification for a ReLU neural network can be
performed by checking the intersections of unsafe regions and output reachable
set described by a union of polyhedra. A numerical example of a randomly
generated ReLU neural network is provided to show the effectiveness of the
approach developed in this paper.
|
Multi-task learning of time series and its application to the travel
demand | We address the problem of modeling and prediction of a set of temporal events
in the context of intelligent transportation systems. To leverage the
information shared by different events, we propose a multi-task learning
framework. We develop a support vector regression model for joint learning of
mutually dependent time series. It is the regularization-based multi-task
learning previously developed for the classification case and extended to time
series. We discuss the relatedness of observed time series and first deploy the
dynamic time warping distance measure to identify groups of similar series.
Then we take into account both time and scale warping and propose to align
multiple time series by inferring their common latent representation. We test
the proposed models on the problem of travel demand prediction in Nancy
(France) public transport system and analyze the benefits of multi-task
learning.
|
Fair Forests: Regularized Tree Induction to Minimize Model Bias | The potential lack of fairness in the outputs of machine learning algorithms
has recently gained attention both within the research community as well as in
society more broadly. Surprisingly, there is no prior work developing
tree-induction algorithms for building fair decision trees or fair random
forests. These methods have widespread popularity as they are one of the few to
be simultaneously interpretable, non-linear, and easy-to-use. In this paper we
develop, to our knowledge, the first technique for the induction of fair
decision trees. We show that our "Fair Forest" retains the benefits of the
tree-based approach, while providing both greater accuracy and fairness than
other alternatives, for both "group fairness" and "individual fairness.'" We
also introduce new measures for fairness which are able to handle multinomial
and continues attributes as well as regression problems, as opposed to binary
attributes and labels only. Finally, we demonstrate a new, more robust
evaluation procedure for algorithms that considers the dataset in its entirety
rather than only a specific protected attribute.
|
Block-Diagonal and LT Codes for Distributed Computing With Straggling
Servers | We propose two coded schemes for the distributed computing problem of
multiplying a matrix by a set of vectors. The first scheme is based on
partitioning the matrix into submatrices and applying maximum distance
separable (MDS) codes to each submatrix. For this scheme, we prove that up to a
given number of partitions the communication load and the computational delay
(not including the encoding and decoding delay) are identical to those of the
scheme recently proposed by Li et al., based on a single, long MDS code.
However, due to the use of shorter MDS codes, our scheme yields a significantly
lower overall computational delay when the delay incurred by encoding and
decoding is also considered. We further propose a second coded scheme based on
Luby Transform (LT) codes under inactivation decoding. Interestingly, LT codes
may reduce the delay over the partitioned scheme at the expense of an increased
communication load. We also consider distributed computing under a deadline and
show numerically that the proposed schemes outperform other schemes in the
literature, with the LT code-based scheme yielding the best performance for the
scenarios considered.
|
Interventions over Predictions: Reframing the Ethical Debate for
Actuarial Risk Assessment | Actuarial risk assessments might be unduly perceived as a neutral way to
counteract implicit bias and increase the fairness of decisions made at almost
every juncture of the criminal justice system, from pretrial release to
sentencing, parole and probation. In recent times these assessments have come
under increased scrutiny, as critics claim that the statistical techniques
underlying them might reproduce existing patterns of discrimination and
historical biases that are reflected in the data. Much of this debate is
centered around competing notions of fairness and predictive accuracy, resting
on the contested use of variables that act as "proxies" for characteristics
legally protected against discrimination, such as race and gender. We argue
that a core ethical debate surrounding the use of regression in risk
assessments is not simply one of bias or accuracy. Rather, it's one of purpose.
If machine learning is operationalized merely in the service of predicting
individual future crime, then it becomes difficult to break cycles of
criminalization that are driven by the iatrogenic effects of the criminal
justice system itself. We posit that machine learning should not be used for
prediction, but rather to surface covariates that are fed into a causal model
for understanding the social, structural and psychological drivers of crime. We
propose an alternative application of machine learning and causal inference
away from predicting risk scores to risk mitigation.
|
How Well Can Generative Adversarial Networks Learn Densities: A
Nonparametric View | We study in this paper the rate of convergence for learning densities under
the Generative Adversarial Networks (GAN) framework, borrowing insights from
nonparametric statistics. We introduce an improved GAN estimator that achieves
a faster rate, through simultaneously leveraging the level of smoothness in the
target density and the evaluation metric, which in theory remedies the mode
collapse problem reported in the literature. A minimax lower bound is
constructed to show that when the dimension is large, the exponent in the rate
for the new GAN estimator is near optimal. One can view our results as
answering in a quantitative way how well GAN learns a wide range of densities
with different smoothness properties, under a hierarchy of evaluation metrics.
As a byproduct, we also obtain improved generalization bounds for GAN with
deeper ReLU discriminator network.
|
ReabsNet: Detecting and Revising Adversarial Examples | Though deep neural network has hit a huge success in recent studies and
applica- tions, it still remains vulnerable to adversarial perturbations which
are imperceptible to humans. To address this problem, we propose a novel
network called ReabsNet to achieve high classification accuracy in the face of
various attacks. The approach is to augment an existing classification network
with a guardian network to detect if a sample is natural or has been
adversarially perturbed. Critically, instead of simply rejecting adversarial
examples, we revise them to get their true labels. We exploit the observation
that a sample containing adversarial perturbations has a possibility of
returning to its true class after revision. We demonstrate that our ReabsNet
outperforms the state-of-the-art defense method under various adversarial
attacks.
|
Linear centralization classifier | A classification algorithm, called the Linear Centralization Classifier
(LCC), is introduced. The algorithm seeks to find a transformation that best
maps instances from the feature space to a space where they concentrate towards
the center of their own classes, while maximimizing the distance between class
centers. We formulate the classifier as a quadratic program with quadratic
constraints. We then simplify this formulation to a linear program that can be
solved effectively using a linear programming solver (e.g., simplex-dual). We
extend the formulation for LCC to enable the use of kernel functions for
non-linear classification applications. We compare our method with two standard
classification methods (support vector machine and linear discriminant
analysis) and four state-of-the-art classification methods when they are
applied to eight standard classification datasets. Our experimental results
show that LCC is able to classify instances more accurately (based on the area
under the receiver operating characteristic) in comparison to other tested
methods on the chosen datasets. We also report the results for LCC with a
particular kernel to solve for synthetic non-linear classification problems.
|
Recurrent Pixel Embedding for Instance Grouping | We introduce a differentiable, end-to-end trainable framework for solving
pixel-level grouping problems such as instance segmentation consisting of two
novel components. First, we regress pixels into a hyper-spherical embedding
space so that pixels from the same group have high cosine similarity while
those from different groups have similarity below a specified margin. We
analyze the choice of embedding dimension and margin, relating them to
theoretical results on the problem of distributing points uniformly on the
sphere. Second, to group instances, we utilize a variant of mean-shift
clustering, implemented as a recurrent neural network parameterized by kernel
bandwidth. This recurrent grouping module is differentiable, enjoys convergent
dynamics and probabilistic interpretability. Backpropagating the group-weighted
loss through this module allows learning to focus on only correcting embedding
errors that won't be resolved during subsequent clustering. Our framework,
while conceptually simple and theoretically abundant, is also practically
effective and computationally efficient. We demonstrate substantial
improvements over state-of-the-art instance segmentation for object proposal
generation, as well as demonstrating the benefits of grouping loss for
classification tasks such as boundary detection and semantic segmentation.
|
Learning and Transferring IDs Representation in E-commerce | Many machine intelligence techniques are developed in E-commerce and one of
the most essential components is the representation of IDs, including user ID,
item ID, product ID, store ID, brand ID, category ID etc. The classical
encoding based methods (like one-hot encoding) are inefficient in that it
suffers sparsity problems due to its high dimension, and it cannot reflect the
relationships among IDs, either homogeneous or heterogeneous ones. In this
paper, we propose an embedding based framework to learn and transfer the
representation of IDs. As the implicit feedbacks of users, a tremendous amount
of item ID sequences can be easily collected from the interactive sessions. By
jointly using these informative sequences and the structural connections among
IDs, all types of IDs can be embedded into one low-dimensional semantic space.
Subsequently, the learned representations are utilized and transferred in four
scenarios: (i) measuring the similarity between items, (ii) transferring from
seen items to unseen items, (iii) transferring across different domains, (iv)
transferring across different tasks. We deploy and evaluate the proposed
approach in Hema App and the results validate its effectiveness.
|
Benchmarking Decoupled Neural Interfaces with Synthetic Gradients | Artifical Neural Networks are a particular class of learning systems modeled
after biological neural functions with an interesting penchant for Hebbian
learning, that is "neurons that wire together, fire together". However, unlike
their natural counterparts, artificial neural networks have a close and
stringent coupling between the modules of neurons in the network. This coupling
or locking imposes upon the network a strict and inflexible structure that
prevent layers in the network from updating their weights until a full
feed-forward and backward pass has occurred. Such a constraint though may have
sufficed for a while, is now no longer feasible in the era of very-large-scale
machine learning, coupled with the increased desire for parallelization of the
learning process across multiple computing infrastructures. To solve this
problem, synthetic gradients (SG) with decoupled neural interfaces (DNI) are
introduced as a viable alternative to the backpropagation algorithm. This paper
performs a speed benchmark to compare the speed and accuracy capabilities of
SG-DNI as opposed to a standard neural interface using multilayer perceptron
MLP. SG-DNI shows good promise, in that it not only captures the learning
problem, it is also over 3-fold faster due to it asynchronous learning
capabilities.
|
Inverse Classification for Comparison-based Interpretability in Machine
Learning | In the context of post-hoc interpretability, this paper addresses the task of
explaining the prediction of a classifier, considering the case where no
information is available, neither on the classifier itself, nor on the
processed data (neither the training nor the test data). It proposes an
instance-based approach whose principle consists in determining the minimal
changes needed to alter a prediction: given a data point whose classification
must be explained, the proposed method consists in identifying a close
neighbour classified differently, where the closeness definition integrates a
sparsity constraint. This principle is implemented using observation generation
in the Growing Spheres algorithm. Experimental results on two datasets
illustrate the relevance of the proposed approach that can be used to gain
knowledge about the classifier.
|
True Asymptotic Natural Gradient Optimization | We introduce a simple algorithm, True Asymptotic Natural Gradient
Optimization (TANGO), that converges to a true natural gradient descent in the
limit of small learning rates, without explicit Fisher matrix estimation.
For quadratic models the algorithm is also an instance of averaged stochastic
gradient, where the parameter is a moving average of a "fast", constant-rate
gradient descent. TANGO appears as a particular de-linearization of averaged
SGD, and is sometimes quite different on non-quadratic models. This further
connects averaged SGD and natural gradient, both of which are arguably optimal
asymptotically.
In large dimension, small learning rates will be required to approximate the
natural gradient well. Still, this shows it is possible to get arbitrarily
close to exact natural gradient descent with a lightweight algorithm.
|
Diversifying Support Vector Machines for Boosting using Kernel
Perturbation: Applications to Class Imbalance and Small Disjuncts | The diversification (generating slightly varying separating discriminators)
of Support Vector Machines (SVMs) for boosting has proven to be a challenge due
to the strong learning nature of SVMs. Based on the insight that perturbing the
SVM kernel may help in diversifying SVMs, we propose two kernel perturbation
based boosting schemes where the kernel is modified in each round so as to
increase the resolution of the kernel-induced Reimannian metric in the vicinity
of the datapoints misclassified in the previous round. We propose a method for
identifying the disjuncts in a dataset, dispelling the dependence on rule-based
learning methods for identifying the disjuncts. We also present a new
performance measure called Geometric Small Disjunct Index (GSDI) to quantify
the performance on small disjuncts for balanced as well as class imbalanced
datasets. Experimental comparison with a variety of state-of-the-art algorithms
is carried out using the best classifiers of each type selected by a new
approach inspired by multi-criteria decision making. The proposed method is
found to outperform the contending state-of-the-art methods on different
datasets (ranging from mildly imbalanced to highly imbalanced and characterized
by varying number of disjuncts) in terms of three different performance indices
(including the proposed GSDI).
|
Contemporary machine learning: a guide for practitioners in the physical
sciences | Machine learning is finding increasingly broad application in the physical
sciences. This most often involves building a model relationship between a
dependent, measurable output and an associated set of controllable, but
complicated, independent inputs. We present a tutorial on current techniques in
machine learning -- a jumping-off point for interested researchers to advance
their work. We focus on deep neural networks with an emphasis on demystifying
deep learning. We begin with background ideas in machine learning and some
example applications from current research in plasma physics. We discuss
supervised learning techniques for modeling complicated functions, beginning
with familiar regression schemes, then advancing to more sophisticated deep
learning methods. We also address unsupervised learning and techniques for
reducing the dimensionality of input spaces. Along the way, we describe methods
for practitioners to help ensure that their models generalize from their
training data to as-yet-unseen test data. We describe classes of tasks --
predicting scalars, handling images, fitting time-series -- and prepare the
reader to choose an appropriate technique. We finally point out some
limitations to modern machine learning and speculate on some ways that
practitioners from the physical sciences may be particularly suited to help.
|
Adaptive Stochastic Dual Coordinate Ascent for Conditional Random Fields | This work investigates the training of conditional random fields (CRFs) via
the stochastic dual coordinate ascent (SDCA) algorithm of Shalev-Shwartz and
Zhang (2016). SDCA enjoys a linear convergence rate and a strong empirical
performance for binary classification problems. However, it has never been used
to train CRFs. Yet it benefits from an `exact' line search with a single
marginalization oracle call, unlike previous approaches. In this paper, we
adapt SDCA to train CRFs, and we enhance it with an adaptive non-uniform
sampling strategy based on block duality gaps. We perform experiments on four
standard sequence prediction tasks. SDCA demonstrates performances on par with
the state of the art, and improves over it on three of the four datasets, which
have in common the use of sparse features.
|
Learning the Kernel for Classification and Regression | We investigate a series of learning kernel problems with polynomial
combinations of base kernels, which will help us solve regression and
classification problems. We also perform some numerical experiments of
polynomial kernels with regression and classification tasks on different
datasets.
|
Learning in the Machine: the Symmetries of the Deep Learning Channel | In a physical neural system, learning rules must be local both in space and
time. In order for learning to occur, non-local information must be
communicated to the deep synapses through a communication channel, the deep
learning channel. We identify several possible architectures for this learning
channel (Bidirectional, Conjoined, Twin, Distinct) and six symmetry challenges:
1) symmetry of architectures; 2) symmetry of weights; 3) symmetry of neurons;
4) symmetry of derivatives; 5) symmetry of processing; and 6) symmetry of
learning rules. Random backpropagation (RBP) addresses the second and third
symmetry, and some of its variations, such as skipped RBP (SRBP) address the
first and the fourth symmetry. Here we address the last two desirable
symmetries showing through simulations that they can be achieved and that the
learning channel is particularly robust to symmetry variations. Specifically,
random backpropagation and its variations can be performed with the same
non-linear neurons used in the main input-output forward channel, and the
connections in the learning channel can be adapted using the same algorithm
used in the forward channel, removing the need for any specialized hardware in
the learning channel. Finally, we provide mathematical results in simple cases
showing that the learning equations in the forward and backward channels
converge to fixed points, for almost any initial conditions. In symmetric
architectures, if the weights in both channels are small at initialization,
adaptation in both channels leads to weights that are essentially symmetric
during and after learning. Biological connections are discussed.
|
Obtaining Accurate Probabilistic Causal Inference by Post-Processing
Calibration | Discovery of an accurate causal Bayesian network structure from observational
data can be useful in many areas of science. Often the discoveries are made
under uncertainty, which can be expressed as probabilities. To guide the use of
such discoveries, including directing further investigation, it is important
that those probabilities be well-calibrated. In this paper, we introduce a
novel framework to derive calibrated probabilities of causal relationships from
observational data. The framework consists of three components: (1) an
approximate method for generating initial probability estimates of the edge
types for each pair of variables, (2) the availability of a relatively small
number of the causal relationships in the network for which the truth status is
known, which we call a calibration training set, and (3) a calibration method
for using the approximate probability estimates and the calibration training
set to generate calibrated probabilities for the many remaining pairs of
variables. We also introduce a new calibration method based on a shallow neural
network. Our experiments on simulated data support that the proposed approach
improves the calibration of causal edge predictions. The results also support
that the approach often improves the precision and recall of predictions.
|
Least-Squares Temporal Difference Learning for the Linear Quadratic
Regulator | Reinforcement learning (RL) has been successfully used to solve many
continuous control tasks. Despite its impressive results however, fundamental
questions regarding the sample complexity of RL on continuous problems remain
open. We study the performance of RL in this setting by considering the
behavior of the Least-Squares Temporal Difference (LSTD) estimator on the
classic Linear Quadratic Regulator (LQR) problem from optimal control. We give
the first finite-time analysis of the number of samples needed to estimate the
value function for a fixed static state-feedback policy to within
$\varepsilon$-relative error. In the process of deriving our result, we give a
general characterization for when the minimum eigenvalue of the empirical
covariance matrix formed along the sample path of a fast-mixing stochastic
process concentrates above zero, extending a result by Koltchinskii and
Mendelson in the independent covariates setting. Finally, we provide
experimental evidence indicating that our analysis correctly captures the
qualitative behavior of LSTD on several LQR instances.
|
Dropout Feature Ranking for Deep Learning Models | Deep neural networks (DNNs) achieve state-of-the-art results in a variety of
domains. Unfortunately, DNNs are notorious for their non-interpretability, and
thus limit their applicability in hypothesis-driven domains such as biology and
healthcare. Moreover, in the resource-constraint setting, it is critical to
design tests relying on fewer more informative features leading to high
accuracy performance within reasonable budget. We aim to close this gap by
proposing a new general feature ranking method for deep learning. We show that
our simple yet effective method performs on par or compares favorably to eight
strawman, classical and deep-learning feature ranking methods in two
simulations and five very different datasets on tasks ranging from
classification to regression, in both static and time series scenarios. We also
illustrate the use of our method on a drug response dataset and show that it
identifies genes relevant to the drug-response.
|
A short variational proof of equivalence between policy gradients and
soft Q learning | Two main families of reinforcement learning algorithms, Q-learning and policy
gradients, have recently been proven to be equivalent when using a softmax
relaxation on one part, and an entropic regularization on the other. We relate
this result to the well-known convex duality of Shannon entropy and the softmax
function. Such a result is also known as the Donsker-Varadhan formula. This
provides a short proof of the equivalence. We then interpret this duality
further, and use ideas of convex analysis to prove a new policy inequality
relative to soft Q-learning.
|
Variational Autoencoders for Learning Latent Representations of Speech
Emotion: A Preliminary Study | Learning the latent representation of data in unsupervised fashion is a very
interesting process that provides relevant features for enhancing the
performance of a classifier. For speech emotion recognition tasks, generating
effective features is crucial. Currently, handcrafted features are mostly used
for speech emotion recognition, however, features learned automatically using
deep learning have shown strong success in many problems, especially in image
processing. In particular, deep generative models such as Variational
Autoencoders (VAEs) have gained enormous success for generating features for
natural images. Inspired by this, we propose VAEs for deriving the latent
representation of speech signals and use this representation to classify
emotions. To the best of our knowledge, we are the first to propose VAEs for
speech emotion classification. Evaluations on the IEMOCAP dataset demonstrate
that features learned by VAEs can produce state-of-the-art results for speech
emotion classification.
|
Query-limited Black-box Attacks to Classifiers | We study black-box attacks on machine learning classifiers where each query
to the model incurs some cost or risk of detection to the adversary. We focus
explicitly on minimizing the number of queries as a major objective.
Specifically, we consider the problem of attacking machine learning classifiers
subject to a budget of feature modification cost while minimizing the number of
queries, where each query returns only a class and confidence score. We
describe an approach that uses Bayesian optimization to minimize the number of
queries, and find that the number of queries can be reduced to approximately
one tenth of the number needed through a random strategy for scenarios where
the feature modification cost budget is low.
|
Online Forecasting Matrix Factorization | In this paper the problem of forecasting high dimensional time series is
considered. Such time series can be modeled as matrices where each column
denotes a measurement. In addition, when missing values are present, low rank
matrix factorization approaches are suitable for predicting future values. This
paper formally defines and analyzes the forecasting problem in the online
setting, i.e. where the data arrives as a stream and only a single pass is
allowed. We present and analyze novel matrix factorization techniques which can
learn low-dimensional embeddings effectively in an online manner. Based on
these embeddings a recursive minimum mean square error estimator is derived,
which learns an autoregressive model on them. Experiments with two real
datasets with tens of millions of measurements show the benefits of the
proposed approach.
|
An Approximate Bayesian Long Short-Term Memory Algorithm for Outlier
Detection | Long Short-Term Memory networks trained with gradient descent and
back-propagation have received great success in various applications. However,
point estimation of the weights of the networks is prone to over-fitting
problems and lacks important uncertainty information associated with the
estimation. However, exact Bayesian neural network methods are intractable and
non-applicable for real-world applications. In this study, we propose an
approximate estimation of the weights uncertainty using Ensemble Kalman Filter,
which is easily scalable to a large number of weights. Furthermore, we optimize
the covariance of the noise distribution in the ensemble update step using
maximum likelihood estimation. To assess the proposed algorithm, we apply it to
outlier detection in five real-world events retrieved from the Twitter
platform.
|
Transfer Regression via Pairwise Similarity Regularization | Transfer learning methods address the situation where little labeled training
data from the "target" problem exists, but much training data from a related
"source" domain is available. However, the overwhelming majority of transfer
learning methods are designed for simple settings where the source and target
predictive functions are almost identical, limiting the applicability of
transfer learning methods to real world data. We propose a novel, weaker,
property of the source domain that can be transferred even when the source and
target predictive functions diverge. Our method assumes the source and target
functions share a Pairwise Similarity property, where if the source function
makes similar predictions on a pair of instances, then so will the target
function. We propose Pairwise Similarity Regularization Transfer, a flexible
graph-based regularization framework which can incorporate this modeling
assumption into standard supervised learning algorithms. We show how users can
encode domain knowledge into our regularizer in the form of spatial continuity,
pairwise "similarity constraints" and how our method can be scaled to large
data sets using the Nystrom approximation. Finally, we present positive and
negative results on real and synthetic data sets and discuss when our Pairwise
Similarity transfer assumption seems to hold in practice.
|
Neural Network Multitask Learning for Traffic Flow Forecasting | Traditional neural network approaches for traffic flow forecasting are
usually single task learning (STL) models, which do not take advantage of the
information provided by related tasks. In contrast to STL, multitask learning
(MTL) has the potential to improve generalization by transferring information
in training signals of extra tasks. In this paper, MTL based neural networks
are used for traffic flow forecasting. For neural network MTL, a
backpropagation (BP) network is constructed by incorporating traffic flows at
several contiguous time instants into an output layer. Nodes in the output
layer can be seen as outputs of different but closely related STL tasks.
Comprehensive experiments on urban vehicular traffic flow data and comparisons
with STL show that MTL in BP neural networks is a promising and effective
approach for traffic flow forecasting.
|
Weighted Data Normalization Based on Eigenvalues for Artificial Neural
Network Classification | Artificial neural network (ANN) is a very useful tool in solving learning
problems. Boosting the performances of ANN can be mainly concluded from two
aspects: optimizing the architecture of ANN and normalizing the raw data for
ANN. In this paper, a novel method which improves the effects of ANN by
preprocessing the raw data is proposed. It totally leverages the fact that
different features should play different roles. The raw data set is firstly
preprocessed by principle component analysis (PCA), and then its principle
components are weighted by their corresponding eigenvalues. Several aspects of
analysis are carried out to analyze its theory and the applicable occasions.
Three classification problems are launched by an active learning algorithm to
verify the proposed method. From the empirical results, conclusion comes to the
fact that the proposed method can significantly improve the performance of ANN.
|
Bayesian Nonparametric Causal Inference: Information Rates and Learning
Algorithms | We investigate the problem of estimating the causal effect of a treatment on
individual subjects from observational data, this is a central problem in
various application domains, including healthcare, social sciences, and online
advertising. Within the Neyman Rubin potential outcomes model, we use the
Kullback Leibler (KL) divergence between the estimated and true distributions
as a measure of accuracy of the estimate, and we define the information rate of
the Bayesian causal inference procedure as the (asymptotic equivalence class of
the) expected value of the KL divergence between the estimated and true
distributions as a function of the number of samples. Using Fano method, we
establish a fundamental limit on the information rate that can be achieved by
any Bayesian estimator, and show that this fundamental limit is independent of
the selection bias in the observational data. We characterize the Bayesian
priors on the potential (factual and counterfactual) outcomes that achieve the
optimal information rate. As a consequence, we show that a particular class of
priors that have been widely used in the causal inference literature cannot
achieve the optimal information rate. On the other hand, a broader class of
priors can achieve the optimal information rate. We go on to propose a prior
adaptation procedure (which we call the information based empirical Bayes
procedure) that optimizes the Bayesian prior by maximizing an information
theoretic criterion on the recovered causal effects rather than maximizing the
marginal likelihood of the observed (factual) data. Building on our analysis,
we construct an information optimal Bayesian causal inference algorithm.
|
Spurious Local Minima are Common in Two-Layer ReLU Neural Networks | We consider the optimization problem associated with training simple ReLU
neural networks of the form $\mathbf{x}\mapsto
\sum_{i=1}^{k}\max\{0,\mathbf{w}_i^\top \mathbf{x}\}$ with respect to the
squared loss. We provide a computer-assisted proof that even if the input
distribution is standard Gaussian, even if the dimension is arbitrarily large,
and even if the target values are generated by such a network, with orthonormal
parameter vectors, the problem can still have spurious local minima once $6\le
k\le 20$. By a concentration of measure argument, this implies that in high
input dimensions, \emph{nearly all} target networks of the relevant sizes lead
to spurious local minima. Moreover, we conduct experiments which show that the
probability of hitting such local minima is quite high, and increasing with the
network size. On the positive side, mild over-parameterization appears to
drastically reduce such local minima, indicating that an over-parameterization
assumption is necessary to get a positive result in this setting.
|
Mean Field Residual Networks: On the Edge of Chaos | We study randomly initialized residual networks using mean field theory and
the theory of difference equations. Classical feedforward neural networks, such
as those with tanh activations, exhibit exponential behavior on the average
when propagating inputs forward or gradients backward. The exponential forward
dynamics causes rapid collapsing of the input space geometry, while the
exponential backward dynamics causes drastic vanishing or exploding gradients.
We show, in contrast, that by adding skip connections, the network will,
depending on the nonlinearity, adopt subexponential forward and backward
dynamics, and in many cases in fact polynomial. The exponents of these
polynomials are obtained through analytic methods and proved and verified
empirically to be correct. In terms of the "edge of chaos" hypothesis, these
subexponential and polynomial laws allow residual networks to "hover over the
boundary between stability and chaos," thus preserving the geometry of the
input space and the gradient information flow. In our experiments, for each
activation function we study here, we initialize residual networks with
different hyperparameters and train them on MNIST. Remarkably, our
initialization time theory can accurately predict test time performance of
these networks, by tracking either the expected amount of gradient explosion or
the expected squared distance between the images of two input vectors.
Importantly, we show, theoretically as well as empirically, that common
initializations such as the Xavier or the He schemes are not optimal for
residual networks, because the optimal initialization variances depend on the
depth. Finally, we have made mathematical contributions by deriving several new
identities for the kernels of powers of ReLU functions by relating them to the
zeroth Bessel function of the second kind.
|
Learning to Run with Actor-Critic Ensemble | We introduce an Actor-Critic Ensemble(ACE) method for improving the
performance of Deep Deterministic Policy Gradient(DDPG) algorithm. At inference
time, our method uses a critic ensemble to select the best action from
proposals of multiple actors running in parallel. By having a larger candidate
set, our method can avoid actions that have fatal consequences, while staying
deterministic. Using ACE, we have won the 2nd place in NIPS'17 Learning to Run
competition, under the name of "Megvii-hzwer".
|
Leveraging Native Language Speech for Accent Identification using Deep
Siamese Networks | The problem of automatic accent identification is important for several
applications like speaker profiling and recognition as well as for improving
speech recognition systems. The accented nature of speech can be primarily
attributed to the influence of the speaker's native language on the given
speech recording. In this paper, we propose a novel accent identification
system whose training exploits speech in native languages along with the
accented speech. Specifically, we develop a deep Siamese network-based model
which learns the association between accented speech recordings and the native
language speech recordings. The Siamese networks are trained with i-vector
features extracted from the speech recordings using either an unsupervised
Gaussian mixture model (GMM) or a supervised deep neural network (DNN) model.
We perform several accent identification experiments using the CSLU Foreign
Accented English (FAE) corpus. In these experiments, our proposed approach
using deep Siamese networks yield significant relative performance improvements
of 15.4 percent on a 10-class accent identification task, over a baseline
DNN-based classification system that uses GMM i-vectors. Furthermore, we
present a detailed error analysis of the proposed accent identification system.
|
Kernel Regression with Sparse Metric Learning | Kernel regression is a popular non-parametric fitting technique. It aims at
learning a function which estimates the targets for test inputs as precise as
possible. Generally, the function value for a test input is estimated by a
weighted average of the surrounding training examples. The weights are
typically computed by a distance-based kernel function and they strongly depend
on the distances between examples. In this paper, we first review the latest
developments of sparse metric learning and kernel regression. Then a novel
kernel regression method involving sparse metric learning, which is called
kernel regression with sparse metric learning (KR$\_$SML), is proposed. The
sparse kernel regression model is established by enforcing a mixed $(2,1)$-norm
regularization over the metric matrix. It learns a Mahalanobis distance metric
by a gradient descent procedure, which can simultaneously conduct
dimensionality reduction and lead to good prediction results. Our work is the
first to combine kernel regression with sparse metric learning. To verify the
effectiveness of the proposed method, it is evaluated on 19 data sets for
regression. Furthermore, the new method is also applied to solving practical
problems of forecasting short-term traffic flows. In the end, we compare the
proposed method with other three related kernel regression methods on all test
data sets under two criterions. Experimental results show that the proposed
method is much more competitive.
|
Efficient Algorithms for t-distributed Stochastic Neighborhood Embedding | t-distributed Stochastic Neighborhood Embedding (t-SNE) is a method for
dimensionality reduction and visualization that has become widely popular in
recent years. Efficient implementations of t-SNE are available, but they scale
poorly to datasets with hundreds of thousands to millions of high dimensional
data-points. We present Fast Fourier Transform-accelerated Interpolation-based
t-SNE (FIt-SNE), which dramatically accelerates the computation of t-SNE. The
most time-consuming step of t-SNE is a convolution that we accelerate by
interpolating onto an equispaced grid and subsequently using the fast Fourier
transform to perform the convolution. We also optimize the computation of input
similarities in high dimensions using multi-threaded approximate nearest
neighbors. We further present a modification to t-SNE called "late
exaggeration," which allows for easier identification of clusters in t-SNE
embeddings. Finally, for datasets that cannot be loaded into the memory, we
present out-of-core randomized principal component analysis (oocPCA), so that
the top principal components of a dataset can be computed without ever fully
loading the matrix, hence allowing for t-SNE of large datasets to be computed
on resource-limited machines.
|
Stochastic Multi-armed Bandits in Constant Space | We consider the stochastic bandit problem in the sublinear space setting,
where one cannot record the win-loss record for all $K$ arms. We give an
algorithm using $O(1)$ words of space with regret \[
\sum_{i=1}^{K}\frac{1}{\Delta_i}\log \frac{\Delta_i}{\Delta}\log T \] where
$\Delta_i$ is the gap between the best arm and arm $i$ and $\Delta$ is the gap
between the best and the second-best arms. If the rewards are bounded away from
$0$ and $1$, this is within an $O(\log 1/\Delta)$ factor of the optimum regret
possible without space constraints.
|
Null Dynamical State Models of Human Cognitive Dysfunction | The hard problem in artificial intelligence asks how the shuffling of
syntactical symbols in a program can lead to systems which experience semantics
and qualia. We address this question in three stages. First, we introduce a new
class of human semantic symbols which appears when unexpected and drastic
environmental change causes humans to become surprised, confused, uncertain,
and in extreme cases, unresponsive, passive and dysfunctional. For this class
of symbols, pre-learned programs become inoperative so these syntactical
programs cannot be the source of experienced qualia. Second, we model the
dysfunctional human response to a radically changed environment as being the
natural response of any learning machine facing novel inputs from well outside
its previous training set. In this situation, learning machines are unable to
extract information from their input and will typically enter a dynamical state
characterized by null outputs and a lack of response. This state immediately
predicts and explains the characteristics of the semantic experiences of humans
in similar circumstances. In the third stage, we consider learning machines
trained to implement multiple functions in simple sequential programs using
environmental data to specify subroutine names, control flow instructions,
memory calls, and so on. Drastic change in any of these environmental inputs
can again lead to inoperative programs. By examining changes specific to people
or locations we can model human cognitive symbols featuring these dependencies,
such as attachment and grief. Our approach links known dynamical machines
states with human qualia and thus offers new insight into the hard problem of
artificial intelligence.
|
Neural Collaborative Autoencoder | In recent years, deep neural networks have yielded state-of-the-art
performance on several tasks. Although some recent works have focused on
combining deep learning with recommendation, we highlight three issues of
existing models. First, these models cannot work on both explicit and implicit
feedback, since the network structures are specially designed for one
particular case. Second, due to the difficulty on training deep neural
networks, existing explicit models do not fully exploit the expressive
potential of deep learning. Third, neural network models are easier to overfit
on the implicit setting than shallow models. To tackle these issues, we present
a generic recommender framework called Neural Collaborative Autoencoder (NCAE)
to perform collaborative filtering, which works well for both explicit feedback
and implicit feedback. NCAE can effectively capture the subtle hidden
relationships between interactions via a non-linear matrix factorization
process. To optimize the deep architecture of NCAE, we develop a three-stage
pre-training mechanism that combines supervised and unsupervised feature
learning. Moreover, to prevent overfitting on the implicit setting, we propose
an error reweighting module and a sparsity-aware data-augmentation strategy.
Extensive experiments on three real-world datasets demonstrate that NCAE can
significantly advance the state-of-the-art.
|
On Connecting Stochastic Gradient MCMC and Differential Privacy | Significant success has been realized recently on applying machine learning
to real-world applications. There have also been corresponding concerns on the
privacy of training data, which relates to data security and confidentiality
issues. Differential privacy provides a principled and rigorous privacy
guarantee on machine learning models. While it is common to design a model
satisfying a required differential-privacy property by injecting noise, it is
generally hard to balance the trade-off between privacy and utility. We show
that stochastic gradient Markov chain Monte Carlo (SG-MCMC) -- a class of
scalable Bayesian posterior sampling algorithms proposed recently -- satisfies
strong differential privacy with carefully chosen step sizes. We develop theory
on the performance of the proposed differentially-private SG-MCMC method. We
conduct experiments to support our analysis and show that a standard SG-MCMC
sampler without any modification (under a default setting) can reach
state-of-the-art performance in terms of both privacy and utility on Bayesian
learning.
|
SAGA: A Submodular Greedy Algorithm For Group Recommendation | In this paper, we propose a unified framework and an algorithm for the
problem of group recommendation where a fixed number of items or alternatives
can be recommended to a group of users. The problem of group recommendation
arises naturally in many real world contexts, and is closely related to the
budgeted social choice problem studied in economics. We frame the group
recommendation problem as choosing a subgraph with the largest group consensus
score in a completely connected graph defined over the item affinity matrix. We
propose a fast greedy algorithm with strong theoretical guarantees, and show
that the proposed algorithm compares favorably to the state-of-the-art group
recommendation algorithms according to commonly used relevance and coverage
performance measures on benchmark dataset.
|
Strongly Hierarchical Factorization Machines and ANOVA Kernel Regression | High-order parametric models that include terms for feature interactions are
applied to various data mining tasks, where ground truth depends on
interactions of features. However, with sparse data, the high- dimensional
parameters for feature interactions often face three issues: expensive
computation, difficulty in parameter estimation and lack of structure. Previous
work has proposed approaches which can partially re- solve the three issues. In
particular, models with factorized parameters (e.g. Factorization Machines) and
sparse learning algorithms (e.g. FTRL-Proximal) can tackle the first two issues
but fail to address the third. Regarding to unstructured parameters,
constraints or complicated regularization terms are applied such that
hierarchical structures can be imposed. However, these methods make the
optimization problem more challenging. In this work, we propose Strongly
Hierarchical Factorization Machines and ANOVA kernel regression where all the
three issues can be addressed without making the optimization problem more
difficult. Experimental results show the proposed models significantly
outperform the state-of-the-art in two data mining tasks: cold-start user
response time prediction and stock volatility prediction.
|
The Robust Manifold Defense: Adversarial Training using Generative
Models | We propose a new type of attack for finding adversarial examples for image
classifiers. Our method exploits spanners, i.e. deep neural networks whose
input space is low-dimensional and whose output range approximates the set of
images of interest. Spanners may be generators of GANs or decoders of VAEs. The
key idea in our attack is to search over latent code pairs to find ones that
generate nearby images with different classifier outputs. We argue that our
attack is stronger than searching over perturbations of real images. Moreover,
we show that our stronger attack can be used to reduce the accuracy of
Defense-GAN to 3\%, resolving an open problem from the well-known paper by
Athalye et al. We combine our attack with normal adversarial training to obtain
the most robust known MNIST classifier, significantly improving the state of
the art against PGD attacks. Our formulation involves solving a min-max
problem, where the min player sets the parameters of the classifier and the max
player is running our attack, and is thus searching for adversarial examples in
the {\em low-dimensional} input space of the spanner.
All code and models are available at
\url{https://github.com/ajiljalal/manifold-defense.git}
|
Algorithmic Regularization in Over-parameterized Matrix Sensing and
Neural Networks with Quadratic Activations | We show that the gradient descent algorithm provides an implicit
regularization effect in the learning of over-parameterized matrix
factorization models and one-hidden-layer neural networks with quadratic
activations. Concretely, we show that given $\tilde{O}(dr^{2})$ random linear
measurements of a rank $r$ positive semidefinite matrix $X^{\star}$, we can
recover $X^{\star}$ by parameterizing it by $UU^\top$ with $U\in \mathbb
R^{d\times d}$ and minimizing the squared loss, even if $r \ll d$. We prove
that starting from a small initialization, gradient descent recovers
$X^{\star}$ in $\tilde{O}(\sqrt{r})$ iterations approximately. The results
solve the conjecture of Gunasekar et al.'17 under the restricted isometry
property. The technique can be applied to analyzing neural networks with
one-hidden-layer quadratic activations with some technical modifications.
|
Who is Smarter? Intelligence Measure of Learning-based Cognitive Radios | Cognitive radio (CR) is considered as a key enabling technology for dynamic
spectrum access to improve spectrum efficiency. Although the CR concept was
invented with the core idea of realizing cognition, the research on measuring
CR cognitive capabilities and intelligence is largely open. Deriving the
intelligence measure of CR not only can lead to the development of new CR
technologies, but also makes it possible to better configure the networks by
integrating CRs with different cognitive capabilities. In this paper, for the
first time, we propose a data-driven methodology to quantitatively measure the
intelligence factors of the CR with learning capabilities. The basic idea of
our methodology is to run various tests on the CR in different spectrum
environments under different settings and obtain various performance data on
different metrics. Then we apply factor analysis on the performance data to
identify and quantize the intelligence factors and cognitive capabilities of
the CR. More specifically, we present a case study consisting of 144 different
types of CRs. The CRs are different in terms of learning-based dynamic spectrum
access strategies, number of sensors, sensing accuracy, processing speed, and
algorithmic complexity. Five intelligence factors are identified for the CRs
through our data analysis.We show that these factors comply well with the
nature of the tested CRs, which validates the proposed intelligence measure
methodology.
|
Building Robust Deep Neural Networks for Road Sign Detection | Deep Neural Networks are built to generalize outside of training set in mind
by using techniques such as regularization, early stopping and dropout. But
considerations to make them more resilient to adversarial examples are rarely
taken. As deep neural networks become more prevalent in mission-critical and
real-time systems, miscreants start to attack them by intentionally making deep
neural networks to misclassify an object of one type to be seen as another
type. This can be catastrophic in some scenarios where the classification of a
deep neural network can lead to a fatal decision by a machine. In this work, we
used GTSRB dataset to craft adversarial samples by Fast Gradient Sign Method
and Jacobian Saliency Method, used those crafted adversarial samples to attack
another Deep Convolutional Neural Network and built the attacked network to be
more resilient against adversarial attacks by making it more robust by
Defensive Distillation and Adversarial Training
|
Predicting protein-protein interactions based on rotation of proteins in
3D-space | Protein-Protein Interactions (PPIs) perform essential roles in biological
functions. Although some experimental techniques have been developed to detect
PPIs, they suffer from high false positive and high false negative rates.
Consequently, efforts have been devoted during recent years to develop
computational approaches to predict the interactions utilizing various sources
of information. Therefore, a unique category of prediction approaches has been
devised which is based on the protein sequence information. However, finding an
appropriate feature encoding to characterize the sequence of proteins is a
major challenge in such methods. In presented work, a sequence based method is
proposed to predict protein-protein interactions using N-Gram encoding
approaches to describe amino acids and a Relaxed Variable Kernel Density
Estimator (RVKDE) as a machine learning tool. Moreover, since proteins can
rotate in 3D-space, amino acid compositions have been considered with
"undirected" property which leads to reduce dimensions of the vector space. The
results show that our proposed method achieves the superiority of prediction
performance with improving an F-measure of 2.5% on Human Protein Reference
Dataset (HPRD).
|
Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization
properties of Entropy-SGD and data-dependent priors | We show that Entropy-SGD (Chaudhari et al., 2017), when viewed as a learning
algorithm, optimizes a PAC-Bayes bound on the risk of a Gibbs (posterior)
classifier, i.e., a randomized classifier obtained by a risk-sensitive
perturbation of the weights of a learned classifier. Entropy-SGD works by
optimizing the bound's prior, violating the hypothesis of the PAC-Bayes theorem
that the prior is chosen independently of the data. Indeed, available
implementations of Entropy-SGD rapidly obtain zero training error on random
labels and the same holds of the Gibbs posterior. In order to obtain a valid
generalization bound, we rely on a result showing that data-dependent priors
obtained by stochastic gradient Langevin dynamics (SGLD) yield valid PAC-Bayes
bounds provided the target distribution of SGLD is {\epsilon}-differentially
private. We observe that test error on MNIST and CIFAR10 falls within the
(empirically nonvacuous) risk bounds computed under the assumption that SGLD
reaches stationarity. In particular, Entropy-SGLD can be configured to yield
relatively tight generalization bounds and still fit real labels, although
these same settings do not obtain state-of-the-art performance.
|
IHT dies hard: Provable accelerated Iterative Hard Thresholding | We study --both in theory and practice-- the use of momentum motions in
classic iterative hard thresholding (IHT) methods. By simply modifying plain
IHT, we investigate its convergence behavior on convex optimization criteria
with non-convex constraints, under standard assumptions. In diverse scenaria,
we observe that acceleration in IHT leads to significant improvements, compared
to state of the art projected gradient descent and Frank-Wolfe variants. As a
byproduct of our inspection, we study the impact of selecting the momentum
parameter: similar to convex settings, two modes of behavior are observed
--"rippling" and linear-- depending on the level of momentum.
|
RLlib: Abstractions for Distributed Reinforcement Learning | Reinforcement learning (RL) algorithms involve the deep nesting of highly
irregular computation patterns, each of which typically exhibits opportunities
for distributed computation. We argue for distributing RL components in a
composable way by adapting algorithms for top-down hierarchical control,
thereby encapsulating parallelism and resource requirements within
short-running compute tasks. We demonstrate the benefits of this principle
through RLlib: a library that provides scalable software primitives for RL.
These primitives enable a broad range of algorithms to be implemented with high
performance, scalability, and substantial code reuse. RLlib is available at
https://rllib.io/.
|
Horn-ICE Learning for Synthesizing Invariants and Contracts | We design learning algorithms for synthesizing invariants using Horn
implication counterexamples (Horn-ICE), extending the ICE-learning model. In
particular, we describe a decision-tree learning algorithm that learns from
Horn-ICE samples, works in polynomial time, and uses statistical heuristics to
learn small trees that satisfy the samples. Since most verification proofs can
be modeled using Horn clauses, Horn-ICE learning is a more robust technique to
learn inductive annotations that prove programs correct. Our experiments show
that an implementation of our algorithm is able to learn adequate inductive
invariants and contracts efficiently for a variety of sequential and concurrent
programs.
|
Sketching for Kronecker Product Regression and P-splines | TensorSketch is an oblivious linear sketch introduced in Pagh'13 and later
used in Pham, Pagh'13 in the context of SVMs for polynomial kernels. It was
shown in Avron, Nguyen, Woodruff'14 that TensorSketch provides a subspace
embedding, and therefore can be used for canonical correlation analysis, low
rank approximation, and principal component regression for the polynomial
kernel. We take TensorSketch outside of the context of polynomials kernels, and
show its utility in applications in which the underlying design matrix is a
Kronecker product of smaller matrices. This allows us to solve Kronecker
product regression and non-negative Kronecker product regression, as well as
regularized spline regression. Our main technical result is then in extending
TensorSketch to other norms. That is, TensorSketch only provides input sparsity
time for Kronecker product regression with respect to the $2$-norm. We show how
to solve Kronecker product regression with respect to the $1$-norm in time
sublinear in the time required for computing the Kronecker product, as well as
for more general $p$-norms.
|
Robust Loss Functions under Label Noise for Deep Neural Networks | In many applications of classifier learning, training data suffers from label
noise. Deep networks are learned using huge training data where the problem of
noisy labels is particularly relevant. The current techniques proposed for
learning deep networks under label noise focus on modifying the network
architecture and on algorithms for estimating true labels from noisy labels. An
alternate approach would be to look for loss functions that are inherently
noise-tolerant. For binary classification there exist theoretical results on
loss functions that are robust to label noise. In this paper, we provide some
sufficient conditions on a loss function so that risk minimization under that
loss function would be inherently tolerant to label noise for multiclass
classification problems. These results generalize the existing results on
noise-tolerant loss functions for binary classification. We study some of the
widely used loss functions in deep networks and show that the loss function
based on mean absolute value of error is inherently robust to label noise. Thus
standard back propagation is enough to learn the true classifier even under
label noise. Through experiments, we illustrate the robustness of risk
minimization with such loss functions for learning neural networks.
|
Exploring the Space of Black-box Attacks on Deep Neural Networks | Existing black-box attacks on deep neural networks (DNNs) so far have largely
focused on transferability, where an adversarial instance generated for a
locally trained model can "transfer" to attack other learning models. In this
paper, we propose novel Gradient Estimation black-box attacks for adversaries
with query access to the target model's class probabilities, which do not rely
on transferability. We also propose strategies to decouple the number of
queries required to generate each adversarial sample from the dimensionality of
the input. An iterative variant of our attack achieves close to 100%
adversarial success rates for both targeted and untargeted attacks on DNNs. We
carry out extensive experiments for a thorough comparative evaluation of
black-box attacks and show that the proposed Gradient Estimation attacks
outperform all transferability based black-box attacks we tested on both MNIST
and CIFAR-10 datasets, achieving adversarial success rates similar to well
known, state-of-the-art white-box attacks. We also apply the Gradient
Estimation attacks successfully against a real-world Content Moderation
classifier hosted by Clarifai. Furthermore, we evaluate black-box attacks
against state-of-the-art defenses. We show that the Gradient Estimation attacks
are very effective even against these defenses.
|
Tensor Regression Networks with various Low-Rank Tensor Approximations | Tensor regression networks achieve high compression rate of neural networks
while having slight impact on performances. They do so by imposing low tensor
rank structure on the weight matrices of fully connected layers. In recent
years, tensor regression networks have been investigated from the perspective
of their compressive power, however, the regularization effect of enforcing
low-rank tensor structure has not been investigated enough. We study tensor
regression networks using various low-rank tensor approximations, aiming to
compare the compressive and regularization power of different low-rank
constraints. We evaluate the compressive and regularization performances of the
proposed model with both deep and shallow convolutional neural networks. The
outcome of our experiment suggests the superiority of Global Average Pooling
Layer over Tensor Regression Layer when applied to deep convolutional neural
network with CIFAR-10 dataset. On the contrary, shallow convolutional neural
networks with tensor regression layer and dropout achieved lower test error
than both Global Average Pooling and fully-connected layer with dropout
function when trained with a small number of samples.
|
Co-Morbidity Exploration on Wearables Activity Data Using Unsupervised
Pre-training and Multi-Task Learning | Physical activity and sleep play a major role in the prevention and
management of many chronic conditions. It is not a trivial task to understand
their impact on chronic conditions. Currently, data from electronic health
records (EHRs), sleep lab studies, and activity/sleep logs are used. The rapid
increase in the popularity of wearable health devices provides a significant
new data source, making it possible to track the user's lifestyle real-time
through web interfaces, both to consumer as well as their healthcare provider,
potentially. However, at present there is a gap between lifestyle data (e.g.,
sleep, physical activity) and clinical outcomes normally captured in EHRs. This
is a critical barrier for the use of this new source of signal for healthcare
decision making. Applying deep learning to wearables data provides a new
opportunity to overcome this barrier.
To address the problem of the unavailability of clinical data from a major
fraction of subjects and unrepresentative subject populations, we propose a
novel unsupervised (task-agnostic) time-series representation learning
technique called act2vec. act2vec learns useful features by taking into account
the co-occurrence of activity levels along with periodicity of human activity
patterns. The learned representations are then exploited to boost the
performance of disorder-specific supervised learning models. Furthermore, since
many disorders are often related to each other, a phenomenon referred to as
co-morbidity, we use a multi-task learning framework for exploiting the shared
structure of disorder inducing life-style choices partially captured in the
wearables data. Empirical evaluation using actigraphy data from 4,124 subjects
shows that our proposed method performs and generalizes substantially better
than the conventional time-series symbolic representational methods and
task-specific deep learning models.
|
DeepIEP: a Peptide Sequence Model of Isoelectric Point (IEP/pI) using
Recurrent Neural Networks (RNNs) | The isoelectric point (IEP or pI) is the pH where the net charge on the
molecular ensemble of peptides and proteins is zero. This physical-chemical
property is dependent on protonable/deprotonable sidechains and their pKa
values. Here an pI prediction model is trained from a database of peptide
sequences and pIs using a recurrent neural network (RNN) with long short-term
memory (LSTM) cells. The trained model obtains an RMSE and R$^2$ of 0.28 and
0.95 for the external test set. The model is not based on pKa values, but
prediction of constructed test sequences show similar rankings as already known
pKa values. The prediction depends mostly on the existence of known acidic and
basic amino acids with fine-adjusted based on the neighboring sequence and
position of the charged amino acids in the peptide chain.
|
An empirical evaluation for the intrusion detection features based on
machine learning and feature selection methods | Despite the great developments in information technology, particularly the
Internet, computer networks, global information exchange, and its positive
impact in all areas of daily life, it has also contributed to the development
of penetration and intrusion which forms a high risk to the security of
information organizations, government agencies, and causes large economic
losses. There are many techniques designed for protection such as firewall and
intrusion detection systems (IDS). IDS is a set of software and/or hardware
techniques used to detect hacker's activities in computer systems. Two types of
anomalies are used in IDS to detect intrusive activities different from normal
user behavior. Misuse relies on the knowledge base that contains all known
attack techniques and intrusion is discovered through research in this
knowledge base. Artificial intelligence techniques have been introduced to
improve the performance of these systems. The importance of IDS is to identify
unauthorized access attempting to compromise confidentiality, integrity or
availability of the computer network. This paper investigates the Intrusion
Detection (ID) problem using three machine learning algorithms namely, BayesNet
algorithm, Multi-Layer Perceptron (MLP), and Support Vector Machine (SVM). The
algorithms are applied on a real, Management Information Based (MIB) dataset
that is collected from real life environment. To enhance the detection process
accuracy, a set of feature selection approaches is used; Infogain (IG), ReleifF
(RF), and Genetic Search (GS). Our experiments show that the three feature
selection methods have enhanced the classification performance. GS with
bayesNet, MLP and SVM give high accuracy rates, more specifically the BayesNet
with the GS accuracy rate is 99.9%.
|
On Convergence of some Gradient-based Temporal-Differences Algorithms
for Off-Policy Learning | We consider off-policy temporal-difference (TD) learning methods for policy
evaluation in Markov decision processes with finite spaces and discounted
reward criteria, and we present a collection of convergence results for several
gradient-based TD algorithms with linear function approximation. The algorithms
we analyze include: (i) two basic forms of two-time-scale gradient-based TD
algorithms, which we call GTD and which minimize the mean squared projected
Bellman error using stochastic gradient-descent; (ii) their "robustified"
biased variants; (iii) their mirror-descent versions which combine the
mirror-descent idea with TD learning; and (iv) a single-time-scale version of
GTD that solves minimax problems formulated for approximate policy evaluation.
We derive convergence results for three types of stepsizes: constant
stepsize, slowly diminishing stepsize, as well as the standard type of
diminishing stepsize with a square-summable condition. For the first two types
of stepsizes, we apply the weak convergence method from stochastic
approximation theory to characterize the asymptotic behavior of the algorithms,
and for the standard type of stepsize, we analyze the algorithmic behavior with
respect to a stronger mode of convergence, almost sure convergence. Our
convergence results are for the aforementioned TD algorithms with three general
ways of setting their $\lambda$-parameters: (i) state-dependent $\lambda$; (ii)
a recently proposed scheme of using history-dependent $\lambda$ to keep the
eligibility traces of the algorithms bounded while allowing for relatively
large values of $\lambda$; and (iii) a composite scheme of setting the
$\lambda$-parameters that combines the preceding two schemes and allows a
broader class of generalized Bellman operators to be used for approximate
policy evaluation with TD methods.
|
The information bottleneck and geometric clustering | The information bottleneck (IB) approach to clustering takes a joint
distribution $P\!\left(X,Y\right)$ and maps the data $X$ to cluster labels $T$
which retain maximal information about $Y$ (Tishby et al., 1999). This
objective results in an algorithm that clusters data points based upon the
similarity of their conditional distributions $P\!\left(Y\mid X\right)$. This
is in contrast to classic "geometric clustering'' algorithms such as $k$-means
and gaussian mixture models (GMMs) which take a set of observed data points
$\left\{ \mathbf{x}_{i}\right\} _{i=1:N}$ and cluster them based upon their
geometric (typically Euclidean) distance from one another. Here, we show how to
use the deterministic information bottleneck (DIB) (Strouse and Schwab, 2017),
a variant of IB, to perform geometric clustering, by choosing cluster labels
that preserve information about data point location on a smoothed dataset. We
also introduce a novel method to choose the number of clusters, based on
identifying solutions where the tradeoff between number of clusters used and
spatial information preserved is strongest. We apply this approach to a variety
of simple clustering problems, showing that DIB with our model selection
procedure recovers the generative cluster labels. We also show that, in
particular limits of our model parameters, clustering with DIB and IB is
equivalent to $k$-means and EM fitting of a GMM with hard and soft assignments,
respectively. Thus, clustering with (D)IB generalizes and provides an
information-theoretic perspective on these classic algorithms.
|
CNN Is All You Need | The Convolution Neural Network (CNN) has demonstrated the unique advantage in
audio, image and text learning; recently it has also challenged Recurrent
Neural Networks (RNNs) with long short-term memory cells (LSTM) in
sequence-to-sequence learning, since the computations involved in CNN are
easily parallelizable whereas those involved in RNN are mostly sequential,
leading to a performance bottleneck. However, unlike RNN, the native CNN lacks
the history sensitivity required for sequence transformation; therefore
enhancing the sequential order awareness, or position-sensitivity, becomes the
key to make CNN the general deep learning model. In this work we introduce an
extended CNN model with strengthen position-sensitivity, called PoseNet. A
notable feature of PoseNet is the asymmetric treatment of position information
in the encoder and the decoder. Experiments shows that PoseNet allows us to
improve the accuracy of CNN based sequence-to-sequence learning significantly,
achieving around 33-36 BLEU scores on the WMT 2014 English-to-German
translation task, and around 44-46 BLEU scores on the English-to-French
translation task.
|
Momentum and Stochastic Momentum for Stochastic Gradient, Newton,
Proximal Point and Subspace Descent Methods | In this paper we study several classes of stochastic optimization algorithms
enriched with heavy ball momentum. Among the methods studied are: stochastic
gradient descent, stochastic Newton, stochastic proximal point and stochastic
dual subspace ascent. This is the first time momentum variants of several of
these methods are studied. We choose to perform our analysis in a setting in
which all of the above methods are equivalent. We prove global nonassymptotic
linear convergence rates for all methods and various measures of success,
including primal function values, primal iterates (in L2 sense), and dual
function values. We also show that the primal iterates converge at an
accelerated linear rate in the L1 sense. This is the first time a linear rate
is shown for the stochastic heavy ball method (i.e., stochastic gradient
descent method with momentum). Under somewhat weaker conditions, we establish a
sublinear convergence rate for Cesaro averages of primal iterates. Moreover, we
propose a novel concept, which we call stochastic momentum, aimed at decreasing
the cost of performing the momentum step. We prove linear convergence of
several stochastic methods with stochastic momentum, and show that in some
sparse data regimes and for sufficiently small momentum parameters, these
methods enjoy better overall complexity than methods with deterministic
momentum. Finally, we perform extensive numerical testing on artificial and
real datasets, including data coming from average consensus problems.
|
Combining Representation Learning with Logic for Language Processing | The current state-of-the-art in many natural language processing and
automated knowledge base completion tasks is held by representation learning
methods which learn distributed vector representations of symbols via
gradient-based optimization. They require little or no hand-crafted features,
thus avoiding the need for most preprocessing steps and task-specific
assumptions. However, in many cases representation learning requires a large
amount of annotated training data to generalize well to unseen data. Such
labeled training data is provided by human annotators who often use formal
logic as the language for specifying annotations. This thesis investigates
different combinations of representation learning methods with logic for
reducing the need for annotated training data, and for improving
generalization.
|
Deep learning for universal linear embeddings of nonlinear dynamics | Identifying coordinate transformations that make strongly nonlinear dynamics
approximately linear is a central challenge in modern dynamical systems. These
transformations have the potential to enable prediction, estimation, and
control of nonlinear systems using standard linear theory. The Koopman operator
has emerged as a leading data-driven embedding, as eigenfunctions of this
operator provide intrinsic coordinates that globally linearize the dynamics.
However, identifying and representing these eigenfunctions has proven to be
mathematically and computationally challenging. This work leverages the power
of deep learning to discover representations of Koopman eigenfunctions from
trajectory data of dynamical systems. Our network is parsimonious and
interpretable by construction, embedding the dynamics on a low-dimensional
manifold that is of the intrinsic rank of the dynamics and parameterized by the
Koopman eigenfunctions. In particular, we identify nonlinear coordinates on
which the dynamics are globally linear using a modified auto-encoder. We also
generalize Koopman representations to include a ubiquitous class of systems
that exhibit continuous spectra, ranging from the simple pendulum to nonlinear
optics and broadband turbulence. Our framework parametrizes the continuous
frequency using an auxiliary network, enabling a compact and efficient
embedding at the intrinsic rank, while connecting our models to half a century
of asymptotics. In this way, we benefit from the power and generality of deep
learning, while retaining the physical interpretability of Koopman embeddings.
|
Learning More Universal Representations for Transfer-Learning | A representation is supposed universal if it encodes any element of the
visual world (e.g., objects, scenes) in any configuration (e.g., scale,
context). While not expecting pure universal representations, the goal in the
literature is to improve the universality level, starting from a representation
with a certain level. To do so, the state-of-the-art consists in learning
CNN-based representations on a diversified training problem (e.g., ImageNet
modified by adding annotated data). While it effectively increases
universality, such approach still requires a large amount of efforts to satisfy
the needs in annotated data. In this work, we propose two methods to improve
universality, but pay special attention to limit the need of annotated data. We
also propose a unified framework of the methods based on the diversifying of
the training problem. Finally, to better match Atkinson's cognitive study about
universal human representations, we proposed to rely on the transfer-learning
scheme as well as a new metric to evaluate universality. This latter, aims us
to demonstrates the interest of our methods on 10 target-problems, relating to
the classification task and a variety of visual domains.
|
Extrapolating Expected Accuracies for Large Multi-Class Problems | The difficulty of multi-class classification generally increases with the
number of classes. Using data from a subset of the classes, can we predict how
well a classifier will scale with an increased number of classes? Under the
assumptions that the classes are sampled identically and independently from a
population, and that the classifier is based on independently learned scoring
functions, we show that the expected accuracy when the classifier is trained on
k classes is the (k-1)st moment of a certain distribution that can be estimated
from data. We present an unbiased estimation method based on the theory, and
demonstrate its application on a facial recognition example.
|
Directional Statistics and Filtering Using libDirectional | In this paper, we present libDirectional, a MATLAB library for directional
statistics and directional estimation. It supports a variety of commonly used
distributions on the unit circle, such as the von Mises, wrapped normal, and
wrapped Cauchy distributions. Furthermore, various distributions on
higher-dimensional manifolds such as the unit hypersphere and the hypertorus
are available. Based on these distributions, several recursive filtering
algorithms in libDirectional allow estimation on these manifolds. The
functionality is implemented in a clear, well-documented, and object-oriented
structure that is both easy to use and easy to extend.
|
PixelSNAIL: An Improved Autoregressive Generative Model | Autoregressive generative models consistently achieve the best results in
density estimation tasks involving high dimensional data, such as images or
audio. They pose density estimation as a sequence modeling task, where a
recurrent neural network (RNN) models the conditional distribution over the
next element conditioned on all previous elements. In this paradigm, the
bottleneck is the extent to which the RNN can model long-range dependencies,
and the most successful approaches rely on causal convolutions, which offer
better access to earlier parts of the sequence than conventional RNNs. Taking
inspiration from recent work in meta reinforcement learning, where dealing with
long-range dependencies is also essential, we introduce a new generative model
architecture that combines causal convolutions with self attention. In this
note, we describe the resulting model and present state-of-the-art
log-likelihood results on CIFAR-10 (2.85 bits per dim) and $32 \times 32$
ImageNet (3.80 bits per dim). Our implementation is available at
https://github.com/neocxi/pixelsnail-public
|
Differentially Private Matrix Completion Revisited | We provide the first provably joint differentially private algorithm with
formal utility guarantees for the problem of user-level privacy-preserving
collaborative filtering. Our algorithm is based on the Frank-Wolfe method, and
it consistently estimates the underlying preference matrix as long as the
number of users $m$ is $\omega(n^{5/4})$, where $n$ is the number of items, and
each user provides her preference for at least $\sqrt{n}$ randomly selected
items. Along the way, we provide an optimal differentially private algorithm
for singular vector computation, based on the celebrated Oja's method, that
provides significant savings in terms of space and time while operating on
sparse matrices. We also empirically evaluate our algorithm on a suite of
datasets, and show that it consistently outperforms the state-of-the-art
private algorithms.
|
Automatic Analysis of EEGs Using Big Data and Hybrid Deep Learning
Architectures | Objective: A clinical decision support tool that automatically interprets
EEGs can reduce time to diagnosis and enhance real-time applications such as
ICU monitoring. Clinicians have indicated that a sensitivity of 95% with a
specificity below 5% was the minimum requirement for clinical acceptance. We
propose a highperformance classification system based on principles of big data
and machine learning. Methods: A hybrid machine learning system that uses
hidden Markov models (HMM) for sequential decoding and deep learning networks
for postprocessing is proposed. These algorithms were trained and evaluated
using the TUH EEG Corpus, which is the world's largest publicly available
database of clinical EEG data. Results: Our approach delivers a sensitivity
above 90% while maintaining a specificity below 5%. This system detects three
events of clinical interest: (1) spike and/or sharp waves, (2) periodic
lateralized epileptiform discharges, (3) generalized periodic epileptiform
discharges. It also detects three events used to model background noise: (1)
artifacts, (2) eye movement (3) background. Conclusions: A hybrid HMM/deep
learning system can deliver a low false alarm rate on EEG event detection,
making automated analysis a viable option for clinicians. Significance: The TUH
EEG Corpus enables application of highly data consumptive machine learning
algorithms to EEG analysis. Performance is approaching clinical acceptance for
real-time applications.
|
Deep Architectures for Automated Seizure Detection in Scalp EEGs | Automated seizure detection using clinical electroencephalograms is a
challenging machine learning problem because the multichannel signal often has
an extremely low signal to noise ratio. Events of interest such as seizures are
easily confused with signal artifacts (e.g, eye movements) or benign variants
(e.g., slowing). Commercially available systems suffer from unacceptably high
false alarm rates. Deep learning algorithms that employ high dimensional models
have not previously been effective due to the lack of big data resources. In
this paper, we use the TUH EEG Seizure Corpus to evaluate a variety of hybrid
deep structures including Convolutional Neural Networks and Long Short-Term
Memory Networks. We introduce a novel recurrent convolutional architecture that
delivers 30% sensitivity at 7 false alarms per 24 hours. We have also evaluated
our system on a held-out evaluation set based on the Duke University Seizure
Corpus and demonstrate that performance trends are similar to the TUH EEG
Seizure Corpus. This is a significant finding because the Duke corpus was
collected with different instrumentation and at different hospitals. Our work
shows that deep learning architectures that integrate spatial and temporal
contexts are critical to achieving state of the art performance and will enable
a new generation of clinically-acceptable technology.
|
Topic Compositional Neural Language Model | We propose a Topic Compositional Neural Language Model (TCNLM), a novel
method designed to simultaneously capture both the global semantic meaning and
the local word ordering structure in a document. The TCNLM learns the global
semantic coherence of a document via a neural topic model, and the probability
of each learned latent topic is further used to build a Mixture-of-Experts
(MoE) language model, where each expert (corresponding to one topic) is a
recurrent neural network (RNN) that accounts for learning the local structure
of a word sequence. In order to train the MoE model efficiently, a matrix
factorization method is applied, by extending each weight matrix of the RNN to
be an ensemble of topic-dependent weight matrices. The degree to which each
member of the ensemble is used is tied to the document-dependent probability of
the corresponding topics. Experimental results on several corpora show that the
proposed approach outperforms both a pure RNN-based model and other
topic-guided language models. Further, our model yields sensible topics, and
also has the capacity to generate meaningful sentences conditioned on given
topics.
|
Learning to Customize Network Security Rules | Security is a major concern for organizations who wish to leverage cloud
computing. In order to reduce security vulnerabilities, public cloud providers
offer firewall functionalities. When properly configured, a firewall protects
cloud networks from cyber-attacks. However, proper firewall configuration
requires intimate knowledge of the protected system, high expertise and
on-going maintenance.
As a result, many organizations do not use firewalls effectively, leaving
their cloud resources vulnerable. In this paper, we present a novel supervised
learning method, and prototype, which compute recommendations for firewall
rules. Recommendations are based on sampled network traffic meta-data (NetFlow)
collected from a public cloud provider. Labels are extracted from firewall
configurations deemed to be authored by experts. NetFlow is collected from
network routers, avoiding expensive collection from cloud VMs, as well as
relieving privacy concerns.
The proposed method captures network routines and dependencies between
resources and firewall configuration. The method predicts IPs to be allowed by
the firewall. A grouping algorithm is subsequently used to generate a
manageable number of IP ranges. Each range is a parameter for a firewall rule.
We present results of experiments on real data, showing ROC AUC of 0.92,
compared to 0.58 for an unsupervised baseline. The results prove the hypothesis
that firewall rules can be automatically generated based on router data, and
that an automated method can be effective in blocking a high percentage of
malicious traffic.
|
Accurate Bayesian Data Classification without Hyperparameter
Cross-validation | We extend the standard Bayesian multivariate Gaussian generative data
classifier by considering a generalization of the conjugate, normal-Wishart
prior distribution and by deriving the hyperparameters analytically via
evidence maximization. The behaviour of the optimal hyperparameters is explored
in the high-dimensional data regime. The classification accuracy of the
resulting generalized model is competitive with state-of-the art Bayesian
discriminant analysis methods, but without the usual computational burden of
cross-validation.
|
Connecting Software Metrics across Versions to Predict Defects | Accurate software defect prediction could help software practitioners
allocate test resources to defect-prone modules effectively and efficiently. In
the last decades, much effort has been devoted to build accurate defect
prediction models, including developing quality defect predictors and modeling
techniques. However, current widely used defect predictors such as code metrics
and process metrics could not well describe how software modules change over
the project evolution, which we believe is important for defect prediction. In
order to deal with this problem, in this paper, we propose to use the
Historical Version Sequence of Metrics (HVSM) in continuous software versions
as defect predictors. Furthermore, we leverage Recurrent Neural Network (RNN),
a popular modeling technique, to take HVSM as the input to build software
prediction models. The experimental results show that, in most cases, the
proposed HVSM-based RNN model has a significantly better effort-aware ranking
effectiveness than the commonly used baseline models.
|
Visualizing the Loss Landscape of Neural Nets | Neural network training relies on our ability to find "good" minimizers of
highly non-convex loss functions. It is well-known that certain network
architecture designs (e.g., skip connections) produce loss functions that train
easier, and well-chosen training parameters (batch size, learning rate,
optimizer) produce minimizers that generalize better. However, the reasons for
these differences, and their effects on the underlying loss landscape, are not
well understood. In this paper, we explore the structure of neural loss
functions, and the effect of loss landscapes on generalization, using a range
of visualization methods. First, we introduce a simple "filter normalization"
method that helps us visualize loss function curvature and make meaningful
side-by-side comparisons between loss functions. Then, using a variety of
visualizations, we explore how network architecture affects the loss landscape,
and how training parameters affect the shape of minimizers.
|
Rapid Adaptation with Conditionally Shifted Neurons | We describe a mechanism by which artificial neural networks can learn rapid
adaptation - the ability to adapt on the fly, with little data, to new tasks -
that we call conditionally shifted neurons. We apply this mechanism in the
framework of metalearning, where the aim is to replicate some of the
flexibility of human learning in machines. Conditionally shifted neurons modify
their activation values with task-specific shifts retrieved from a memory
module, which is populated rapidly based on limited task experience. On
metalearning benchmarks from the vision and language domains, models augmented
with conditionally shifted neurons achieve state-of-the-art results.
|
Gradient Regularization Improves Accuracy of Discriminative Models | Regularizing the gradient norm of the output of a neural network with respect
to its inputs is a powerful technique, rediscovered several times. This paper
presents evidence that gradient regularization can consistently improve
classification accuracy on vision tasks, using modern deep neural networks,
especially when the amount of training data is small. We introduce our
regularizers as members of a broader class of Jacobian-based regularizers. We
demonstrate empirically on real and synthetic data that the learning process
leads to gradients controlled beyond the training points, and results in
solutions that generalize well.
|
Minimizing Polarization and Disagreement in Social Networks | The rise of social media and online social networks has been a disruptive
force in society. Opinions are increasingly shaped by interactions on online
social media, and social phenomena including disagreement and polarization are
now tightly woven into everyday life. In this work we initiate the study of the
following question: given $n$ agents, each with its own initial opinion that
reflects its core value on a topic, and an opinion dynamics model, what is the
structure of a social network that minimizes {\em polarization} and {\em
disagreement} simultaneously?
This question is central to recommender systems: should a recommender system
prefer a link suggestion between two online users with similar mindsets in
order to keep disagreement low, or between two users with different opinions in
order to expose each to the other's viewpoint of the world, and decrease
overall levels of polarization? Our contributions include a mathematical
formalization of this question as an optimization problem and an exact,
time-efficient algorithm. We also prove that there always exists a network with
$O(n/\epsilon^2)$ edges that is a $(1+\epsilon)$ approximation to the optimum.
For a fixed graph, we additionally show how to optimize our objective function
over the agents' innate opinions in polynomial time.
We perform an empirical study of our proposed methods on synthetic and
real-world data that verify their value as mining tools to better understand
the trade-off between of disagreement and polarization. We find that there is a
lot of space to reduce both polarization and disagreement in real-world
networks; for instance, on a Reddit network where users exchange comments on
politics, our methods achieve a $\sim 60\,000$-fold reduction in polarization
and disagreement.
|
Random Feature-based Online Multi-kernel Learning in Environments with
Unknown Dynamics | Kernel-based methods exhibit well-documented performance in various nonlinear
learning tasks. Most of them rely on a preselected kernel, whose prudent choice
presumes task-specific prior information. Especially when the latter is not
available, multi-kernel learning has gained popularity thanks to its
flexibility in choosing kernels from a prescribed kernel dictionary. Leveraging
the random feature approximation and its recent orthogonality-promoting
variant, the present contribution develops a scalable multi-kernel learning
scheme (termed Raker) to obtain the sought nonlinear learning function `on the
fly,' first for static environments. To further boost performance in dynamic
environments, an adaptive multi-kernel learning scheme (termed AdaRaker) is
developed. AdaRaker accounts not only for data-driven learning of kernel
combination, but also for the unknown dynamics. Performance is analyzed in
terms of both static and dynamic regrets. AdaRaker is uniquely capable of
tracking nonlinear learning functions in environments with unknown dynamics,
and with with analytic performance guarantees. Tests with synthetic and real
datasets are carried out to showcase the effectiveness of the novel algorithms.
|
Parallel Active Subspace Decomposition for Scalable and Efficient Tensor
Robust Principal Component Analysis | Tensor robust principal component analysis (TRPCA) has received a substantial
amount of attention in various fields. Most existing methods, normally relying
on tensor nuclear norm minimization, need to pay an expensive computational
cost due to multiple singular value decompositions (SVDs) at each iteration. To
overcome the drawback, we propose a scalable and efficient method, named
Parallel Active Subspace Decomposition (PASD), which divides the unfolding
along each mode of the tensor into a columnwise orthonormal matrix (active
subspace) and another small-size matrix in parallel. Such a transformation
leads to a nonconvex optimization problem in which the scale of nulcear norm
minimization is generally much smaller than that in the original problem.
Furthermore, we introduce an alternating direction method of multipliers (ADMM)
method to solve the reformulated problem and provide rigorous analyses for its
convergence and suboptimality. Experimental results on synthetic and real-world
data show that our algorithm is more accurate than the state-of-the-art
approaches, and is orders of magnitude faster.
|
Debiased Machine Learning of Set-Identified Linear Models | This paper provides estimation and inference methods for an identified set's
boundary (i.e., support function) where the selection among a very large number
of covariates is based on modern regularized tools. I characterize the boundary
using a semiparametric moment equation. Combining Neyman-orthogonality and
sample splitting ideas, I construct a root-N consistent, uniformly
asymptotically Gaussian estimator of the boundary and propose a multiplier
bootstrap procedure to conduct inference. I apply this result to the partially
linear model and the partially linear IV model with an interval-valued outcome.
|
Robust Covariate Shift Prediction with General Losses and Feature Views | Covariate shift relaxes the widely-employed independent and identically
distributed (IID) assumption by allowing different training and testing input
distributions. Unfortunately, common methods for addressing covariate shift by
trying to remove the bias between training and testing distributions using
importance weighting often provide poor performance guarantees in theory and
unreliable predictions with high variance in practice. Recently developed
methods that construct a predictor that is inherently robust to the
difficulties of learning under covariate shift are restricted to minimizing
logloss and can be too conservative when faced with high-dimensional learning
tasks. We address these limitations in two ways: by robustly minimizing various
loss functions, including non-convex ones, under the testing distribution; and
by separately shaping the influence of covariate shift according to different
feature-based views of the relationship between input variables and example
labels. These generalizations make robust covariate shift prediction applicable
to more task scenarios. We demonstrate the benefits on classification under
covariate shift tasks.
|
Kernel Robust Bias-Aware Prediction under Covariate Shift | Under covariate shift, training (source) data and testing (target) data
differ in input space distribution, but share the same conditional label
distribution. This poses a challenging machine learning task. Robust Bias-Aware
(RBA) prediction provides the conditional label distribution that is robust to
the worstcase logarithmic loss for the target distribution while matching
feature expectation constraints from the source distribution. However,
employing RBA with insufficient feature constraints may result in high
certainty predictions for much of the source data, while leaving too much
uncertainty for target data predictions. To overcome this issue, we extend the
representer theorem to the RBA setting, enabling minimization of regularized
expected target risk by a reweighted kernel expectation under the source
distribution. By applying kernel methods, we establish consistency guarantees
and demonstrate better performance of the RBA classifier than competing methods
on synthetically biased UCI datasets as well as datasets that have natural
covariate shift.
|
Multi-timescale memory dynamics in a reinforcement learning network with
attention-gated memory | Learning and memory are intertwined in our brain and their relationship is at
the core of several recent neural network models. In particular, the
Attention-Gated MEmory Tagging model (AuGMEnT) is a reinforcement learning
network with an emphasis on biological plausibility of memory dynamics and
learning. We find that the AuGMEnT network does not solve some hierarchical
tasks, where higher-level stimuli have to be maintained over a long time, while
lower-level stimuli need to be remembered and forgotten over a shorter
timescale. To overcome this limitation, we introduce hybrid AuGMEnT, with leaky
or short-timescale and non-leaky or long-timescale units in memory, that allow
to exchange lower-level information while maintaining higher-level one, thus
solving both hierarchical and distractor tasks.
|
Application of Convolutional Neural Network to Predict Airfoil Lift
Coefficient | The adaptability of the convolutional neural network (CNN) technique for
aerodynamic meta-modeling tasks is probed in this work. The primary objective
is to develop suitable CNN architecture for variable flow conditions and object
geometry, in addition to identifying a sufficient data preparation process.
Multiple CNN structures were trained to learn the lift coefficients of the
airfoils with a variety of shapes in multiple flow Mach numbers, Reynolds
numbers, and diverse angles of attack. This is conducted to illustrate the
concept of the technique. A multi-layered perceptron (MLP) is also used for the
training sets. The MLP results are compared with that of the CNN results. The
newly proposed meta-modeling concept has been found to be comparable with the
MLP in learning capability; and more importantly, our CNN model exhibits a
competitive prediction accuracy with minimal constraints in a geometric
representation.
|
Objective evaluation metrics for automatic classification of EEG events | The evaluation of machine learning algorithms in biomedical fields for
applications involving sequential data lacks standardization. Common
quantitative scalar evaluation metrics such as sensitivity and specificity can
often be misleading depending on the requirements of the application.
Evaluation metrics must ultimately reflect the needs of users yet be
sufficiently sensitive to guide algorithm development. Feedback from critical
care clinicians who use automated event detection software in clinical
applications has been overwhelmingly emphatic that a low false alarm rate,
typically measured in units of the number of errors per 24 hours, is the single
most important criterion for user acceptance. Though using a single metric is
not often as insightful as examining performance over a range of operating
conditions, there is a need for a single scalar figure of merit. In this paper,
we discuss the deficiencies of existing metrics for a seizure detection task
and propose several new metrics that offer a more balanced view of performance.
We demonstrate these metrics on a seizure detection task based on the TUH EEG
Corpus. We show that two promising metrics are a measure based on a concept
borrowed from the spoken term detection literature, Actual Term-Weighted Value
(ATWV), and a new metric, Time-Aligned Event Scoring (TAES), that accounts for
the temporal alignment of the hypothesis to the reference annotation. We also
demonstrate that state of the art technology based on deep learning, though
impressive in its performance, still needs significant improvement before it
meets very strict user acceptance criteria.
|
The Multilinear Structure of ReLU Networks | We study the loss surface of neural networks equipped with a hinge loss
criterion and ReLU or leaky ReLU nonlinearities. Any such network defines a
piecewise multilinear form in parameter space. By appealing to harmonic
analysis we show that all local minima of such network are non-differentiable,
except for those minima that occur in a region of parameter space where the
loss surface is perfectly flat. Non-differentiable minima are therefore not
technicalities or pathologies; they are heart of the problem when investigating
the loss of ReLU networks. As a consequence, we must employ techniques from
nonsmooth analysis to study these loss surfaces. We show how to apply these
techniques in some illustrative cases.
|
Non-linear motor control by local learning in spiking neural networks | Learning weights in a spiking neural network with hidden neurons, using
local, stable and online rules, to control non-linear body dynamics is an open
problem. Here, we employ a supervised scheme, Feedback-based Online Local
Learning Of Weights (FOLLOW), to train a network of heterogeneous spiking
neurons with hidden layers, to control a two-link arm so as to reproduce a
desired state trajectory. The network first learns an inverse model of the
non-linear dynamics, i.e. from state trajectory as input to the network, it
learns to infer the continuous-time command that produced the trajectory.
Connection weights are adjusted via a local plasticity rule that involves
pre-synaptic firing and post-synaptic feedback of the error in the inferred
command. We choose a network architecture, termed differential feedforward,
that gives the lowest test error from different feedforward and recurrent
architectures. The learned inverse model is then used to generate a
continuous-time motor command to control the arm, given a desired trajectory.
|
Deep Learning Interior Tomography for Region-of-Interest Reconstruction | Interior tomography for the region-of-interest (ROI) imaging has advantages
of using a small detector and reducing X-ray radiation dose. However, standard
analytic reconstruction suffers from severe cupping artifacts due to existence
of null space in the truncated Radon transform. Existing penalized
reconstruction methods may address this problem but they require extensive
computations due to the iterative reconstruction. Inspired by the recent deep
learning approaches to low-dose and sparse view CT, here we propose a deep
learning architecture that removes null space signals from the FBP
reconstruction. Experimental results have shown that the proposed method
provides near-perfect reconstruction with about 7-10 dB improvement in PSNR
over existing methods in spite of significantly reduced run-time complexity.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.