title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Fast Counting in Machine Learning Applications | We propose scalable methods to execute counting queries in machine learning
applications. To achieve memory and computational efficiency, we abstract
counting queries and their context such that the counts can be aggregated as a
stream. We demonstrate performance and scalability of the resulting approach on
random queries, and through extensive experimentation using Bayesian networks
learning and association rule mining. Our methods significantly outperform
commonly used ADtrees and hash tables, and are practical alternatives for
processing large-scale data.
|
Machine Learning Enabled Computational Screening of Inorganic Solid
Electrolytes for Dendrite Suppression with Li Metal Anode | Next generation batteries based on lithium (Li) metal anodes have been
plagued by the dendritic electrodeposition of Li metal on the anode during
cycling, resulting in short circuit and capacity loss. Suppression of dendritic
growth through the use of solid electrolytes has emerged as one of the most
promising strategies for enabling the use of Li metal anodes. We perform a
computational screening of over 12,000 inorganic solids based on their ability
to suppress dendrite initiation in contact with Li metal anode. Properties for
mechanically isotropic and anisotropic interfaces that can be used in stability
criteria for determining the propensity of dendrite initiation are usually
obtained from computationally expensive first-principles methods. In order to
obtain a large dataset for screening, we use machine learning models to predict
the mechanical properties of several new solid electrolytes. We train a
convolutional neural network on the shear and bulk moduli purely on structural
features of the material. We use AdaBoost, Lasso and Bayesian ridge regression
to train the elastic constants, where the choice of the model depended on the
size of the training data and the noise that it can handle. Our models give us
direct interpretability by revealing the dominant structural features affecting
the elastic constants. The stiffness is found to increase with a decrease in
volume per atom, increase in minimum anion-anion separation, and increase in
sublattice (all but Li) packing fraction. Cross-validation/test performance
suggests our models generalize well. We predict over 20 mechanically
anisotropic interfaces between Li metal and 6 solid electrolytes which can be
used to suppress dendrite growth. Our screened candidates are generally soft
and highly anisotropic, and present opportunities for simultaneously obtaining
dendrite suppression and high ionic conductivity in solid electrolytes.
|
3D G-CNNs for Pulmonary Nodule Detection | Convolutional Neural Networks (CNNs) require a large amount of annotated data
to learn from, which is often difficult to obtain in the medical domain. In
this paper we show that the sample complexity of CNNs can be significantly
improved by using 3D roto-translation group convolutions (G-Convs) instead of
the more conventional translational convolutions. These 3D G-CNNs were applied
to the problem of false positive reduction for pulmonary nodule detection, and
proved to be substantially more effective in terms of performance, sensitivity
to malignant nodules, and speed of convergence compared to a strong and
comparable baseline architecture with regular convolutions, data augmentation
and a similar number of parameters. For every dataset size tested, the G-CNN
achieved a FROC score close to the CNN trained on ten times more data.
|
Asynch-SGBDT: Asynchronous Parallel Stochastic Gradient Boosting
Decision Tree based on Parameters Server | In AI research and industry, machine learning is the most widely used tool.
One of the most important machine learning algorithms is Gradient Boosting
Decision Tree, i.e. GBDT whose training process needs considerable
computational resources and time. To shorten GBDT training time, many works
tried to apply GBDT on Parameter Server. However, those GBDT algorithms are
synchronous parallel algorithms which fail to make full use of Parameter
Server. In this paper, we examine the possibility of using asynchronous
parallel methods to train GBDT model and name this algorithm as asynch-SGBDT
(asynchronous parallel stochastic gradient boosting decision tree). Our
theoretical and experimental results indicate that the scalability of
asynch-SGBDT is influenced by the sample diversity of datasets, sampling rate,
step length and the setting of GBDT tree. Experimental results also show
asynch-SGBDT training process reaches a linear speedup in asynchronous parallel
manner when datasets and GBDT trees meet high scalability requirements.
|
Efficient Model Identification for Tensegrity Locomotion | This paper aims to identify in a practical manner unknown physical
parameters, such as mechanical models of actuated robot links, which are
critical in dynamical robotic tasks. Key features include the use of an
off-the-shelf physics engine and the Bayesian optimization framework. The task
being considered is locomotion with a high-dimensional, compliant Tensegrity
robot. A key insight, in this case, is the need to project the model
identification challenge into an appropriate lower dimensional space for
efficiency. Comparisons with alternatives indicate that the proposed method can
identify the parameters more accurately within the given time budget, which
also results in more precise locomotion control.
|
Network-based protein structural classification | Experimental determination of protein function is resource-consuming. As an
alternative, computational prediction of protein function has received
attention. In this context, protein structural classification (PSC) can help,
by allowing for determining structural classes of currently unclassified
proteins based on their features, and then relying on the fact that proteins
with similar structures have similar functions. Existing PSC approaches rely on
sequence-based or direct 3-dimensional (3D) structure-based protein features.
In contrast, we first model 3D structures of proteins as protein structure
networks (PSNs). Then, we use network-based features for PSC. We propose the
use of graphlets, state-of-the-art features in many research areas of network
science, in the task of PSC. Moreover, because graphlets can deal only with
unweighted PSNs, and because accounting for edge weights when constructing PSNs
could improve PSC accuracy, we also propose a deep learning framework that
automatically learns network features from weighted PSNs. When evaluated on a
large set of ~9,400 CATH and ~12,800 SCOP protein domains (spanning 36 PSN
sets), our proposed approaches are superior to existing PSC approaches in terms
of accuracy, with comparable running time.
|
Multimodal Unsupervised Image-to-Image Translation | Unsupervised image-to-image translation is an important and challenging
problem in computer vision. Given an image in the source domain, the goal is to
learn the conditional distribution of corresponding images in the target
domain, without seeing any pairs of corresponding images. While this
conditional distribution is inherently multimodal, existing approaches make an
overly simplified assumption, modeling it as a deterministic one-to-one
mapping. As a result, they fail to generate diverse outputs from a given source
domain image. To address this limitation, we propose a Multimodal Unsupervised
Image-to-image Translation (MUNIT) framework. We assume that the image
representation can be decomposed into a content code that is domain-invariant,
and a style code that captures domain-specific properties. To translate an
image to another domain, we recombine its content code with a random style code
sampled from the style space of the target domain. We analyze the proposed
framework and establish several theoretical results. Extensive experiments with
comparisons to the state-of-the-art approaches further demonstrates the
advantage of the proposed framework. Moreover, our framework allows users to
control the style of translation outputs by providing an example style image.
Code and pretrained models are available at https://github.com/nvlabs/MUNIT
|
MOVI: A Model-Free Approach to Dynamic Fleet Management | Modern vehicle fleets, e.g., for ridesharing platforms and taxi companies,
can reduce passengers' waiting times by proactively dispatching vehicles to
locations where pickup requests are anticipated in the future. Yet it is
unclear how to best do this: optimal dispatching requires optimizing over
several sources of uncertainty, including vehicles' travel times to their
dispatched locations, as well as coordinating between vehicles so that they do
not attempt to pick up the same passenger. While prior works have developed
models for this uncertainty and used them to optimize dispatch policies, in
this work we introduce a model-free approach. Specifically, we propose MOVI, a
Deep Q-network (DQN)-based framework that directly learns the optimal vehicle
dispatch policy. Since DQNs scale poorly with a large number of possible
dispatches, we streamline our DQN training and suppose that each individual
vehicle independently learns its own optimal policy, ensuring scalability at
the cost of less coordination between vehicles. We then formulate a centralized
receding-horizon control (RHC) policy to compare with our DQN policies. To
compare these policies, we design and build MOVI as a large-scale realistic
simulator based on 15 million taxi trip records that simulates policy-agnostic
responses to dispatch decisions. We show that the DQN dispatch policy reduces
the number of unserviced requests by 76% compared to without dispatch and 20%
compared to the RHC approach, emphasizing the benefits of a model-free approach
and suggesting that there is limited value to coordinating vehicle actions.
This finding may help to explain the success of ridesharing platforms, for
which drivers make individual decisions.
|
RIPEx: Extracting malicious IP addresses from security forums using
cross-forum learning | Is it possible to extract malicious IP addresses reported in security forums
in an automatic way? This is the question at the heart of our work. We focus on
security forums, where security professionals and hackers share knowledge and
information, and often report misbehaving IP addresses. So far, there have only
been a few efforts to extract information from such security forums. We propose
RIPEx, a systematic approach to identify and label IP addresses in security
forums by utilizing a cross-forum learning method. In more detail, the
challenge is twofold: (a) identifying IP addresses from other numerical
entities, such as software version numbers, and (b) classifying the IP address
as benign or malicious. We propose an integrated solution that tackles both
these problems. A novelty of our approach is that it does not require training
data for each new forum. Our approach does knowledge transfer across forums: we
use a classifier from our source forums to identify seed information for
training a classifier on the target forum. We evaluate our method using data
collected from five security forums with a total of 31K users and 542K posts.
First, RIPEx can distinguish IP address from other numeric expressions with 95%
precision and above 93% recall on average. Second, RIPEx identifies malicious
IP addresses with an average precision of 88% and over 78% recall, using our
cross-forum learning. Our work is a first step towards harnessing the wealth of
useful information that can be found in security forums.
|
A Compact Network Learning Model for Distribution Regression | Despite the superior performance of deep learning in many applications,
challenges remain in the area of regression on function spaces. In particular,
neural networks are unable to encode function inputs compactly as each node
encodes just a real value. We propose a novel idea to address this shortcoming:
to encode an entire function in a single network node. To that end, we design a
compact network representation that encodes and propagates functions in single
nodes for the distribution regression task. Our proposed Distribution
Regression Network (DRN) achieves higher prediction accuracies while being much
more compact and uses fewer parameters than traditional neural networks.
|
Understanding Community Structure in Layered Neural Networks | A layered neural network is now one of the most common choices for the
prediction of high-dimensional practical data sets, where the relationship
between input and output data is complex and cannot be represented well by
simple conventional models. Its effectiveness is shown in various tasks,
however, the lack of interpretability of the trained result by a layered neural
network has limited its application area.
In our previous studies, we proposed methods for extracting a simplified
global structure of a trained layered neural network by classifying the units
into communities according to their connection patterns with adjacent layers.
These methods provided us with knowledge about the strength of the relationship
between communities from the existence of bundled connections, which are
determined by threshold processing of the connection ratio between pairs of
communities.
However, it has been difficult to understand the role of each community
quantitatively by observing the modular structure. We could only know to which
sets of the input and output dimensions each community was mainly connected, by
tracing the bundled connections from the community to the input and output
layers. Another problem is that the finally obtained modular structure is
changed greatly depending on the setting of the threshold hyperparameter used
for determining bundled connections.
In this paper, we propose a new method for interpreting quantitatively the
role of each community in inference, by defining the effect of each input
dimension on a community, and the effect of a community on each output
dimension. We show experimentally that our proposed method can reveal the role
of each part of a layered neural network by applying the neural networks to
three types of data sets, extracting communities from the trained network, and
applying the proposed method to the community structure.
|
Adversarial Clustering: A Grid Based Clustering Algorithm Against Active
Adversaries | Nowadays more and more data are gathered for detecting and preventing cyber
attacks. In cyber security applications, data analytics techniques have to deal
with active adversaries that try to deceive the data analytics models and avoid
being detected. The existence of such adversarial behavior motivates the
development of robust and resilient adversarial learning techniques for various
tasks. Most of the previous work focused on adversarial classification
techniques, which assumed the existence of a reasonably large amount of
carefully labeled data instances. However, in practice, labeling the data
instances often requires costly and time-consuming human expertise and becomes
a significant bottleneck. Meanwhile, a large number of unlabeled instances can
also be used to understand the adversaries' behavior. To address the above
mentioned challenges, in this paper, we develop a novel grid based adversarial
clustering algorithm. Our adversarial clustering algorithm is able to identify
the core normal regions, and to draw defensive walls around the centers of the
normal objects utilizing game theoretic ideas. Our algorithm also identifies
sub-clusters of attack objects, the overlapping areas within clusters, and
outliers which may be potential anomalies.
|
Fast, Parameter free Outlier Identification for Robust PCA | Robust PCA, the problem of PCA in the presence of outliers has been
extensively investigated in the last few years. Here we focus on Robust PCA in
the column sparse outlier model. The existing methods for column sparse outlier
model assumes either the knowledge of the dimension of the lower dimensional
subspace or the fraction of outliers in the system. However in many
applications knowledge of these parameters is not available. Motivated by this
we propose a parameter free outlier identification method for robust PCA which
a) does not require the knowledge of outlier fraction, b) does not require the
knowledge of the dimension of the underlying subspace, c) is computationally
simple and fast. Further, analytical guarantees are derived for outlier
identification and the performance of the algorithm is compared with the
existing state of the art methods.
|
{\mu}-cuDNN: Accelerating Deep Learning Frameworks with Micro-Batching | NVIDIA cuDNN is a low-level library that provides GPU kernels frequently used
in deep learning. Specifically, cuDNN implements several equivalent convolution
algorithms, whose performance and memory footprint may vary considerably,
depending on the layer dimensions. When an algorithm is automatically selected
by cuDNN, the decision is performed on a per-layer basis, and thus it often
resorts to slower algorithms that fit the workspace size constraints. We
present {\mu}-cuDNN, a transparent wrapper library for cuDNN, which divides
layers' mini-batch computation into several micro-batches. Based on Dynamic
Programming and Integer Linear Programming, {\mu}-cuDNN enables faster
algorithms by decreasing the workspace requirements. At the same time,
{\mu}-cuDNN keeps the computational semantics unchanged, so that it decouples
statistical efficiency from the hardware efficiency safely. We demonstrate the
effectiveness of {\mu}-cuDNN over two frameworks, Caffe and TensorFlow,
achieving speedups of 1.63x for AlexNet and 1.21x for ResNet-18 on P100-SXM2
GPU. These results indicate that using micro-batches can seamlessly increase
the performance of deep learning, while maintaining the same memory footprint.
|
The unreasonable effectiveness of the forget gate | Given the success of the gated recurrent unit, a natural question is whether
all the gates of the long short-term memory (LSTM) network are necessary.
Previous research has shown that the forget gate is one of the most important
gates in the LSTM. Here we show that a forget-gate-only version of the LSTM
with chrono-initialized biases, not only provides computational savings but
outperforms the standard LSTM on multiple benchmark datasets and competes with
some of the best contemporary models. Our proposed network, the JANET, achieves
accuracies of 99% and 92.5% on the MNIST and pMNIST datasets, outperforming the
standard LSTM which yields accuracies of 98.5% and 91%.
|
Learning Contracting Vector Fields For Stable Imitation Learning | We propose a new non-parametric framework for learning incrementally stable
dynamical systems x' = f(x) from a set of sampled trajectories. We construct a
rich family of smooth vector fields induced by certain classes of matrix-valued
kernels, whose equilibria are placed exactly at a desired set of locations and
whose local contraction and curvature properties at various points can be
explicitly controlled using convex optimization. With curl-free kernels, our
framework may also be viewed as a mechanism to learn potential fields and
gradient flows. We develop large-scale techniques using randomized kernel
approximations in this context. We demonstrate our approach, called contracting
vector fields (CVF), on imitation learning tasks involving complex
point-to-point human handwriting motions.
|
Scalable and Interpretable One-class SVMs with Deep Learning and Random
Fourier features | One-class support vector machine (OC-SVM) for a long time has been one of the
most effective anomaly detection methods and extensively adopted in both
research as well as industrial applications. The biggest issue for OC-SVM is
yet the capability to operate with large and high-dimensional datasets due to
optimization complexity. Those problems might be mitigated via dimensionality
reduction techniques such as manifold learning or autoencoder. However,
previous work often treats representation learning and anomaly prediction
separately. In this paper, we propose autoencoder based one-class support
vector machine (AE-1SVM) that brings OC-SVM, with the aid of random Fourier
features to approximate the radial basis kernel, into deep learning context by
combining it with a representation learning architecture and jointly exploit
stochastic gradient descent to obtain end-to-end training. Interestingly, this
also opens up the possible use of gradient-based attribution methods to explain
the decision making for anomaly detection, which has ever been challenging as a
result of the implicit mappings between the input space and the kernel space.
To the best of our knowledge, this is the first work to study the
interpretability of deep learning in anomaly detection. We evaluate our method
on a wide range of unsupervised anomaly detection tasks in which our end-to-end
training architecture achieves a performance significantly better than the
previous work using separate training.
|
Distributed Collaborative Hashing and Its Applications in Ant Financial | Collaborative filtering, especially latent factor model, has been popularly
used in personalized recommendation. Latent factor model aims to learn user and
item latent factors from user-item historic behaviors. To apply it into real
big data scenarios, efficiency becomes the first concern, including offline
model training efficiency and online recommendation efficiency. In this paper,
we propose a Distributed Collaborative Hashing (DCH) model which can
significantly improve both efficiencies. Specifically, we first propose a
distributed learning framework, following the state-of-the-art parameter server
paradigm, to learn the offline collaborative model. Our model can be learnt
efficiently by distributedly computing subgradients in minibatches on workers
and updating model parameters on servers asynchronously. We then adopt hashing
technique to speedup the online recommendation procedure. Recommendation can be
quickly made through exploiting lookup hash tables. We conduct thorough
experiments on two real large-scale datasets. The experimental results
demonstrate that, comparing with the classic and state-of-the-art (distributed)
latent factor models, DCH has comparable performance in terms of recommendation
accuracy but has both fast convergence speed in offline model training
procedure and realtime efficiency in online recommendation procedure.
Furthermore, the encouraging performance of DCH is also shown for several
real-world applications in Ant Financial.
|
DeepFM: An End-to-End Wide & Deep Learning Framework for CTR Prediction | Learning sophisticated feature interactions behind user behaviors is critical
in maximizing CTR for recommender systems. Despite great progress, existing
methods have a strong bias towards low- or high-order interactions, or rely on
expertise feature engineering. In this paper, we show that it is possible to
derive an end-to-end learning model that emphasizes both low- and high-order
feature interactions. The proposed framework, DeepFM, combines the power of
factorization machines for recommendation and deep learning for feature
learning in a new neural network architecture. Compared to the latest Wide &
Deep model from Google, DeepFM has a shared raw feature input to both its
"wide" and "deep" components, with no need of feature engineering besides raw
features. DeepFM, as a general learning framework, can incorporate various
network architectures in its deep component. In this paper, we study two
instances of DeepFM where its "deep" component is DNN and PNN respectively, for
which we denote as DeepFM-D and DeepFM-P. Comprehensive experiments are
conducted to demonstrate the effectiveness of DeepFM-D and DeepFM-P over the
existing models for CTR prediction, on both benchmark data and commercial data.
We conduct online A/B test in Huawei App Market, which reveals that DeepFM-D
leads to more than 10% improvement of click-through rate in the production
environment, compared to a well-engineered LR model. We also covered related
practice in deploying our framework in Huawei App Market.
|
Online Fall Detection using Recurrent Neural Networks | Unintentional falls can cause severe injuries and even death, especially if
no immediate assistance is given. The aim of Fall Detection Systems (FDSs) is
to detect an occurring fall. This information can be used to trigger the
necessary assistance in case of injury. This can be done by using either
ambient-based sensors, e.g. cameras, or wearable devices. The aim of this work
is to study the technical aspects of FDSs based on wearable devices and
artificial intelligence techniques, in particular Deep Learning (DL), to
implement an effective algorithm for on-line fall detection. The proposed
classifier is based on a Recurrent Neural Network (RNN) model with underlying
Long Short-Term Memory (LSTM) blocks. The method is tested on the publicly
available SisFall dataset, with extended annotation, and compared with the
results obtained by the SisFall authors.
|
Roster Evaluation Based on Classifiers for the Nurse Rostering Problem | The personnel scheduling problem is a well-known NP-hard combinatorial
problem. Due to the complexity of this problem and the size of the real-world
instances, it is not possible to use exact methods, and thus heuristics,
meta-heuristics, or hyper-heuristics must be employed. The majority of
heuristic approaches are based on iterative search, where the quality of
intermediate solutions must be calculated. Unfortunately, this is
computationally highly expensive because these problems have many constraints
and some are very complex. In this study, we propose a machine learning
technique as a tool to accelerate the evaluation phase in heuristic approaches.
The solution is based on a simple classifier, which is able to determine
whether the changed solution (more precisely, the changed part of the solution)
is better than the original or not. This decision is made much faster than a
standard cost-oriented evaluation process. However, the classification process
cannot guarantee 100% correctness. Therefore, our approach, which is
illustrated using a tabu search algorithm in this study, includes a filtering
mechanism, where the classifier rejects the majority of the potentially bad
solutions and the remaining solutions are then evaluated in a standard manner.
We also show how the boosting algorithms can improve the quality of the final
solution compared with a simple classifier. We verified our proposed approach
and premises, based on standard and real-world benchmark instances, to
demonstrate the significant speedup obtained with comparable solution quality.
|
Representing smooth functions as compositions of near-identity functions
with implications for deep network optimization | We show that any smooth bi-Lipschitz $h$ can be represented exactly as a
composition $h_m \circ ... \circ h_1$ of functions $h_1,...,h_m$ that are close
to the identity in the sense that each $\left(h_i-\mathrm{Id}\right)$ is
Lipschitz, and the Lipschitz constant decreases inversely with the number $m$
of functions composed. This implies that $h$ can be represented to any accuracy
by a deep residual network whose nonlinear layers compute functions with a
small Lipschitz constant. Next, we consider nonlinear regression with a
composition of near-identity nonlinear maps. We show that, regarding Fr\'echet
derivatives with respect to the $h_1,...,h_m$, any critical point of a
quadratic criterion in this near-identity region must be a global minimizer. In
contrast, if we consider derivatives with respect to parameters of a fixed-size
residual network with sigmoid activation functions, we show that there are
near-identity critical points that are suboptimal, even in the realizable case.
Informally, this means that functional gradient methods for residual networks
cannot get stuck at suboptimal critical points corresponding to near-identity
layers, whereas parametric gradient methods for sigmoidal residual networks
suffer from suboptimal critical points in the near-identity region.
|
Connectivity in Random Annulus Graphs and the Geometric Block Model | We provide new connectivity results for {\em vertex-random graphs} or {\em
random annulus graphs} which are significant generalizations of random
geometric graphs. Random geometric graphs (RGG) are one of the most basic
models of random graphs for spatial networks proposed by Gilbert in 1961,
shortly after the introduction of the Erd\H{o}s-R\'{en}yi random graphs. They
resemble social networks in many ways (e.g. by spontaneously creating cluster
of nodes with high modularity). The connectivity properties of RGG have been
studied since its introduction, and analyzing them has been significantly
harder than their Erd\H{o}s-R\'{en}yi counterparts due to correlated edge
formation.
Our next contribution is in using the connectivity of random annulus graphs
to provide necessary and sufficient conditions for efficient recovery of
communities for {\em the geometric block model} (GBM). The GBM is a
probabilistic model for community detection defined over an RGG in a similar
spirit as the popular {\em stochastic block model}, which is defined over an
Erd\H{o}s-R\'{en}yi random graph. The geometric block model inherits the
transitivity properties of RGGs and thus models communities better than a
stochastic block model. However, analyzing them requires fresh perspectives as
all prior tools fail due to correlation in edge formation. We provide a simple
and efficient algorithm that can recover communities in GBM exactly with high
probability in the regime of connectivity.
|
Comparatives, Quantifiers, Proportions: A Multi-Task Model for the
Learning of Quantities from Vision | The present work investigates whether different quantification mechanisms
(set comparison, vague quantification, and proportional estimation) can be
jointly learned from visual scenes by a multi-task computational model. The
motivation is that, in humans, these processes underlie the same cognitive,
non-symbolic ability, which allows an automatic estimation and comparison of
set magnitudes. We show that when information about lower-complexity tasks is
available, the higher-level proportional task becomes more accurate than when
performed in isolation. Moreover, the multi-task model is able to generalize to
unseen combinations of target/non-target objects. Consistently with behavioral
evidence showing the interference of absolute number in the proportional task,
the multi-task model no longer works when asked to provide the number of target
objects in the scene.
|
A Deep Learning Approach to Fast, Format-Agnostic Detection of Malicious
Web Content | Malicious web content is a serious problem on the Internet today. In this
paper we propose a deep learning approach to detecting malevolent web pages.
While past work on web content detection has relied on syntactic parsing or on
emulation of HTML and Javascript to extract features, our approach operates
directly on a language-agnostic stream of tokens extracted directly from static
HTML files with a simple regular expression. This makes it fast enough to
operate in high-frequency data contexts like firewalls and web proxies, and
allows it to avoid the attack surface exposure of complex parsing and emulation
code. Unlike well-known approaches such as bag-of-words models, which ignore
spatial information, our neural network examines content at hierarchical
spatial scales, allowing our model to capture locality and yielding superior
accuracy compared to bag-of-words baselines. Our proposed architecture achieves
a 97.5% detection rate at a 0.1% false positive rate, and classifies
small-batched web pages at a rate of over 100 per second on commodity hardware.
The speed and accuracy of our approach makes it appropriate for deployment to
endpoints, firewalls, and web proxies.
|
Machine Learning in Astronomy: A Case Study in Quasar-Star
Classification | We present the results of various automated classification methods, based on
machine learning (ML), of objects from data releases 6 and 7 (DR6 and DR7) of
the Sloan Digital Sky Survey (SDSS), primarily distinguishing stars from
quasars. We provide a careful scrutiny of approaches available in the
literature and have highlighted the pitfalls in those approaches based on the
nature of data used for the study. The aim is to investigate the
appropriateness of the application of certain ML methods. The manuscript argues
convincingly in favor of the efficacy of asymmetric AdaBoost to classify
photometric data. The paper presents a critical review of existing study and
puts forward an application of asymmetric AdaBoost, as an offspring of that
exercise.
|
Regularized Singular Value Decomposition and Application to Recommender
System | Singular value decomposition (SVD) is the mathematical basis of principal
component analysis (PCA). Together, SVD and PCA are one of the most widely used
mathematical formalism/decomposition in machine learning, data mining, pattern
recognition, artificial intelligence, computer vision, signal processing, etc.
In recent applications, regularization becomes an increasing trend. In this
paper, we present a regularized SVD (RSVD), present an efficient computational
algorithm, and provide several theoretical analysis. We show that although RSVD
is non-convex, it has a closed-form global optimal solution. Finally, we apply
RSVD to the application of recommender system and experimental result show that
RSVD outperforms SVD significantly.
|
A new robust feature selection method using variance-based sensitivity
analysis | Excluding irrelevant features in a pattern recognition task plays an
important role in maintaining a simpler machine learning model and optimizing
the computational efficiency. Nowadays with the rise of large scale datasets,
feature selection is in great demand as it becomes a central issue when facing
high-dimensional datasets. The present study provides a new measure of saliency
for features by employing a Sensitivity Analysis (SA) technique called the
extended Fourier amplitude sensitivity test, and a well-trained Feedforward
Neural Network (FNN) model, which ultimately leads to the selection of a
promising optimal feature subset. Ideas of the paper are mainly demonstrated
based on adopting FNN model for feature selection in classification problems.
But in the end, a generalization framework is discussed in order to give
insights into the usage in regression problems as well as expressing how other
function approximate models can be deployed. Effectiveness of the proposed
method is verified by result analysis and data visualization for a series of
experiments over several well-known datasets drawn from UCI machine learning
repository.
|
Robust Dual View Deep Agent | Motivated by recent advance of machine learning using Deep Reinforcement
Learning this paper proposes a modified architecture that produces more robust
agents and speeds up the training process. Our architecture is based on
Asynchronous Advantage Actor-Critic (A3C) algorithm where the total input
dimensionality is halved by dividing the input into two independent streams. We
use ViZDoom, 3D world software that is based on the classical first person
shooter video game, Doom, as a test case. The experiments show that in
comparison to single input agents, the proposed architecture succeeds to have
the same playing performance and shows more robust behavior, achieving
significant reduction in the number of training parameters of almost 30%.
|
A comparison of methods for model selection when estimating individual
treatment effects | Practitioners in medicine, business, political science, and other fields are
increasingly aware that decisions should be personalized to each patient,
customer, or voter. A given treatment (e.g. a drug or advertisement) should be
administered only to those who will respond most positively, and certainly not
to those who will be harmed by it. Individual-level treatment effects can be
estimated with tools adapted from machine learning, but different models can
yield contradictory estimates. Unlike risk prediction models, however,
treatment effect models cannot be easily evaluated against each other using a
held-out test set because the true treatment effect itself is never directly
observed. Besides outcome prediction accuracy, several metrics that can
leverage held-out data to evaluate treatment effects models have been proposed,
but they are not widely used. We provide a didactic framework that elucidates
the relationships between the different approaches and compare them all using a
variety of simulations of both randomized and observational data. Our results
show that researchers estimating heterogenous treatment effects need not limit
themselves to a single model-fitting algorithm. Instead of relying on a single
method, multiple models fit by a diverse set of algorithms should be evaluated
against each other using an objective function learned from the validation set.
The model minimizing that objective should be used for estimating the
individual treatment effect for future individuals.
|
Exploring the Encoding Layer and Loss Function in End-to-End Speaker and
Language Recognition System | In this paper, we explore the encoding/pooling layer and loss function in the
end-to-end speaker and language recognition system. First, a unified and
interpretable end-to-end system for both speaker and language recognition is
developed. It accepts variable-length input and produces an utterance level
result. In the end-to-end system, the encoding layer plays a role in
aggregating the variable-length input sequence into an utterance level
representation. Besides the basic temporal average pooling, we introduce a
self-attentive pooling layer and a learnable dictionary encoding layer to get
the utterance level representation. In terms of loss function for open-set
speaker verification, to get more discriminative speaker embedding, center loss
and angular softmax loss is introduced in the end-to-end system. Experimental
results on Voxceleb and NIST LRE 07 datasets show that the performance of
end-to-end learning system could be significantly improved by the proposed
encoding layer and loss function.
|
Model-Free Information Extraction in Enriched Nonlinear Phase-Space | Detecting anomalies and discovering driving signals is an essential component
of scientific research and industrial practice. Often the underlying mechanism
is highly complex, involving hidden evolving nonlinear dynamics and noise
contamination. When representative physical models and large labeled data sets
are unavailable, as is the case with most real-world applications,
model-dependent Bayesian approaches would yield misleading results, and most
supervised learning machines would also fail to reliably resolve the
intricately evolving systems. Here, we propose an unsupervised machine-learning
approach that operates in a well-constructed function space, whereby the
evolving nonlinear dynamics are captured through a linear functional
representation determined by the Koopman operator. This breakthrough leverages
on the time-feature embedding and the ensuing reconstruction of a phase-space
representation of the dynamics, thereby permitting the reliable identification
of critical global signatures from the whole trajectory. This dramatically
improves over commonly used static local features, which are vulnerable to
unknown transitions or noise. Thanks to its data-driven nature, our method
excludes any prior models and training corpus. We benchmark the astonishing
accuracy of our method on three diverse and challenging problems in: biology,
medicine, and engineering. In all cases, it outperforms existing
state-of-the-art methods. As a new unsupervised information processing
paradigm, it is suitable for ubiquitous nonlinear dynamical systems or
end-users with little expertise, which permits an unbiased excavation of
underlying working principles or intrinsic correlations submerged in unlabeled
data flows.
|
Motion-based Object Segmentation based on Dense RGB-D Scene Flow | Given two consecutive RGB-D images, we propose a model that estimates a dense
3D motion field, also known as scene flow. We take advantage of the fact that
in robot manipulation scenarios, scenes often consist of a set of rigidly
moving objects. Our model jointly estimates (i) the segmentation of the scene
into an unknown but finite number of objects, (ii) the motion trajectories of
these objects and (iii) the object scene flow. We employ an hourglass, deep
neural network architecture. In the encoding stage, the RGB and depth images
undergo spatial compression and correlation. In the decoding stage, the model
outputs three images containing a per-pixel estimate of the corresponding
object center as well as object translation and rotation. This forms the basis
for inferring the object segmentation and final object scene flow. To evaluate
our model, we generated a new and challenging, large-scale, synthetic dataset
that is specifically targeted at robotic manipulation: It contains a large
number of scenes with a very diverse set of simultaneously moving 3D objects
and is recorded with a simulated, static RGB-D camera. In quantitative
experiments, we show that we outperform state-of-the-art scene flow and
motion-segmentation methods on this data set. In qualitative experiments, we
show how our learned model transfers to challenging real-world scenes, visually
generating better results than existing methods.
|
Fast Optimal Bandwidth Selection for RBF Kernel using Reproducing Kernel
Hilbert Space Operators for Kernel Based Classifiers | Kernel based methods have shown effective performance in many remote sensing
classification tasks. However their performance significantly depend on its
hyper-parameters. The conventional technique to estimate the parameter comes
with high computational complexity. Thus, the objective of this letter is to
propose an fast and efficient method to select the bandwidth parameter of the
Gaussian kernel in the kernel based classification methods. The proposed method
is developed based on the operators in the reproducing kernel Hilbert space and
it is evaluated on Support vector machines and PerTurbo classification method.
Experiments conducted with hyperspectral datasets show that our proposed method
outperforms the state-of-art method in terms in computational time and
classification performance.
|
An interpretable LSTM neural network for autoregressive exogenous model | In this paper, we propose an interpretable LSTM recurrent neural network,
i.e., multi-variable LSTM for time series with exogenous variables. Currently,
widely used attention mechanism in recurrent neural networks mostly focuses on
the temporal aspect of data and falls short of characterizing variable
importance. To this end, our multi-variable LSTM equipped with tensorized
hidden states is developed to learn variable specific representations, which
give rise to both temporal and variable level attention. Preliminary
experiments demonstrate comparable prediction performance of multi-variable
LSTM w.r.t. encoder-decoder based baselines. More interestingly, variable
importance in real datasets characterized by the variable attention is highly
in line with that determined by statistical Granger causality test, which
exhibits the prospect of multi-variable LSTM as a simple and uniform end-to-end
framework for both forecasting and knowledge discovery.
|
ClassiNet -- Predicting Missing Features for Short-Text Classification | The fundamental problem in short-text classification is \emph{feature
sparseness} -- the lack of feature overlap between a trained model and a test
instance to be classified. We propose \emph{ClassiNet} -- a network of
classifiers trained for predicting missing features in a given instance, to
overcome the feature sparseness problem. Using a set of unlabeled training
instances, we first learn binary classifiers as feature predictors for
predicting whether a particular feature occurs in a given instance. Next, each
feature predictor is represented as a vertex $v_i$ in the ClassiNet where a
one-to-one correspondence exists between feature predictors and vertices. The
weight of the directed edge $e_{ij}$ connecting a vertex $v_i$ to a vertex
$v_j$ represents the conditional probability that given $v_i$ exists in an
instance, $v_j$ also exists in the same instance. We show that ClassiNets
generalize word co-occurrence graphs by considering implicit co-occurrences
between features. We extract numerous features from the trained ClassiNet to
overcome feature sparseness. In particular, for a given instance $\vec{x}$, we
find similar features from ClassiNet that did not appear in $\vec{x}$, and
append those features in the representation of $\vec{x}$. Moreover, we propose
a method based on graph propagation to find features that are indirectly
related to a given short-text. We evaluate ClassiNets on several benchmark
datasets for short-text classification. Our experimental results show that by
using ClassiNet, we can statistically significantly improve the accuracy in
short-text classification tasks, without having to use any external resources
such as thesauri for finding related features.
|
Low-Precision Floating-Point Schemes for Neural Network Training | The use of low-precision fixed-point arithmetic along with stochastic
rounding has been proposed as a promising alternative to the commonly used
32-bit floating point arithmetic to enhance training neural networks training
in terms of performance and energy efficiency. In the first part of this paper,
the behaviour of the 12-bit fixed-point arithmetic when training a
convolutional neural network with the CIFAR-10 dataset is analysed, showing
that such arithmetic is not the most appropriate for the training phase. After
that, the paper presents and evaluates, under the same conditions, alternative
low-precision arithmetics, starting with the 12-bit floating-point arithmetic.
These two representations are then leveraged using local scaling in order to
increase accuracy and get closer to the baseline 32-bit floating-point
arithmetic. Finally, the paper introduces a simplified model in which both the
outputs and the gradients of the neural networks are constrained to
power-of-two values, just using 7 bits for their representation. The evaluation
demonstrates a minimal loss in accuracy for the proposed Power-of-Two neural
network, avoiding the use of multiplications and divisions and thereby,
significantly reducing the training time as well as the energy consumption and
memory requirements during the training and inference phases.
|
Adaptive Federated Learning in Resource Constrained Edge Computing
Systems | Emerging technologies and applications including Internet of Things (IoT),
social networking, and crowd-sourcing generate large amounts of data at the
network edge. Machine learning models are often built from the collected data,
to enable the detection, classification, and prediction of future events. Due
to bandwidth, storage, and privacy concerns, it is often impractical to send
all the data to a centralized location. In this paper, we consider the problem
of learning model parameters from data distributed across multiple edge nodes,
without sending raw data to a centralized place. Our focus is on a generic
class of machine learning models that are trained using gradient-descent based
approaches. We analyze the convergence bound of distributed gradient descent
from a theoretical point of view, based on which we propose a control algorithm
that determines the best trade-off between local update and global parameter
aggregation to minimize the loss function under a given resource budget. The
performance of the proposed algorithm is evaluated via extensive experiments
with real datasets, both on a networked prototype system and in a larger-scale
simulated environment. The experimentation results show that our proposed
approach performs near to the optimum with various machine learning models and
different data distributions.
|
OmicsMapNet: Transforming omics data to take advantage of Deep
Convolutional Neural Network for discovery | We developed OmicsMapNet approach to take advantage of existing deep leaning
frameworks to analyze high-dimensional omics data as 2-dimensional images. The
omics data of individual samples were first rearranged into 2D images in which
molecular features related in functions, ontologies, or other relationships
were organized in spatially adjacent and patterned locations. Deep learning
neural networks were trained to classify the images. Molecular features
informative of classes of different phenotypes were subsequently identified. As
an example, we used the KEGG BRITE database to rearrange RNA-Seq expression
data of TCGA diffuse glioma samples as treemaps to capture the functional
hierarchical structure of genes in 2D images. Deep Convolutional Neural
Networks (CNN) were derived using tools from TensorFlow to learn the grade of
TCGA LGG and GBM samples with relatively high accuracy. The most contributory
features in the trained CNN were confirmed in pathway analysis for their
plausible functional involvement.
|
Adversarial Attacks Against Medical Deep Learning Systems | The discovery of adversarial examples has raised concerns about the practical
deployment of deep learning systems. In this paper, we demonstrate that
adversarial examples are capable of manipulating deep learning systems across
three clinical domains. For each of our representative medical deep learning
classifiers, both white and black box attacks were highly successful. Our
models are representative of the current state of the art in medical computer
vision and, in some cases, directly reflect architectures already seeing
deployment in real world clinical settings. In addition to the technical
contribution of our paper, we synthesize a large body of knowledge about the
healthcare system to argue that medicine may be uniquely susceptible to
adversarial attacks, both in terms of monetary incentives and technical
vulnerability. To this end, we outline the healthcare economy and the
incentives it creates for fraud and provide concrete examples of how and why
such attacks could be realistically carried out. We urge practitioners to be
aware of current vulnerabilities when deploying deep learning systems in
clinical settings, and encourage the machine learning community to further
investigate the domain-specific characteristics of medical learning systems.
|
From CDF to PDF --- A Density Estimation Method for High Dimensional
Data | CDF2PDF is a method of PDF estimation by approximating CDF. The original idea
of it was previously proposed in [1] called SIC. However, SIC requires
additional hyper-parameter tunning, and no algorithms for computing higher
order derivative from a trained NN are provided in [1]. CDF2PDF improves SIC by
avoiding the time-consuming hyper-parameter tuning part and enabling higher
order derivative computation to be done in polynomial time. Experiments of this
method for one-dimensional data shows promising results.
|
Generative Adversarial Network based Autoencoder: Application to fault
detection problem for closed loop dynamical systems | Fault detection problem for closed loop uncertain dynamical systems, is
investigated in this paper, using different deep learning based methods.
Traditional classifier based method does not perform well, because of the
inherent difficulty of detecting system level faults for closed loop dynamical
system. Specifically, acting controller in any closed loop dynamical system,
works to reduce the effect of system level faults. A novel Generative
Adversarial based deep Autoencoder is designed to classify datasets under
normal and faulty operating conditions. This proposed network performs
significantly well when compared to any available classifier based methods, and
moreover, does not require labeled fault incorporated datasets for training
purpose. Finally, this aforementioned network's performance is tested on a high
complexity building energy system dataset.
|
Data-Dependent Coresets for Compressing Neural Networks with
Applications to Generalization Bounds | We present an efficient coresets-based neural network compression algorithm
that sparsifies the parameters of a trained fully-connected neural network in a
manner that provably approximates the network's output. Our approach is based
on an importance sampling scheme that judiciously defines a sampling
distribution over the neural network parameters, and as a result, retains
parameters of high importance while discarding redundant ones. We leverage a
novel, empirical notion of sensitivity and extend traditional coreset
constructions to the application of compressing parameters. Our theoretical
analysis establishes guarantees on the size and accuracy of the resulting
compressed network and gives rise to generalization bounds that may provide new
insights into the generalization properties of neural networks. We demonstrate
the practical effectiveness of our algorithm on a variety of neural network
configurations and real-world data sets.
|
Twin Regularization for online speech recognition | Online speech recognition is crucial for developing natural human-machine
interfaces. This modality, however, is significantly more challenging than
off-line ASR, since real-time/low-latency constraints inevitably hinder the use
of future information, that is known to be very helpful to perform robust
predictions. A popular solution to mitigate this issue consists of feeding
neural acoustic models with context windows that gather some future frames.
This introduces a latency which depends on the number of employed look-ahead
features. This paper explores a different approach, based on estimating the
future rather than waiting for it. Our technique encourages the hidden
representations of a unidirectional recurrent network to embed some useful
information about the future. Inspired by a recently proposed technique called
Twin Networks, we add a regularization term that forces forward hidden states
to be as close as possible to cotemporal backward ones, computed by a "twin"
neural network running backwards in time. The experiments, conducted on a
number of datasets, recurrent architectures, input features, and acoustic
conditions, have shown the effectiveness of this approach. One important
advantage is that our method does not introduce any additional computation at
test time if compared to standard unidirectional recurrent networks.
|
Approximating the covariance ellipsoid | We explore ways in which the covariance ellipsoid ${\cal B}=\{v \in
\mathbb{R}^d : \mathbb{E} <X,v>^2 \leq 1\}$ of a centred random vector $X$ in
$\mathbb{R}^d$ can be approximated by a simple set. The data one is given for
constructing the approximating set consists of $X_1,...,X_N$ that are
independent and distributed as $X$.
We present a general method that can be used to construct such approximations
and implement it for two types of approximating sets. We first construct a
(random) set ${\cal K}$ defined by a union of intersections of slabs
$H_{z,\alpha}=\{v \in \mathbb{R}^d : |<z,v>| \leq \alpha\}$ (and therefore
${\cal K}$ is actually the output of a neural network with two hidden layers).
The slabs are generated using $X_1,...,X_N$, and under minimal assumptions on
$X$ (e.g., $X$ can be heavy-tailed) it suffices that $N = c_1d
\eta^{-4}\log(2/\eta)$ to ensure that $(1-\eta) {\cal K} \subset {\cal B}
\subset (1+\eta){\cal K}$. In some cases (e.g., if $X$ is rotation invariant
and has marginals that are well behaved in some weak sense), a smaller sample
size suffices: $N = c_1d\eta^{-2}\log(2/\eta)$.
We then show that if the slabs are replaced by randomly generated ellipsoids
defined using $X_1,...,X_N$, the same degree of approximation is true when $N
\geq c_2d\eta^{-2}\log(2/\eta)$.
The construction we use is based on the small-ball method.
|
Adaptivity for Regularized Kernel Methods by Lepskii's Principle | We address the problem of {\it adaptivity} in the framework of reproducing
kernel Hilbert space (RKHS) regression. More precisely, we analyze estimators
arising from a linear regularization scheme $g_\lam$. In practical
applications, an important task is to choose the regularization parameter
$\lam$ appropriately, i.e. based only on the given data and independently on
unknown structural assumptions on the regression function. An attractive
approach avoiding data-splitting is the {\it Lepskii Principle} (LP), also
known as the {\it Balancing Principle} is this setting. We show that a modified
parameter choice based on (LP) is minimax optimal adaptive, up to
$\log\log(n)$. A convenient result is the fact that balancing in $L^2(\nu)-$
norm, which is easiest, automatically gives optimal balancing in all stronger
norms, interpolating between $L^2(\nu)$ and the RKHS. An analogous result is
open for other classical approaches to data dependent choices of the
regularization parameter, e.g. for Hold-Out.
|
A refinement of Bennett's inequality with applications to portfolio
optimization | A refinement of Bennett's inequality is introduced which is strictly tighter
than the classical bound. The new bound establishes the convergence of the
average of independent random variables to its expected value. It also
carefully exploits information about the potentially heterogeneous mean,
variance, and ceiling of each random variable. The bound is strictly sharper in
the homogeneous setting and very often significantly sharper in the
heterogeneous setting. The improved convergence rates are obtained by
leveraging Lambert's W function. We apply the new bound in a portfolio
optimization setting to allocate a budget across investments with heterogeneous
returns.
|
On Gradient-Based Learning in Continuous Games | We formulate a general framework for competitive gradient-based learning that
encompasses a wide breadth of multi-agent learning algorithms, and analyze the
limiting behavior of competitive gradient-based learning algorithms using
dynamical systems theory. For both general-sum and potential games, we
characterize a non-negligible subset of the local Nash equilibria that will be
avoided if each agent employs a gradient-based learning algorithm. We also shed
light on the issue of convergence to non-Nash strategies in general- and
zero-sum games, which may have no relevance to the underlying game, and arise
solely due to the choice of algorithm. The existence and frequency of such
strategies may explain some of the difficulties encountered when using gradient
descent in zero-sum games as, e.g., in the training of generative adversarial
networks. To reinforce the theoretical contributions, we provide empirical
results that highlight the frequency of linear quadratic dynamic games (a
benchmark for multi-agent reinforcement learning) that admit global Nash
equilibria that are almost surely avoided by policy gradient.
|
Composable Unpaired Image to Image Translation | There has been remarkable recent work in unpaired image-to-image translation.
However, they're restricted to translation on single pairs of distributions,
with some exceptions. In this study, we extend one of these works to a scalable
multidistribution translation mechanism. Our translation models not only
converts from one distribution to another but can be stacked to create
composite translation functions. We show that this composite property makes it
possible to generate images with characteristics not seen in the training set.
We also propose a decoupled training mechanism to train multiple distributions
separately, which we show, generates better samples than isolated joint
training. Further, we do a qualitative and quantitative analysis to assess the
plausibility of the samples. The code is made available at
https://github.com/lgraesser/im2im2im.
|
A Direct Sum Result for the Information Complexity of Learning | How many bits of information are required to PAC learn a class of hypotheses
of VC dimension $d$? The mathematical setting we follow is that of Bassily et
al. (2018), where the value of interest is the mutual information
$\mathrm{I}(S;A(S))$ between the input sample $S$ and the hypothesis outputted
by the learning algorithm $A$. We introduce a class of functions of VC
dimension $d$ over the domain $\mathcal{X}$ with information complexity at
least $\Omega\left(d\log \log \frac{|\mathcal{X}|}{d}\right)$ bits for any
consistent and proper algorithm (deterministic or random). Bassily et al.
proved a similar (but quantitatively weaker) result for the case $d=1$.
The above result is in fact a special case of a more general phenomenon we
explore. We define the notion of information complexity of a given class of
functions $\mathcal{H}$. Intuitively, it is the minimum amount of information
that an algorithm for $\mathcal{H}$ must retain about its input to ensure
consistency and properness. We prove a direct sum result for information
complexity in this context; roughly speaking, the information complexity sums
when combining several classes.
|
Binary Matrix Factorization via Dictionary Learning | Matrix factorization is a key tool in data analysis; its applications include
recommender systems, correlation analysis, signal processing, among others.
Binary matrices are a particular case which has received significant attention
for over thirty years, especially within the field of data mining. Dictionary
learning refers to a family of methods for learning overcomplete basis (also
called frames) in order to efficiently encode samples of a given type; this
area, now also about twenty years old, was mostly developed within the signal
processing field. In this work we propose two binary matrix factorization
methods based on a binary adaptation of the dictionary learning paradigm to
binary matrices. The proposed algorithms focus on speed and scalability; they
work with binary factors combined with bit-wise operations and a few auxiliary
integer ones. Furthermore, the methods are readily applicable to online binary
matrix factorization. Another important issue in matrix factorization is the
choice of rank for the factors; we address this model selection problem with an
efficient method based on the Minimum Description Length principle. Our
preliminary results show that the proposed methods are effective at producing
interpretable factorizations of various data types of different nature.
|
Block Mean Approximation for Efficient Second Order Optimization | Advanced optimization algorithms such as Newton method and AdaGrad benefit
from second order derivative or second order statistics to achieve better
descent directions and faster convergence rates. At their heart, such
algorithms need to compute the inverse or inverse square root of a matrix whose
size is quadratic of the dimensionality of the search space. For high
dimensional search spaces, the matrix inversion or inversion of square root
becomes overwhelming which in turn demands for approximate methods. In this
work, we propose a new matrix approximation method which divides a matrix into
blocks and represents each block by one or two numbers. The method allows
efficient computation of matrix inverse and inverse square root. We apply our
method to AdaGrad in training deep neural networks. Experiments show
encouraging results compared to the diagonal approximation.
|
conformalClassification: A Conformal Prediction R Package for
Classification | The conformalClassification package implements Transductive Conformal
Prediction (TCP) and Inductive Conformal Prediction (ICP) for classification
problems. Conformal Prediction (CP) is a framework that complements the
predictions of machine learning algorithms with reliable measures of
confidence. TCP gives results with higher validity than ICP, however ICP is
computationally faster than TCP. The package conformalClassification is built
upon the random forest method, where votes of the random forest for each class
are considered as the conformity scores for each data point. Although the main
aim of the conformalClassification package is to generate CP errors (p-values)
for classification problems, the package also implements various diagnostic
measures such as deviation from validity, error rate, efficiency, observed
fuzziness and calibration plots. In future releases, we plan to extend the
package to use other machine learning algorithms, (e.g. support vector
machines) for model fitting.
|
Deep Learning on Key Performance Indicators for Predictive Maintenance
in SAP HANA | With a new era of cloud and big data, Database Management Systems (DBMSs)
have become more crucial in numerous enterprise business applications in all
the industries. Accordingly, the importance of their proactive and preventive
maintenance has also increased. However, detecting problems by predefined rules
or stochastic modeling has limitations, particularly when analyzing the data on
high-dimensional Key Performance Indicators (KPIs) from a DBMS. In recent
years, Deep Learning (DL) has opened new opportunities for this complex
analysis. In this paper, we present two complementary DL approaches to detect
anomalies in SAP HANA. A temporal learning approach is used to detect abnormal
patterns based on unlabeled historical data, whereas a spatial learning
approach is used to classify known anomalies based on labeled data. We
implement a system in SAP HANA integrated with Google TensorFlow. The
experimental results with real-world data confirm the effectiveness of the
system and models.
|
Learning Simple Thresholded Features with Sparse Support Recovery | The thresholded feature has recently emerged as an extremely efficient, yet
rough empirical approximation, of the time-consuming sparse coding inference
process. Such an approximation has not yet been rigorously examined, and
standard dictionaries often lead to non-optimal performance when used for
computing thresholded features. In this paper, we first present two theoretical
recovery guarantees for the thresholded feature to exactly recover the nonzero
support of the sparse code. Motivated by them, we then formulate the Dictionary
Learning for Thresholded Features (DLTF) model, which learns an optimized
dictionary for applying the thresholded feature. In particular, for the $(k,
2)$ norm involved, a novel proximal operator with log-linear time complexity
$O(m\log m)$ is derived. We evaluate the performance of DLTF on a vast range of
synthetic and real-data tasks, where DLTF demonstrates remarkable efficiency,
effectiveness and robustness in all experiments. In addition, we briefly
discuss the potential link between DLTF and deep learning building blocks.
|
Building robust prediction models for defective sensor data using
Artificial Neural Networks | Predicting the health of components in complex dynamic systems such as an
automobile poses numerous challenges. The primary aim of such predictive
systems is to use the high-dimensional data acquired from different sensors and
predict the state-of-health of a particular component, e.g., brake pad. The
classical approach involves selecting a smaller set of relevant sensor signals
using feature selection and using them to train a machine learning algorithm.
However, this fails to address two prominent problems: (1) sensors are
susceptible to failure when exposed to extreme conditions over a long periods
of time; (2) sensors are electrical devices that can be affected by noise or
electrical interference. Using the failed and noisy sensor signals as inputs
largely reduce the prediction accuracy. To tackle this problem, it is
advantageous to use the information from all sensor signals, so that the
failure of one sensor can be compensated by another. In this work, we propose
an Artificial Neural Network (ANN) based framework to exploit the information
from a large number of signals. Secondly, our framework introduces a data
augmentation approach to perform accurate predictions in spite of noisy
signals. The plausibility of our framework is validated on real life industrial
application from Robert Bosch GmbH.
|
Constant Step Size Stochastic Gradient Descent for Probabilistic
Modeling | Stochastic gradient methods enable learning probabilistic models from large
amounts of data. While large step-sizes (learning rates) have shown to be best
for least-squares (e.g., Gaussian noise) once combined with parameter
averaging, these are not leading to convergent algorithms in general. In this
paper, we consider generalized linear models, that is, conditional models based
on exponential families. We propose averaging moment parameters instead of
natural parameters for constant-step-size stochastic gradient descent. For
finite-dimensional models, we show that this can sometimes (and surprisingly)
lead to better predictions than the best linear model. For infinite-dimensional
models, we show that it always converges to optimal predictions, while
averaging natural parameters never does. We illustrate our findings with
simulations on synthetic data and classical benchmarks with many observations.
|
SPSA-FSR: Simultaneous Perturbation Stochastic Approximation for Feature
Selection and Ranking | This manuscript presents the following: (1) an improved version of the Binary
Simultaneous Perturbation Stochastic Approximation (SPSA) Method for feature
selection in machine learning (Aksakalli and Malekipirbazari, Pattern
Recognition Letters, Vol. 75, 2016) based on non-monotone iteration gains
computed via the Barzilai and Borwein (BB) method, (2) its adaptation for
feature ranking, and (3) comparison against popular methods on public benchmark
datasets. The improved method, which we call SPSA-FSR, dramatically reduces the
number of iterations required for convergence without impacting solution
quality. SPSA-FSR can be used for feature ranking and feature selection both
for classification and regression problems. After a review of the current
state-of-the-art, we discuss our improvements in detail and present three sets
of computational experiments: (1) comparison of SPSA-FS as a (wrapper) feature
selection method against sequential methods as well as genetic algorithms, (2)
comparison of SPSA-FS as a feature ranking method in a classification setting
against random forest importance, chi-squared, and information main methods,
and (3) comparison of SPSA-FS as a feature ranking method in a regression
setting against minimum redundancy maximum relevance (MRMR), RELIEF, and linear
correlation methods. The number of features in the datasets we use range from a
few dozens to a few thousands. Our results indicate that SPSA-FS converges to a
good feature set in no more than 100 iterations and therefore it is quite fast
for a wrapper method. SPSA-FS also outperforms popular feature selection as
well as feature ranking methods in majority of test cases, sometimes by a large
margin, and it stands as a promising new feature selection and ranking method.
|
Developing Synthesis Flows Without Human Knowledge | Design flows are the explicit combinations of design transformations,
primarily involved in synthesis, placement and routing processes, to accomplish
the design of Integrated Circuits (ICs) and System-on-Chip (SoC). Mostly, the
flows are developed based on the knowledge of the experts. However, due to the
large search space of design flows and the increasing design complexity,
developing Intellectual Property (IP)-specific synthesis flows providing high
Quality of Result (QoR) is extremely challenging. This work presents a fully
autonomous framework that artificially produces design-specific synthesis flows
without human guidance and baseline flows, using Convolutional Neural Network
(CNN). The demonstrations are made by successfully designing logic synthesis
flows of three large scaled designs.
|
RFCDE: Random Forests for Conditional Density Estimation | Random forests is a common non-parametric regression technique which performs
well for mixed-type data and irrelevant covariates, while being robust to
monotonic variable transformations. Existing random forest implementations
target regression or classification. We introduce the RFCDE package for fitting
random forest models optimized for nonparametric conditional density
estimation, including joint densities for multiple responses. This enables
analysis of conditional probability distributions which is useful for
propagating uncertainty and of joint distributions that describe relationships
between multiple responses and covariates. RFCDE is released under the MIT
open-source license and can be accessed at https://github.com/tpospisi/rfcde .
Both R and Python versions, which call a common C++ library, are available.
|
BELIEF: A distance-based redundancy-proof feature selection method for
Big Data | With the advent of Big Data era, data reduction methods are highly demanded
given its ability to simplify huge data, and ease complex learning processes.
Concretely, algorithms that are able to filter relevant dimensions from a set
of millions are of huge importance. Although effective, these techniques suffer
from the "scalability" curse as well.
In this work, we propose a distributed feature weighting algorithm, which is
able to rank millions of features in parallel using large samples. This method,
inspired by the well-known RELIEF algorithm, introduces a novel redundancy
elimination measure that provides similar schemes to those based on entropy at
a much lower cost. It also allows smooth scale up when more instances are
demanded in feature estimations. Empirical tests performed on our method show
its estimation ability in manifold huge sets --both in number of features and
instances--, as well as its simplified runtime cost (specially, at the
redundancy detection step).
|
Global Robustness Evaluation of Deep Neural Networks with Provable
Guarantees for the $L_0$ Norm | Deployment of deep neural networks (DNNs) in safety- or security-critical
systems requires provable guarantees on their correct behaviour. A common
requirement is robustness to adversarial perturbations in a neighbourhood
around an input. In this paper we focus on the $L_0$ norm and aim to compute,
for a trained DNN and an input, the maximal radius of a safe norm ball around
the input within which there are no adversarial examples. Then we define global
robustness as an expectation of the maximal safe radius over a test data set.
We first show that the problem is NP-hard, and then propose an approximate
approach to iteratively compute lower and upper bounds on the network's
robustness. The approach is \emph{anytime}, i.e., it returns intermediate
bounds and robustness estimates that are gradually, but strictly, improved as
the computation proceeds; \emph{tensor-based}, i.e., the computation is
conducted over a set of inputs simultaneously, instead of one by one, to enable
efficient GPU computation; and has \emph{provable guarantees}, i.e., both the
bounds and the robustness estimates can converge to their optimal values.
Finally, we demonstrate the utility of the proposed approach in practice to
compute tight bounds by applying and adapting the anytime algorithm to a set of
challenging problems, including global robustness evaluation, competitive $L_0$
attacks, test case generation for DNNs, and local robustness evaluation on
large-scale ImageNet DNNs. We release the code of all case studies via GitHub.
|
Deep Embedding Kernel | In this paper, we propose a novel supervised learning method that is called
Deep Embedding Kernel (DEK). DEK combines the advantages of deep learning and
kernel methods in a unified framework. More specifically, DEK is a learnable
kernel represented by a newly designed deep architecture. Compared with
pre-defined kernels, this kernel can be explicitly trained to map data to an
optimized high-level feature space where data may have favorable features
toward the application. Compared with typical deep learning using SoftMax or
logistic regression as the top layer, DEK is expected to be more generalizable
to new data. Experimental results show that DEK has superior performance than
typical machine learning methods in identity detection, classification,
regression, dimension reduction, and transfer learning.
|
ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object
Detector | Given the ability to directly manipulate image pixels in the digital input
space, an adversary can easily generate imperceptible perturbations to fool a
Deep Neural Network (DNN) image classifier, as demonstrated in prior work. In
this work, we propose ShapeShifter, an attack that tackles the more challenging
problem of crafting physical adversarial perturbations to fool image-based
object detectors like Faster R-CNN. Attacking an object detector is more
difficult than attacking an image classifier, as it needs to mislead the
classification results in multiple bounding boxes with different scales.
Extending the digital attack to the physical world adds another layer of
difficulty, because it requires the perturbation to be robust enough to survive
real-world distortions due to different viewing distances and angles, lighting
conditions, and camera limitations. We show that the Expectation over
Transformation technique, which was originally proposed to enhance the
robustness of adversarial perturbations in image classification, can be
successfully adapted to the object detection setting. ShapeShifter can generate
adversarially perturbed stop signs that are consistently mis-detected by Faster
R-CNN as other objects, posing a potential threat to autonomous vehicles and
other safety-critical computer vision systems.
|
Models for Capturing Temporal Smoothness in Evolving Networks for
Learning Latent Representation of Nodes | In a dynamic network, the neighborhood of the vertices evolve across
different temporal snapshots of the network. Accurate modeling of this temporal
evolution can help solve complex tasks involving real-life social and
interaction networks. However, existing models for learning latent
representation are inadequate for obtaining the representation vectors of the
vertices for different time-stamps of a dynamic network in a meaningful way. In
this paper, we propose latent representation learning models for dynamic
networks which overcome the above limitation by considering two different kinds
of temporal smoothness: (i) retrofitted, and (ii) linear transformation. The
retrofitted model tracks the representation vector of a vertex over time,
facilitating vertex-based temporal analysis of a network. On the other hand,
linear transformation based model provides a smooth transition operator which
maps the representation vectors of all vertices from one temporal snapshot to
the next (unobserved) snapshot-this facilitates prediction of the state of a
network in a future time-stamp. We validate the performance of our proposed
models by employing them for solving the temporal link prediction task.
Experiments on 9 real-life networks from various domains validate that the
proposed models are significantly better than the existing models for
predicting the dynamics of an evolving network.
|
CytonRL: an Efficient Reinforcement Learning Open-source Toolkit
Implemented in C++ | This paper presents an open-source enforcement learning toolkit named CytonRL
(https://github.com/arthurxlw/cytonRL). The toolkit implements four recent
advanced deep Q-learning algorithms from scratch using C++ and NVIDIA's
GPU-accelerated libraries. The code is simple and elegant, owing to an
open-source general-purpose neural network library named CytonLib. Benchmark
shows that the toolkit achieves competitive performances on the popular Atari
game of Breakout.
|
Walk-Steered Convolution for Graph Classification | Graph classification is a fundamental but challenging issue for numerous
real-world applications. Despite recent great progress in image/video
classification, convolutional neural networks (CNNs) cannot yet cater to graphs
well because of graphical non-Euclidean topology. In this work, we propose a
walk-steered convolutional (WSC) network to assemble the essential success of
standard convolutional neural networks as well as the powerful representation
ability of random walk. Instead of deterministic neighbor searching used in
previous graphical CNNs, we construct multi-scale walk fields (a.k.a. local
receptive fields) with random walk paths to depict subgraph structures and
advocate graph scalability. To express the internal variations of a walk field,
Gaussian mixture models are introduced to encode principal components of walk
paths therein. As an analogy to a standard convolution kernel on image,
Gaussian models implicitly coordinate those unordered vertices/nodes and edges
in a local receptive field after projecting to the gradient space of Gaussian
parameters. We further stack graph coarsening upon Gaussian encoding by using
dynamic clustering, such that high-level semantics of graph can be well learned
like the conventional pooling on image. The experimental results on several
public datasets demonstrate the superiority of our proposed WSC method over
many state-of-the-arts for graph classification.
|
BigDL: A Distributed Deep Learning Framework for Big Data | This paper presents BigDL (a distributed deep learning framework for Apache
Spark), which has been used by a variety of users in the industry for building
deep learning applications on production big data platforms. It allows deep
learning applications to run on the Apache Hadoop/Spark cluster so as to
directly process the production data, and as a part of the end-to-end data
analysis pipeline for deployment and management. Unlike existing deep learning
frameworks, BigDL implements distributed, data parallel training directly on
top of the functional compute model (with copy-on-write and coarse-grained
operations) of Spark. We also share real-world experience and "war stories" of
users that have adopted BigDL to address their challenges(i.e., how to easily
build end-to-end data analysis and deep learning pipelines for their production
data).
|
Non-Vacuous Generalization Bounds at the ImageNet Scale: A PAC-Bayesian
Compression Approach | Modern neural networks are highly overparameterized, with capacity to
substantially overfit to training data. Nevertheless, these networks often
generalize well in practice. It has also been observed that trained networks
can often be "compressed" to much smaller representations. The purpose of this
paper is to connect these two empirical observations. Our main technical result
is a generalization bound for compressed networks based on the compressed size.
Combined with off-the-shelf compression algorithms, the bound leads to state of
the art generalization guarantees; in particular, we provide the first
non-vacuous generalization guarantees for realistic architectures applied to
the ImageNet classification problem. As additional evidence connecting
compression and generalization, we show that compressibility of models that
tend to overfit is limited: We establish an absolute limit on expected
compressibility as a function of expected generalization error, where the
expectations are over the random choice of training examples. The bounds are
complemented by empirical results that show an increase in overfitting implies
an increase in the number of bits required to describe a trained network.
|
Neural Models for Reasoning over Multiple Mentions using Coreference | Many problems in NLP require aggregating information from multiple mentions
of the same entity which may be far apart in the text. Existing Recurrent
Neural Network (RNN) layers are biased towards short-term dependencies and
hence not suited to such tasks. We present a recurrent layer which is instead
biased towards coreferent dependencies. The layer uses coreference annotations
extracted from an external system to connect entity mentions belonging to the
same cluster. Incorporating this layer into a state-of-the-art reading
comprehension model improves performance on three datasets -- Wikihop, LAMBADA
and the bAbi AI tasks -- with large gains when training data is scarce.
|
UCBoost: A Boosting Approach to Tame Complexity and Optimality for
Stochastic Bandits | In this work, we address the open problem of finding low-complexity
near-optimal multi-armed bandit algorithms for sequential decision making
problems. Existing bandit algorithms are either sub-optimal and computationally
simple (e.g., UCB1) or optimal and computationally complex (e.g., kl-UCB). We
propose a boosting approach to Upper Confidence Bound based algorithms for
stochastic bandits, that we call UCBoost. Specifically, we propose two types of
UCBoost algorithms. We show that UCBoost($D$) enjoys $O(1)$ complexity for each
arm per round as well as regret guarantee that is $1/e$-close to that of the
kl-UCB algorithm. We propose an approximation-based UCBoost algorithm,
UCBoost($\epsilon$), that enjoys a regret guarantee $\epsilon$-close to that of
kl-UCB as well as $O(\log(1/\epsilon))$ complexity for each arm per round.
Hence, our algorithms provide practitioners a practical way to trade optimality
with computational complexity. Finally, we present numerical results which show
that UCBoost($\epsilon$) can achieve the same regret performance as the
standard kl-UCB while incurring only $1\%$ of the computational cost of kl-UCB.
|
MaxGain: Regularisation of Neural Networks by Constraining Activation
Magnitudes | Effective regularisation of neural networks is essential to combat
overfitting due to the large number of parameters involved. We present an
empirical analogue to the Lipschitz constant of a feed-forward neural network,
which we refer to as the maximum gain. We hypothesise that constraining the
gain of a network will have a regularising effect, similar to how constraining
the Lipschitz constant of a network has been shown to improve generalisation. A
simple algorithm is provided that involves rescaling the weight matrix of each
layer after each parameter update. We conduct a series of studies on common
benchmark datasets, and also a novel dataset that we introduce to enable easier
significance testing for experiments using convolutional networks. Performance
on these datasets compares favourably with other common regularisation
techniques.
|
A Univariate Bound of Area Under ROC | Area under ROC (AUC) is an important metric for binary classification and
bipartite ranking problems. However, it is difficult to directly optimizing AUC
as a learning objective, so most existing algorithms are based on optimizing a
surrogate loss to AUC. One significant drawback of these surrogate losses is
that they require pairwise comparisons among training data, which leads to slow
running time and increasing local storage for online learning. In this work, we
describe a new surrogate loss based on a reformulation of the AUC risk, which
does not require pairwise comparison but rankings of the predictions. We
further show that the ranking operation can be avoided, and the learning
objective obtained based on this surrogate enjoys linear complexity in time and
storage. We perform experiments to demonstrate the effectiveness of the online
and batch algorithms for AUC optimization based on the proposed surrogate loss.
|
Model-Free Linear Quadratic Control via Reduction to Expert Prediction | Model-free approaches for reinforcement learning (RL) and continuous control
find policies based only on past states and rewards, without fitting a model of
the system dynamics. They are appealing as they are general purpose and easy to
implement; however, they also come with fewer theoretical guarantees than
model-based RL. In this work, we present a new model-free algorithm for
controlling linear quadratic (LQ) systems, and show that its regret scales as
$O(T^{\xi+2/3})$ for any small $\xi>0$ if time horizon satisfies $T>C^{1/\xi}$
for a constant $C$. The algorithm is based on a reduction of control of Markov
decision processes to an expert prediction problem. In practice, it corresponds
to a variant of policy iteration with forced exploration, where the policy in
each phase is greedy with respect to the average of all previous value
functions. This is the first model-free algorithm for adaptive control of LQ
systems that provably achieves sublinear regret and has a polynomial
computation cost. Empirically, our algorithm dramatically outperforms standard
policy iteration, but performs worse than a model-based approach.
|
A Boosting Framework of Factorization Machine | Recently, Factorization Machines (FM) has become more and more popular for
recommendation systems, due to its effectiveness in finding informative
interactions between features. Usually, the weights for the interactions is
learnt as a low rank weight matrix, which is formulated as an inner product of
two low rank matrices. This low rank can help improve the generalization
ability of Factorization Machines. However, to choose the rank properly, it
usually needs to run the algorithm for many times using different ranks, which
clearly is inefficient for some large-scale datasets. To alleviate this issue,
we propose an Adaptive Boosting framework of Factorization Machines (AdaFM),
which can adaptively search for proper ranks for different datasets without
re-training. Instead of using a fixed rank for FM, the proposed algorithm will
adaptively gradually increases its rank according to its performance until the
performance does not grow, using boosting strategy. To verify the performance
of our proposed framework, we conduct an extensive set of experiments on many
real-world datasets. Encouraging empirical results shows that the proposed
algorithms are generally more effective than state-of-the-art other
Factorization Machines.
|
Deep Learning on Operational Facility Data Related to Large-Scale
Distributed Area Scientific Workflows | Distributed computing platforms provide a robust mechanism to perform
large-scale computations by splitting the task and data among multiple
locations, possibly located thousands of miles apart geographically. Although
such distribution of resources can lead to benefits, it also comes with its
associated problems such as rampant duplication of file transfers increasing
congestion, long job completion times, unexpected site crashing, suboptimal
data transfer rates, unpredictable reliability in a time range, and suboptimal
usage of storage elements. In addition, each sub-system becomes a potential
failure node that can trigger system wide disruptions. In this vision paper, we
outline our approach to leveraging Deep Learning algorithms to discover
solutions to unique problems that arise in a system with computational
infrastructure that is spread over a wide area. The presented vision, motivated
by a real scientific use case from Belle II experiments, is to develop
multilayer neural networks to tackle forecasting, anomaly detection and
optimization challenges in a complex and distributed data movement environment.
Through this vision based on Deep Learning principles, we aim to achieve
reduced congestion events, faster file transfer rates, and enhanced site
reliability.
|
Parametric Models for Mutual Kernel Matrix Completion | Recent studies utilize multiple kernel learning to deal with incomplete-data
problem. In this study, we introduce new methods that do not only complete
multiple incomplete kernel matrices simultaneously, but also allow control of
the flexibility of the model by parameterizing the model matrix. By imposing
restrictions on the model covariance, overfitting of the data is avoided. A
limitation of kernel matrix estimations done via optimization of an objective
function is that the positive definiteness of the result is not guaranteed. In
view of this limitation, our proposed methods employ the LogDet divergence,
which ensures the positive definiteness of the resulting inferred kernel
matrix. We empirically show that our proposed restricted covariance models,
employed with LogDet divergence, yield significant improvements in the
generalization performance of previous completion methods.
|
A Support Tensor Train Machine | There has been growing interest in extending traditional vector-based machine
learning techniques to their tensor forms. An example is the support tensor
machine (STM) that utilizes a rank-one tensor to capture the data structure,
thereby alleviating the overfitting and curse of dimensionality problems in the
conventional support vector machine (SVM). However, the expressive power of a
rank-one tensor is restrictive for many real-world data. To overcome this
limitation, we introduce a support tensor train machine (STTM) by replacing the
rank-one tensor in an STM with a tensor train. Experiments validate and confirm
the superiority of an STTM over the SVM and STM.
|
VC-Dimension Based Generalization Bounds for Relational Learning | In many applications of relational learning, the available data can be seen
as a sample from a larger relational structure (e.g. we may be given a small
fragment from some social network). In this paper we are particularly concerned
with scenarios in which we can assume that (i) the domain elements appearing in
the given sample have been uniformly sampled without replacement from the
(unknown) full domain and (ii) the sample is complete for these domain elements
(i.e. it is the full substructure induced by these elements). Within this
setting, we study bounds on the error of sufficient statistics of relational
models that are estimated on the available data. As our main result, we prove a
bound based on a variant of the Vapnik-Chervonenkis dimension which is suitable
for relational data.
|
MetaBags: Bagged Meta-Decision Trees for Regression | Ensembles are popular methods for solving practical supervised learning
problems. They reduce the risk of having underperforming models in
production-grade software. Although critical, methods for learning
heterogeneous regression ensembles have not been proposed at large scale,
whereas in classical ML literature, stacking, cascading and voting are mostly
restricted to classification problems. Regression poses distinct learning
challenges that may result in poor performance, even when using well
established homogeneous ensemble schemas such as bagging or boosting.
In this paper, we introduce MetaBags, a novel, practically useful stacking
framework for regression. MetaBags is a meta-learning algorithm that learns a
set of meta-decision trees designed to select one base model (i.e. expert) for
each query, and focuses on inductive bias reduction. A set of meta-decision
trees are learned using different types of meta-features, specially created for
this purpose - to then be bagged at meta-level. This procedure is designed to
learn a model with a fair bias-variance trade-off, and its improvement over
base model performance is correlated with the prediction diversity of different
experts on specific input space subregions. The proposed method and
meta-features are designed in such a way that they enable good predictive
performance even in subregions of space which are not adequately represented in
the available training data.
An exhaustive empirical testing of the method was performed, evaluating both
generalization error and scalability of the approach on synthetic, open and
real-world application datasets. The obtained results show that our method
significantly outperforms existing state-of-the-art approaches.
|
Learning Sparse Latent Representations with the Deep Copula Information
Bottleneck | Deep latent variable models are powerful tools for representation learning.
In this paper, we adopt the deep information bottleneck model, identify its
shortcomings and propose a model that circumvents them. To this end, we apply a
copula transformation which, by restoring the invariance properties of the
information bottleneck method, leads to disentanglement of the features in the
latent space. Building on that, we show how this transformation translates to
sparsity of the latent space in the new model. We evaluate our method on
artificial and real data.
|
Hierarchical correlation reconstruction with missing data, for example
for biology-inspired neuron | Machine learning often needs to model density from a multidimensional data
sample, including correlations between coordinates. Additionally, we often have
missing data case: that data points can miss values for some of coordinates.
This article adapts rapid parametric density estimation approach for this
purpose: modelling density as a linear combination of orthonormal functions,
for which $L^2$ optimization says that (independently) estimated coefficient
for a given function is just average over the sample of value of this function.
Hierarchical correlation reconstruction first models probability density for
each separate coordinate using all its appearances in data sample, then adds
corrections from independently modelled pairwise correlations using all samples
having both coordinates, and so on independently adding correlations for
growing numbers of variables using often decreasing evidence in data sample. A
basic application of such modelled multidimensional density can be imputation
of missing coordinates: by inserting known coordinates to the density, and
taking expected values for the missing coordinates, or even their entire joint
probability distribution. Presented method can be compared with cascade
correlations approach, offering several advantages in flexibility and accuracy.
It can be also used as artificial neuron: maximizing prediction capabilities
for only local behavior - modelling and predicting local connections.
|
A Comparison of Machine Learning Algorithms for the Surveillance of
Autism Spectrum Disorder | The Centers for Disease Control and Prevention (CDC) coordinates a
labor-intensive process to measure the prevalence of autism spectrum disorder
(ASD) among children in the United States. Random forests methods have shown
promise in speeding up this process, but they lag behind human classification
accuracy by about 5%. We explore whether more recently available document
classification algorithms can close this gap. We applied 8 supervised learning
algorithms to predict whether children meet the case definition for ASD based
solely on the words in their evaluations. We compared the algorithms'
performance across 10 random train-test splits of the data, using
classification accuracy, F1 score, and number of positive calls to evaluate
their potential use for surveillance. Across the 10 train-test cycles, the
random forest and support vector machine with Naive Bayes features (NB-SVM)
each achieved slightly more than 87% mean accuracy. The NB-SVM produced
significantly more false negatives than false positives (P = 0.027), but the
random forest did not, making its prevalence estimates very close to the true
prevalence in the data. The best-performing neural network performed similarly
to the random forest on both measures. The random forest performed as well as
more recently available models like the NB-SVM and the neural network, and it
also produced good prevalence estimates. NB-SVM may not be a good candidate for
use in a fully-automated surveillance workflow due to increased false
negatives. More sophisticated algorithms, like hierarchical convolutional
neural networks, may not be feasible to train due to characteristics of the
data. Current algorithms might perform better if the data are abstracted and
processed differently and if they take into account information about the
children in addition to their evaluations.
|
Cluster Analysis on Locally Asymptotically Self-similar Processes with
Known Number of Clusters | We conduct cluster analysis on a class of locally asymptotically self-similar
stochastic processes, which includes multifractional Brownian motion as a
representative. When the true number of clusters is supposed to be known, a new
covariance-based dissimilarity measure is introduced, from which we obtain the
approximately asymptotically consistent clustering algorithms. In simulation
studies, clustering data sampled from multifractional Brownian motions with
distinct functional Hurst parameters illustrates the approximated asymptotic
consistency of the proposed algorithms. Clustering global financial markets'
equity indexes returns and sovereign CDS spreads provides a successful real
world application.
|
Analysis of Extremely Obese Individuals Using Deep Learning Stacked
Autoencoders and Genome-Wide Genetic Data | The aetiology of polygenic obesity is multifactorial, which indicates that
life-style and environmental factors may influence multiples genes to aggravate
this disorder. Several low-risk single nucleotide polymorphisms (SNPs) have
been associated with BMI. However, identified loci only explain a small
proportion of the variation ob-served for this phenotype. The linear nature of
genome wide association studies (GWAS) used to identify associations between
genetic variants and the phenotype have had limited success in explaining the
heritability variation of BMI and shown low predictive capacity in
classification studies. GWAS ignores the epistatic interactions that less
significant variants have on the phenotypic outcome. In this paper we utilise a
novel deep learning-based methodology to reduce the high dimensional space in
GWAS and find epistatic interactions between SNPs for classification purposes.
SNPs were filtered based on the effects associations have with BMI. Since
Bonferroni adjustment for multiple testing is highly conservative, an important
proportion of SNPs involved in SNP-SNP interactions are ignored. Therefore,
only SNPs with p-values < 1x10-2 were considered for subsequent epistasis
analysis using stacked auto encoders (SAE). This allows the nonlinearity
present in SNP-SNP interactions to be discovered through progressively smaller
hidden layer units and to initialise a multi-layer feedforward artificial
neural network (ANN) classifier. The classifier is fine-tuned to classify
extremely obese and non-obese individuals. The best results were obtained with
2000 compressed units (SE=0.949153, SP=0.933014, Gini=0.949936,
Lo-gloss=0.1956, AUC=0.97497 and MSE=0.054057). Using 50 compressed units it
was possible to achieve (SE=0.785311, SP=0.799043, Gini=0.703566,
Logloss=0.476864, AUC=0.85178 and MSE=0.156315).
|
PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in
Spatiotemporal Predictive Learning | We present PredRNN++, an improved recurrent network for video predictive
learning. In pursuit of a greater spatiotemporal modeling capability, our
approach increases the transition depth between adjacent states by leveraging a
novel recurrent unit, which is named Causal LSTM for re-organizing the spatial
and temporal memories in a cascaded mechanism. However, there is still a
dilemma in video predictive learning: increasingly deep-in-time models have
been designed for capturing complex variations, while introducing more
difficulties in the gradient back-propagation. To alleviate this undesirable
effect, we propose a Gradient Highway architecture, which provides alternative
shorter routes for gradient flows from outputs back to long-range inputs. This
architecture works seamlessly with causal LSTMs, enabling PredRNN++ to capture
short-term and long-term dependencies adaptively. We assess our model on both
synthetic and real video datasets, showing its ability to ease the vanishing
gradient problem and yield state-of-the-art prediction results even in a
difficult objects occlusion scenario.
|
On Improving Deep Reinforcement Learning for POMDPs | Deep Reinforcement Learning (RL) recently emerged as one of the most
competitive approaches for learning in sequential decision making problems with
fully observable environments, e.g., computer Go. However, very little work has
been done in deep RL to handle partially observable environments. We propose a
new architecture called Action-specific Deep Recurrent Q-Network (ADRQN) to
enhance learning performance in partially observable domains. Actions are
encoded by a fully connected layer and coupled with a convolutional observation
to form an action-observation pair. The time series of action-observation pairs
are then integrated by an LSTM layer that learns latent states based on which a
fully connected layer computes Q-values as in conventional Deep Q-Networks
(DQNs). We demonstrate the effectiveness of our new architecture in several
partially observable domains, including flickering Atari games.
|
High Dimensional Time Series Generators | Multidimensional time series are sequences of real valued vectors. They occur
in different areas, for example handwritten characters, GPS tracking, and
gestures of modern virtual reality motion controllers. Within these areas, a
common task is to search for similar time series. Dynamic Time Warping (DTW) is
a common distance function to compare two time series. The Edit Distance with
Real Penalty (ERP) and the Dog Keeper Distance (DK) are two more distance
functions on time series. Their behaviour has been analyzed on 1-dimensional
time series. However, it is not easy to evaluate their behaviour in relation to
growing dimensionality. For this reason we propose two new data synthesizers
generating multidimensional time series. The first synthesizer extends the well
known cylinder-bell-funnel (CBF) dataset to multidimensional time series. Here,
each time series has an arbitrary type (cylinder, bell, or funnel) in each
dimension, thus for $d$-dimensional time series there are $3^{d}$ different
classes. The second synthesizer (RAM) creates time series with ideas adapted
from Brownian motions which is a common model of movement in physics. Finally,
we evaluate the applicability of a 1-nearest neighbor classifier using DTW on
datasets generated by our synthesizers.
|
Graph-based Selective Outlier Ensembles | An ensemble technique is characterized by the mechanism that generates the
components and by the mechanism that combines them. A common way to achieve the
consensus is to enable each component to equally participate in the aggregation
process. A problem with this approach is that poor components are likely to
negatively affect the quality of the consensus result. To address this issue,
alternatives have been explored in the literature to build selective classifier
and cluster ensembles, where only a subset of the components contributes to the
computation of the consensus. Of the family of ensemble methods, outlier
ensembles are the least studied. Only recently, the selection problem for
outlier ensembles has been discussed. In this work we define a new graph-based
class of ranking selection methods. A method in this class is characterized by
two main steps: (1) Mapping the rankings onto a graph structure; and (2) Mining
the resulting graph to identify a subset of rankings. We define a specific
instance of the graph-based ranking selection class. Specifically, we map the
problem of selecting ensemble components onto a mining problem in a graph. An
extensive evaluation was conducted on a variety of heterogeneous data and
methods. Our empirical results show that our approach outperforms
state-of-the-art selective outlier ensemble techniques.
|
Multi-Reward Reinforced Summarization with Saliency and Entailment | Abstractive text summarization is the task of compressing and rewriting a
long document into a short summary while maintaining saliency, directed logical
entailment, and non-redundancy. In this work, we address these three important
aspects of a good summary via a reinforcement learning approach with two novel
reward functions: ROUGESal and Entail, on top of a coverage-based baseline. The
ROUGESal reward modifies the ROUGE metric by up-weighting the salient
phrases/words detected via a keyphrase classifier. The Entail reward gives high
(length-normalized) scores to logically-entailed summaries using an entailment
classifier. Further, we show superior performance improvement when these
rewards are combined with traditional metric (ROUGE) based rewards, via our
novel and effective multi-reward approach of optimizing multiple rewards
simultaneously in alternate mini-batches. Our method achieves the new
state-of-the-art results (including human evaluation) on the CNN/Daily Mail
dataset as well as strong improvements in a test-only transfer setup on
DUC-2002.
|
On Learning Intrinsic Rewards for Policy Gradient Methods | In many sequential decision making tasks, it is challenging to design reward
functions that help an RL agent efficiently learn behavior that is considered
good by the agent designer. A number of different formulations of the
reward-design problem, or close variants thereof, have been proposed in the
literature. In this paper we build on the Optimal Rewards Framework of Singh
et.al. that defines the optimal intrinsic reward function as one that when used
by an RL agent achieves behavior that optimizes the task-specifying or
extrinsic reward function. Previous work in this framework has shown how good
intrinsic reward functions can be learned for lookahead search based planning
agents. Whether it is possible to learn intrinsic reward functions for learning
agents remains an open problem. In this paper we derive a novel algorithm for
learning intrinsic rewards for policy-gradient based learning agents. We
compare the performance of an augmented agent that uses our algorithm to
provide additive intrinsic rewards to an A2C-based policy learner (for Atari
games) and a PPO-based policy learner (for Mujoco domains) with a baseline
agent that uses the same policy learners but with only extrinsic rewards. Our
results show improved performance on most but not all of the domains.
|
An Adaptive Clipping Approach for Proximal Policy Optimization | Very recently proximal policy optimization (PPO) algorithms have been
proposed as first-order optimization methods for effective reinforcement
learning. While PPO is inspired by the same learning theory that justifies
trust region policy optimization (TRPO), PPO substantially simplifies algorithm
design and improves data efficiency by performing multiple epochs of
\emph{clipped policy optimization} from sampled data. Although clipping in PPO
stands for an important new mechanism for efficient and reliable policy update,
it may fail to adaptively improve learning performance in accordance with the
importance of each sampled state. To address this issue, a new surrogate
learning objective featuring an adaptive clipping mechanism is proposed in this
paper, enabling us to develop a new algorithm, known as PPO-$\lambda$.
PPO-$\lambda$ optimizes policies repeatedly based on a theoretical target for
adaptive policy improvement. Meanwhile, destructively large policy update can
be effectively prevented through both clipping and adaptive control of a
hyperparameter $\lambda$ in PPO-$\lambda$, ensuring high learning reliability.
PPO-$\lambda$ enjoys the same simple and efficient design as PPO. Empirically
on several Atari game playing tasks and benchmark control tasks, PPO-$\lambda$
also achieved clearly better performance than PPO.
|
Unlearn What You Have Learned: Adaptive Crowd Teaching with
Exponentially Decayed Memory Learners | With the increasing demand for large amount of labeled data, crowdsourcing
has been used in many large-scale data mining applications. However, most
existing works in crowdsourcing mainly focus on label inference and incentive
design. In this paper, we address a different problem of adaptive crowd
teaching, which is a sub-area of machine teaching in the context of
crowdsourcing. Compared with machines, human beings are extremely good at
learning a specific target concept (e.g., classifying the images into given
categories) and they can also easily transfer the learned concepts into similar
learning tasks. Therefore, a more effective way of utilizing crowdsourcing is
by supervising the crowd to label in the form of teaching. In order to perform
the teaching and expertise estimation simultaneously, we propose an adaptive
teaching framework named JEDI to construct the personalized optimal teaching
set for the crowdsourcing workers. In JEDI teaching, the teacher assumes that
each learner has an exponentially decayed memory. Furthermore, it ensures
comprehensiveness in the learning process by carefully balancing teaching
diversity and learner's accurate learning in terms of teaching usefulness.
Finally, we validate the effectiveness and efficacy of JEDI teaching in
comparison with the state-of-the-art techniques on multiple data sets with both
synthetic learners and real crowdsourcing workers.
|
Deep Multimodal Subspace Clustering Networks | We present convolutional neural network (CNN) based approaches for
unsupervised multimodal subspace clustering. The proposed framework consists of
three main stages - multimodal encoder, self-expressive layer, and multimodal
decoder. The encoder takes multimodal data as input and fuses them to a latent
space representation. The self-expressive layer is responsible for enforcing
the self-expressiveness property and acquiring an affinity matrix corresponding
to the data points. The decoder reconstructs the original input data. The
network uses the distance between the decoder's reconstruction and the original
input in its training. We investigate early, late and intermediate fusion
techniques and propose three different encoders corresponding to them for
spatial fusion. The self-expressive layers and multimodal decoders are
essentially the same for different spatial fusion-based approaches. In addition
to various spatial fusion-based methods, an affinity fusion-based network is
also proposed in which the self-expressive layer corresponding to different
modalities is enforced to be the same. Extensive experiments on three datasets
show that the proposed methods significantly outperform the state-of-the-art
multimodal subspace clustering methods.
|
Two-Player Games for Efficient Non-Convex Constrained Optimization | In recent years, constrained optimization has become increasingly relevant to
the machine learning community, with applications including Neyman-Pearson
classification, robust optimization, and fair machine learning. A natural
approach to constrained optimization is to optimize the Lagrangian, but this is
not guaranteed to work in the non-convex setting, and, if using a first-order
method, cannot cope with non-differentiable constraints (e.g. constraints on
rates or proportions).
The Lagrangian can be interpreted as a two-player game played between a
player who seeks to optimize over the model parameters, and a player who wishes
to maximize over the Lagrange multipliers. We propose a non-zero-sum variant of
the Lagrangian formulation that can cope with non-differentiable--even
discontinuous--constraints, which we call the "proxy-Lagrangian". The first
player minimizes external regret in terms of easy-to-optimize "proxy
constraints", while the second player enforces the original constraints by
minimizing swap regret.
For this new formulation, as for the Lagrangian in the non-convex setting,
the result is a stochastic classifier. For both the proxy-Lagrangian and
Lagrangian formulations, however, we prove that this classifier, instead of
having unbounded size, can be taken to be a distribution over no more than m+1
models (where m is the number of constraints). This is a significant
improvement in practical terms.
|
UCNN: Exploiting Computational Reuse in Deep Neural Networks via Weight
Repetition | Convolutional Neural Networks (CNNs) have begun to permeate all corners of
electronic society (from voice recognition to scene generation) due to their
high accuracy and machine efficiency per operation. At their core, CNN
computations are made up of multi-dimensional dot products between weight and
input vectors. This paper studies how weight repetition ---when the same weight
occurs multiple times in or across weight vectors--- can be exploited to save
energy and improve performance during CNN inference. This generalizes a popular
line of work to improve efficiency from CNN weight sparsity, as reducing
computation due to repeated zero weights is a special case of reducing
computation due to repeated weights.
To exploit weight repetition, this paper proposes a new CNN accelerator
called the Unique Weight CNN Accelerator (UCNN). UCNN uses weight repetition to
reuse CNN sub-computations (e.g., dot products) and to reduce CNN model size
when stored in off-chip DRAM ---both of which save energy. UCNN further
improves performance by exploiting sparsity in weights. We evaluate UCNN with
an accelerator-level cycle and energy model and with an RTL implementation of
the UCNN processing element. On three contemporary CNNs, UCNN improves
throughput-normalized energy consumption by 1.2x - 4x, relative to a similarly
provisioned baseline accelerator that uses Eyeriss-style sparsity
optimizations. At the same time, the UCNN processing element adds only 17-24%
area overhead relative to the same baseline.
|
Fast Weight Long Short-Term Memory | Associative memory using fast weights is a short-term memory mechanism that
substantially improves the memory capacity and time scale of recurrent neural
networks (RNNs). As recent studies introduced fast weights only to regular
RNNs, it is unknown whether fast weight memory is beneficial to gated RNNs. In
this work, we report a significant synergy between long short-term memory
(LSTM) networks and fast weight associative memories. We show that this
combination, in learning associative retrieval tasks, results in much faster
training and lower test error, a performance boost most prominent at high
memory task difficulties.
|
Online Non-Additive Path Learning under Full and Partial Information | We study the problem of online path learning with non-additive gains, which
is a central problem appearing in several applications, including ensemble
structured prediction. We present new online algorithms for path learning with
non-additive count-based gains for the three settings of full information,
semi-bandit and full bandit with very favorable regret guarantees. A key
component of our algorithms is the definition and computation of an
intermediate context-dependent automaton that enables us to use existing
algorithms designed for additive gains. We further apply our methods to the
important application of ensemble structured prediction. Finally, beyond
count-based gains, we give an efficient implementation of the EXP3 algorithm
for the full bandit setting with an arbitrary (non-additive) gain.
|
Understanding Convolutional Neural Networks with Information Theory: An
Initial Exploration | The matrix-based Renyi's \alpha-entropy functional and its multivariate
extension were recently developed in terms of the normalized eigenspectrum of a
Hermitian matrix of the projected data in a reproducing kernel Hilbert space
(RKHS). However, the utility and possible applications of these new estimators
are rather new and mostly unknown to practitioners. In this paper, we first
show that our estimators enable straightforward measurement of information flow
in realistic convolutional neural networks (CNN) without any approximation.
Then, we introduce the partial information decomposition (PID) framework and
develop three quantities to analyze the synergy and redundancy in convolutional
layer representations. Our results validate two fundamental data processing
inequalities and reveal some fundamental properties concerning the training of
CNN.
|
Deep Generative Networks For Sequence Prediction | This thesis investigates unsupervised time series representation learning for
sequence prediction problems, i.e. generating nice-looking input samples given
a previous history, for high dimensional input sequences by decoupling the
static input representation from the recurrent sequence representation. We
introduce three models based on Generative Stochastic Networks (GSN) for
unsupervised sequence learning and prediction. Experimental results for these
three models are presented on pixels of sequential handwritten digit (MNIST)
data, videos of low-resolution bouncing balls, and motion capture data. The
main contribution of this thesis is to provide evidence that GSNs are a viable
framework to learn useful representations of complex sequential input data, and
to suggest a new framework for deep generative models to learn complex
sequences by decoupling static input representations from dynamic time
dependency representations.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.