title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Reinforcement Learning of Speech Recognition System Based on Policy
Gradient and Hypothesis Selection | Speech recognition systems have achieved high recognition performance for
several tasks. However, the performance of such systems is dependent on the
tremendously costly development work of preparing vast amounts of task-matched
transcribed speech data for supervised training. The key problem here is the
cost of transcribing speech data. The cost is repeatedly required to support
new languages and new tasks. Assuming broad network services for transcribing
speech data for many users, a system would become more self-sufficient and more
useful if it possessed the ability to learn from very light feedback from the
users without annoying them. In this paper, we propose a general reinforcement
learning framework for speech recognition systems based on the policy gradient
method. As a particular instance of the framework, we also propose a hypothesis
selection-based reinforcement learning method. The proposed framework provides
a new view for several existing training and adaptation methods. The
experimental results show that the proposed method improves the recognition
performance compared to unsupervised adaptation.
|
Online Deep Learning: Learning Deep Neural Networks on the Fly | Deep Neural Networks (DNNs) are typically trained by backpropagation in a
batch learning setting, which requires the entire training data to be made
available prior to the learning task. This is not scalable for many real-world
scenarios where new data arrives sequentially in a stream form. We aim to
address an open challenge of "Online Deep Learning" (ODL) for learning DNNs on
the fly in an online setting. Unlike traditional online learning that often
optimizes some convex objective function with respect to a shallow model (e.g.,
a linear/kernel-based hypothesis), ODL is significantly more challenging since
the optimization of the DNN objective function is non-convex, and regular
backpropagation does not work well in practice, especially for online learning
settings. In this paper, we present a new online deep learning framework that
attempts to tackle the challenges by learning DNN models of adaptive depth from
a sequence of training data in an online learning setting. In particular, we
propose a novel Hedge Backpropagation (HBP) method for online updating the
parameters of DNN effectively, and validate the efficacy of our method on
large-scale data sets, including both stationary and concept drifting
scenarios.
|
Learning under $p$-Tampering Attacks | Recently, Mahloujifar and Mahmoody (TCC'17) studied attacks against learning
algorithms using a special case of Valiant's malicious noise, called
$p$-tampering, in which the adversary gets to change any training example with
independent probability $p$ but is limited to only choose malicious examples
with correct labels. They obtained $p$-tampering attacks that increase the
error probability in the so called targeted poisoning model in which the
adversary's goal is to increase the loss of the trained hypothesis over a
particular test example. At the heart of their attack was an efficient
algorithm to bias the expected value of any bounded real-output function
through $p$-tampering.
In this work, we present new biasing attacks for increasing the expected
value of bounded real-valued functions. Our improved biasing attacks, directly
imply improved $p$-tampering attacks against learners in the targeted poisoning
model. As a bonus, our attacks come with considerably simpler analysis. We also
study the possibility of PAC learning under $p$-tampering attacks in the
non-targeted (aka indiscriminate) setting where the adversary's goal is to
increase the risk of the generated hypothesis (for a random test example). We
show that PAC learning is possible under $p$-tampering poisoning attacks
essentially whenever it is possible in the realizable setting without the
attacks. We further show that PAC learning under "correct-label" adversarial
noise is not possible in general, if the adversary could choose the (still
limited to only $p$ fraction of) tampered examples that she substitutes with
adversarially chosen ones. Our formal model for such "bounded-budget" tampering
attackers is inspired by the notions of (strong) adaptive corruption in secure
multi-party computation.
|
Quantized Memory-Augmented Neural Networks | Memory-augmented neural networks (MANNs) refer to a class of neural network
models equipped with external memory (such as neural Turing machines and memory
networks). These neural networks outperform conventional recurrent neural
networks (RNNs) in terms of learning long-term dependency, allowing them to
solve intriguing AI tasks that would otherwise be hard to address. This paper
concerns the problem of quantizing MANNs. Quantization is known to be effective
when we deploy deep models on embedded systems with limited resources.
Furthermore, quantization can substantially reduce the energy consumption of
the inference procedure. These benefits justify recent developments of
quantized multi layer perceptrons, convolutional networks, and RNNs. However,
no prior work has reported the successful quantization of MANNs. The in-depth
analysis presented here reveals various challenges that do not appear in the
quantization of the other networks. Without addressing them properly, quantized
MANNs would normally suffer from excessive quantization error which leads to
degraded performance. In this paper, we identify memory addressing
(specifically, content-based addressing) as the main reason for the performance
degradation and propose a robust quantization method for MANNs to address the
challenge. In our experiments, we achieved a computation-energy gain of 22x
with 8-bit fixed-point and binary quantization compared to the floating-point
implementation. Measured on the bAbI dataset, the resulting model, named the
quantized MANN (Q-MANN), improved the error rate by 46% and 30% with 8-bit
fixed-point and binary quantization, respectively, compared to the MANN
quantized using conventional techniques.
|
Joint Sentiment/Topic Modeling on Text Data Using Boosted Restricted
Boltzmann Machine | Recently by the development of the Internet and the Web, different types of
social media such as web blogs become an immense source of text data. Through
the processing of these data, it is possible to discover practical information
about different topics, individuals opinions and a thorough understanding of
the society. Therefore, applying models which can automatically extract the
subjective information from the documents would be efficient and helpful. Topic
modeling methods, also sentiment analysis are the most raised topics in the
natural language processing and text mining fields. In this paper a new
structure for joint sentiment-topic modeling based on Restricted Boltzmann
Machine (RBM) which is a type of neural networks is proposed. By modifying the
structure of RBM as well as appending a layer which is analogous to sentiment
of text data to it, we propose a generative structure for joint sentiment topic
modeling based on neutral networks. The proposed method is supervised and
trained by the Contrastive Divergence algorithm. The new attached layer in the
proposed model is a layer with the multinomial probability distribution which
can be used in text data sentiment classification or any other supervised
application. The proposed model is compared with existing models in the
experiments such as evaluating as a generative model, sentiment classification,
information retrieval and the corresponding results demonstrate the efficiency
of the method.
|
LSTM Networks for Data-Aware Remaining Time Prediction of Business
Process Instances | Predicting the completion time of business process instances would be a very
helpful aid when managing processes under service level agreement constraints.
The ability to know in advance the trend of running process instances would
allow business managers to react in time, in order to prevent delays or
undesirable situations. However, making such accurate forecasts is not easy:
many factors may influence the required time to complete a process instance. In
this paper, we propose an approach based on deep Recurrent Neural Networks
(specifically LSTMs) that is able to exploit arbitrary information associated
to single events, in order to produce an as-accurate-as-possible prediction of
the completion time of running instances. Experiments on real-world datasets
confirm the quality of our proposal.
|
Attend and Diagnose: Clinical Time Series Analysis using Attention
Models | With widespread adoption of electronic health records, there is an increased
emphasis for predictive models that can effectively deal with clinical
time-series data. Powered by Recurrent Neural Network (RNN) architectures with
Long Short-Term Memory (LSTM) units, deep neural networks have achieved
state-of-the-art results in several clinical prediction tasks. Despite the
success of RNNs, its sequential nature prohibits parallelized computing, thus
making it inefficient particularly when processing long sequences. Recently,
architectures which are based solely on attention mechanisms have shown
remarkable success in transduction tasks in NLP, while being computationally
superior. In this paper, for the first time, we utilize attention models for
clinical time-series modeling, thereby dispensing recurrence entirely. We
develop the \textit{SAnD} (Simply Attend and Diagnose) architecture, which
employs a masked, self-attention mechanism, and uses positional encoding and
dense interpolation strategies for incorporating temporal order. Furthermore,
we develop a multi-task variant of \textit{SAnD} to jointly infer models with
multiple diagnosis tasks. Using the recent MIMIC-III benchmark datasets, we
demonstrate that the proposed approach achieves state-of-the-art performance in
all tasks, outperforming LSTM models and classical baselines with
hand-engineered features.
|
D-SLATS: Distributed Simultaneous Localization and Time Synchronization | Through the last decade, we have witnessed a surge of Internet of Things
(IoT) devices, and with that a greater need to choreograph their actions across
both time and space. Although these two problems, namely time synchronization
and localization, share many aspects in common, they are traditionally treated
separately or combined on centralized approaches that results in an ineffcient
use of resources, or in solutions that are not scalable in terms of the number
of IoT devices. Therefore, we propose D-SLATS, a framework comprised of three
different and independent algorithms to jointly solve time synchronization and
localization problems in a distributed fashion. The First two algorithms are
based mainly on the distributed Extended Kalman Filter (EKF) whereas the third
one uses optimization techniques. No fusion center is required, and the devices
only communicate with their neighbors. The proposed methods are evaluated on
custom Ultra-Wideband communication Testbed and a quadrotor, representing a
network of both static and mobile nodes. Our algorithms achieve up to three
microseconds time synchronization accuracy and 30 cm localization error.
|
Accelerated Method for Stochastic Composition Optimization with
Nonsmooth Regularization | Stochastic composition optimization draws much attention recently and has
been successful in many emerging applications of machine learning, statistical
analysis, and reinforcement learning. In this paper, we focus on the
composition problem with nonsmooth regularization penalty. Previous works
either have slow convergence rate or do not provide complete convergence
analysis for the general problem. In this paper, we tackle these two issues by
proposing a new stochastic composition optimization method for composition
problem with nonsmooth regularization penalty. In our method, we apply variance
reduction technique to accelerate the speed of convergence. To the best of our
knowledge, our method admits the fastest convergence rate for stochastic
composition optimization: for strongly convex composition problem, our
algorithm is proved to admit linear convergence; for general composition
problem, our algorithm significantly improves the state-of-the-art convergence
rate from $O(T^{-1/2})$ to $O((n_1+n_2)^{{2}/{3}}T^{-1})$. Finally, we apply
our proposed algorithm to portfolio management and policy evaluation in
reinforcement learning. Experimental results verify our theoretical analysis.
|
CARLA: An Open Urban Driving Simulator | We introduce CARLA, an open-source simulator for autonomous driving research.
CARLA has been developed from the ground up to support development, training,
and validation of autonomous urban driving systems. In addition to open-source
code and protocols, CARLA provides open digital assets (urban layouts,
buildings, vehicles) that were created for this purpose and can be used freely.
The simulation platform supports flexible specification of sensor suites and
environmental conditions. We use CARLA to study the performance of three
approaches to autonomous driving: a classic modular pipeline, an end-to-end
model trained via imitation learning, and an end-to-end model trained via
reinforcement learning. The approaches are evaluated in controlled scenarios of
increasing difficulty, and their performance is examined via metrics provided
by CARLA, illustrating the platform's utility for autonomous driving research.
The supplementary video can be viewed at https://youtu.be/Hp8Dz-Zek2E
|
Bayesian Paragraph Vectors | Word2vec (Mikolov et al., 2013) has proven to be successful in natural
language processing by capturing the semantic relationships between different
words. Built on top of single-word embeddings, paragraph vectors (Le and
Mikolov, 2014) find fixed-length representations for pieces of text with
arbitrary lengths, such as documents, paragraphs, and sentences. In this work,
we propose a novel interpretation for neural-network-based paragraph vectors by
developing an unsupervised generative model whose maximum likelihood solution
corresponds to traditional paragraph vectors. This probabilistic formulation
allows us to go beyond point estimates of parameters and to perform Bayesian
posterior inference. We find that the entropy of paragraph vectors decreases
with the length of documents, and that information about posterior uncertainty
improves performance in supervised learning tasks such as sentiment analysis
and paraphrase detection.
|
Breaking the Softmax Bottleneck: A High-Rank RNN Language Model | We formulate language modeling as a matrix factorization problem, and show
that the expressiveness of Softmax-based models (including the majority of
neural language models) is limited by a Softmax bottleneck. Given that natural
language is highly context-dependent, this further implies that in practice
Softmax with distributed word embeddings does not have enough capacity to model
natural language. We propose a simple and effective method to address this
issue, and improve the state-of-the-art perplexities on Penn Treebank and
WikiText-2 to 47.69 and 40.68 respectively. The proposed method also excels on
the large-scale 1B Word dataset, outperforming the baseline by over 5.6 points
in perplexity.
|
Applications of Deep Learning and Reinforcement Learning to Biological
Data | Rapid advances of hardware-based technologies during the past decades have
opened up new possibilities for Life scientists to gather multimodal data in
various application domains (e.g., Omics, Bioimaging, Medical Imaging, and
[Brain/Body]-Machine Interfaces), thus generating novel opportunities for
development of dedicated data intensive machine learning techniques. Overall,
recent research in Deep learning (DL), Reinforcement learning (RL), and their
combination (Deep RL) promise to revolutionize Artificial Intelligence. The
growth in computational power accompanied by faster and increased data storage
and declining computing costs have already allowed scientists in various fields
to apply these techniques on datasets that were previously intractable for
their size and complexity. This review article provides a comprehensive survey
on the application of DL, RL, and Deep RL techniques in mining Biological data.
In addition, we compare performances of DL techniques when applied to different
datasets across various application domains. Finally, we outline open issues in
this challenging research area and discuss future development perspectives.
|
WMRB: Learning to Rank in a Scalable Batch Training Approach | We propose a new learning to rank algorithm, named Weighted Margin-Rank Batch
loss (WMRB), to extend the popular Weighted Approximate-Rank Pairwise loss
(WARP). WMRB uses a new rank estimator and an efficient batch training
algorithm. The approach allows more accurate item rank approximation and
explicit utilization of parallel computation to accelerate training. In three
item recommendation tasks, WMRB consistently outperforms WARP and other
baselines. Moreover, WMRB shows clear time efficiency advantages as data scale
increases.
|
A Batch Learning Framework for Scalable Personalized Ranking | In designing personalized ranking algorithms, it is desirable to encourage a
high precision at the top of the ranked list. Existing methods either seek a
smooth convex surrogate for a non-smooth ranking metric or directly modify
updating procedures to encourage top accuracy. In this work we point out that
these methods do not scale well to a large-scale setting, and this is partly
due to the inaccurate pointwise or pairwise rank estimation. We propose a new
framework for personalized ranking. It uses batch-based rank estimators and
smooth rank-sensitive loss functions. This new batch learning framework leads
to more stable and accurate rank approximations compared to previous work.
Moreover, it enables explicit use of parallel computation to speed up training.
We conduct empirical evaluation on three item recommendation tasks. Our method
shows consistent accuracy improvements over state-of-the-art methods.
Additionally, we observe time efficiency advantages when data scale increases.
|
Deep Within-Class Covariance Analysis for Robust Audio Representation
Learning | Convolutional Neural Networks (CNNs) can learn effective features, though
have been shown to suffer from a performance drop when the distribution of the
data changes from training to test data. In this paper we analyze the internal
representations of CNNs and observe that the representations of unseen data in
each class, spread more (with higher variance) in the embedding space of the
CNN compared to representations of the training data. More importantly, this
difference is more extreme if the unseen data comes from a shifted
distribution. Based on this observation, we objectively evaluate the degree of
representation's variance in each class via eigenvalue decomposition on the
within-class covariance of the internal representations of CNNs and observe the
same behaviour. This can be problematic as larger variances might lead to
mis-classification if the sample crosses the decision boundary of its class. We
apply nearest neighbor classification on the representations and empirically
show that the embeddings with the high variance actually have significantly
worse KNN classification performances, although this could not be foreseen from
their end-to-end classification results. To tackle this problem, we propose
Deep Within-Class Covariance Analysis (DWCCA), a deep neural network layer that
significantly reduces the within-class covariance of a DNN's representation,
improving performance on unseen test data from a shifted distribution. We
empirically evaluate DWCCA on two datasets for Acoustic Scene Classification
(DCASE2016 and DCASE2017). We demonstrate that not only does DWCCA
significantly improve the network's internal representation, it also increases
the end-to-end classification accuracy, especially when the test set exhibits a
distribution shift. By adding DWCCA to a VGG network, we achieve around 6
percentage points improvement in the case of a distribution mismatch.
|
Ocasta: Clustering Configuration Settings For Error Recovery | Effective machine-aided diagnosis and repair of configuration errors
continues to elude computer systems designers. Most of the literature targets
errors that can be attributed to a single erroneous configuration setting.
However, a recent study found that a significant amount of configuration errors
require fixing more than one setting together. To address this limitation,
Ocasta statistically clusters dependent configuration settings based on the
application's accesses to its configuration settings and utilizes the extracted
clustering of configuration settings to fix configuration errors involving more
than one configuration settings. Ocasta treats applications as black-boxes and
only relies on the ability to observe application accesses to their
configuration settings.
We collected traces of real application usage from 24 Linux and 5 Windows
desktops computers and found that Ocasta is able to correctly identify clusters
with 88.6% accuracy. To demonstrate the effectiveness of Ocasta, we evaluated
it on 16 real-world configuration errors of 11 Linux and Windows applications.
Ocasta is able to successfully repair all evaluated configuration errors in 11
minutes on average and only requires the user to examine an average of 3
screenshots of the output of the application to confirm that the error is
repaired. A user study we conducted shows that Ocasta is easy to use by both
expert and non-expert users and is more efficient than manual configuration
error troubleshooting.
|
Few-Shot Learning with Graph Neural Networks | We propose to study the problem of few-shot learning with the prism of
inference on a partially observed graphical model, constructed from a
collection of input images whose label can be either observed or not. By
assimilating generic message-passing inference algorithms with their
neural-network counterparts, we define a graph neural network architecture that
generalizes several of the recently proposed few-shot learning models. Besides
providing improved numerical performance, our framework is easily extended to
variants of few-shot learning, such as semi-supervised or active learning,
demonstrating the ability of graph-based models to operate well on 'relational'
tasks.
|
Kernelized Hashcode Representations for Relation Extraction | Kernel methods have produced state-of-the-art results for a number of NLP
tasks such as relation extraction, but suffer from poor scalability due to the
high cost of computing kernel similarities between natural language structures.
A recently proposed technique, kernelized locality-sensitive hashing (KLSH),
can significantly reduce the computational cost, but is only applicable to
classifiers operating on kNN graphs. Here we propose to use random subspaces of
KLSH codes for efficiently constructing an explicit representation of NLP
structures suitable for general classification methods. Further, we propose an
approach for optimizing the KLSH model for classification problems by
maximizing an approximation of mutual information between the KLSH codes
(feature vectors) and the class labels. We evaluate the proposed approach on
biomedical relation extraction datasets, and observe significant and robust
improvements in accuracy w.r.t. state-of-the-art classifiers, along with
drastic (orders-of-magnitude) speedup compared to conventional kernel methods.
|
Enhancing Network Embedding with Auxiliary Information: An Explicit
Matrix Factorization Perspective | Recent advances in the field of network embedding have shown the
low-dimensional network representation is playing a critical role in network
analysis. However, most of the existing principles of network embedding do not
incorporate auxiliary information such as content and labels of nodes flexibly.
In this paper, we take a matrix factorization perspective of network embedding,
and incorporate structure, content and label information of the network
simultaneously. For structure, we validate that the matrix we construct
preserves high-order proximities of the network. Label information can be
further integrated into the matrix via the process of random walk sampling to
enhance the quality of embedding in an unsupervised manner, i.e., without
leveraging downstream classifiers. In addition, we generalize the Skip-Gram
Negative Sampling model to integrate the content of the network in a matrix
factorization framework. As a consequence, network embedding can be learned in
a unified framework integrating network structure and node content as well as
label information simultaneously. We demonstrate the efficacy of the proposed
model with the tasks of semi-supervised node classification and link prediction
on a variety of real-world benchmark network datasets.
|
Adversarial Training for Disease Prediction from Electronic Health
Records with Missing Data | Electronic health records (EHRs) have contributed to the computerization of
patient records and can thus be used not only for efficient and systematic
medical services, but also for research on biomedical data science. However,
there are many missing values in EHRs when provided in matrix form, which is an
important issue in many biomedical EHR applications. In this paper, we propose
a two-stage framework that includes missing data imputation and disease
prediction to address the missing data problem in EHRs. We compared the disease
prediction performance of generative adversarial networks (GANs) and
conventional learning algorithms in combination with missing data prediction
methods. As a result, we obtained a level of accuracy of 0.9777, sensitivity of
0.9521, specificity of 0.9925, area under the receiver operating characteristic
curve (AUC-ROC) of 0.9889, and F-score of 0.9688 with a stacked autoencoder as
the missing data prediction method and an auxiliary classifier GAN (AC-GAN) as
the disease prediction method. The comparison results show that a combination
of a stacked autoencoder and an AC-GAN significantly outperforms other existing
approaches. Our results suggest that the proposed framework is more robust for
disease prediction from EHRs with missing data.
|
STWalk: Learning Trajectory Representations in Temporal Graphs | Analyzing the temporal behavior of nodes in time-varying graphs is useful for
many applications such as targeted advertising, community evolution and outlier
detection. In this paper, we present a novel approach, STWalk, for learning
trajectory representations of nodes in temporal graphs. The proposed framework
makes use of structural properties of graphs at current and previous time-steps
to learn effective node trajectory representations. STWalk performs random
walks on a graph at a given time step (called space-walk) as well as on graphs
from past time-steps (called time-walk) to capture the spatio-temporal behavior
of nodes. We propose two variants of STWalk to learn trajectory
representations. In one algorithm, we perform space-walk and time-walk as part
of a single step. In the other variant, we perform space-walk and time-walk
separately and combine the learned representations to get the final trajectory
embedding. Extensive experiments on three real-world temporal graph datasets
validate the effectiveness of the learned representations when compared to
three baseline methods. We also show the goodness of the learned trajectory
embeddings for change point detection, as well as demonstrate that arithmetic
operations on these trajectory representations yield interesting and
interpretable results.
|
A Sparse Graph-Structured Lasso Mixed Model for Genetic Association with
Confounding Correction | While linear mixed model (LMM) has shown a competitive performance in
correcting spurious associations raised by population stratification, family
structures, and cryptic relatedness, more challenges are still to be addressed
regarding the complex structure of genotypic and phenotypic data. For example,
geneticists have discovered that some clusters of phenotypes are more
co-expressed than others. Hence, a joint analysis that can utilize such
relatedness information in a heterogeneous data set is crucial for genetic
modeling.
We proposed the sparse graph-structured linear mixed model (sGLMM) that can
incorporate the relatedness information from traits in a dataset with
confounding correction. Our method is capable of uncovering the genetic
associations of a large number of phenotypes together while considering the
relatedness of these phenotypes. Through extensive simulation experiments, we
show that the proposed model outperforms other existing approaches and can
model correlation from both population structure and shared signals. Further,
we validate the effectiveness of sGLMM in the real-world genomic dataset on two
different species from plants and humans. In Arabidopsis thaliana data, sGLMM
behaves better than all other baseline models for 63.4% traits. We also discuss
the potential causal genetic variation of Human Alzheimer's disease discovered
by our model and justify some of the most important genetic loci.
|
Unsupervised Document Embedding With CNNs | We propose a new model for unsupervised document embedding. Leading existing
approaches either require complex inference or use recurrent neural networks
(RNN) that are difficult to parallelize. We take a different route and develop
a convolutional neural network (CNN) embedding model. Our CNN architecture is
fully parallelizable resulting in over 10x speedup in inference time over RNN
models. Parallelizable architecture enables to train deeper models where each
successive layer has increasingly larger receptive field and models longer
range semantic structure within the document. We additionally propose a fully
unsupervised learning algorithm to train this model based on stochastic forward
prediction. Empirical results on two public benchmarks show that our approach
produces comparable to state-of-the-art accuracy at a fraction of computational
cost.
|
CUR Decompositions, Similarity Matrices, and Subspace Clustering | A general framework for solving the subspace clustering problem using the CUR
decomposition is presented. The CUR decomposition provides a natural way to
construct similarity matrices for data that come from a union of unknown
subspaces $\mathscr{U}=\underset{i=1}{\overset{M}\bigcup}S_i$. The similarity
matrices thus constructed give the exact clustering in the noise-free case.
Additionally, this decomposition gives rise to many distinct similarity
matrices from a given set of data, which allow enough flexibility to perform
accurate clustering of noisy data. We also show that two known methods for
subspace clustering can be derived from the CUR decomposition. An algorithm
based on the theoretical construction of similarity matrices is presented, and
experiments on synthetic and real data are presented to test the method.
Additionally, an adaptation of our CUR based similarity matrices is utilized
to provide a heuristic algorithm for subspace clustering; this algorithm yields
the best overall performance to date for clustering the Hopkins155 motion
segmentation dataset.
|
Feature Selection based on the Local Lift Dependence Scale | This paper uses a classical approach to feature selection: minimization of a
cost function applied on estimated joint distributions. However, the search
space in which such minimization is performed is extended. In the original
formulation, the search space is the Boolean lattice of features sets (BLFS),
while, in the present formulation, it is a collection of Boolean lattices of
ordered pairs (features, associated value) (CBLOP), indexed by the elements of
the BLFS. In this approach, we may not only select the features that are most
related to a variable Y, but also select the values of the features that most
influence the variable or that are most prone to have a specific value of Y. A
local formulation of Shanon's mutual information is applied on a CBLOP to
select features, namely, the Local Lift Dependence Scale, an scale for
measuring variable dependence in multiple resolutions. The main contribution of
this paper is to define and apply this local measure, which permits to analyse
local properties of joint distributions that are neglected by the classical
Shanon's global measure. The proposed approach is applied to a dataset
consisting of student performances on a university entrance exam, as well as on
undergraduate courses. The approach is also applied to two datasets of the UCI
Machine Learning Repository.
|
Skyline Identification in Multi-Armed Bandits | We introduce a variant of the classical PAC multi-armed bandit problem. There
is an ordered set of $n$ arms $A[1],\dots,A[n]$, each with some stochastic
reward drawn from some unknown bounded distribution. The goal is to identify
the $skyline$ of the set $A$, consisting of all arms $A[i]$ such that $A[i]$
has larger expected reward than all lower-numbered arms $A[1],\dots,A[i-1]$. We
define a natural notion of an $\varepsilon$-approximate skyline and prove
matching upper and lower bounds for identifying an $\varepsilon$-skyline.
Specifically, we show that in order to identify an $\varepsilon$-skyline from
among $n$ arms with probability $1-\delta$, $$
\Theta\bigg(\frac{n}{\varepsilon^2} \cdot \min\bigg\{
\log\bigg(\frac{1}{\varepsilon \delta}\bigg), \log\bigg(\frac{n}{\delta}\bigg)
\bigg\} \bigg) $$ samples are necessary and sufficient. When $\varepsilon \gg
1/n$, our results improve over the naive algorithm, which draws enough samples
to approximate the expected reward of every arm; the algorithm of (Auer et al.,
AISTATS'16) for Pareto-optimal arm identification is likewise superseded. Our
results show that the sample complexity of the skyline problem lies strictly in
between that of best arm identification (Even-Dar et al., COLT'02) and that of
approximating the expected reward of every arm.
|
D-PCN: Parallel Convolutional Networks for Image Recognition via a
Discriminator | In this paper, we introduce a simple but quite effective recognition
framework dubbed D-PCN, aiming at enhancing feature extracting ability of CNN.
The framework consists of two parallel CNNs, a discriminator and an extra
classifier which takes integrated features from parallel networks and gives
final prediction. The discriminator is core which drives parallel networks to
focus on different regions and learn complementary representations. The
corresponding joint training strategy is introduced which ensures the
utilization of discriminator. We validate D-PCN with several CNN models on two
benchmark datasets: CIFAR-100 and ImageNet32x32, D-PCN enhances all models. In
particular it yields state of the art performance on CIFAR-100 compared with
related works. We also conduct visualization experiment on fine-grained
Stanford Dogs dataset and verify our motivation. Additionally, we apply D-PCN
for segmentation on PASCAL VOC 2012 and also find promotion.
|
Linking Sequences of Events with Sparse or No Common Occurrence across
Data Sets | Data of practical interest - such as personal records, transaction logs, and
medical histories - are sequential collections of events relevant to a
particular source entity. Recent studies have attempted to link sequences that
represent a common entity across data sets to allow more comprehensive
statistical analyses and to identify potential privacy failures. Yet, current
approaches remain tailored to their specific domains of application, and they
fail when co-referent sequences in different data sets contain sparse or no
common events, which occurs frequently in many cases.
To address this, we formalize the general problem of "sequence linkage" and
describe "LDA-Link," a generic solution that is applicable even when
co-referent event sequences contain no common items at all. LDA-Link is built
upon "Split-Document" model, a new mixed-membership probabilistic model for the
generation of event sequence collections. It detects the latent similarity of
sequences and thus achieves robustness particularly when co-referent sequences
share sparse or no event overlap. We apply LDA-Link in the context of social
media profile reconciliation where users make no common posts across platforms,
comparing to the state-of-the-art generic solution to sequence linkage.
|
Unified Spectral Clustering with Optimal Graph | Spectral clustering has found extensive use in many areas. Most traditional
spectral clustering algorithms work in three separate steps: similarity graph
construction; continuous labels learning; discretizing the learned labels by
k-means clustering. Such common practice has two potential flaws, which may
lead to severe information loss and performance degradation. First, predefined
similarity graph might not be optimal for subsequent clustering. It is
well-accepted that similarity graph highly affects the clustering results. To
this end, we propose to automatically learn similarity information from data
and simultaneously consider the constraint that the similarity matrix has exact
c connected components if there are c clusters. Second, the discrete solution
may deviate from the spectral solution since k-means method is well-known as
sensitive to the initialization of cluster centers. In this work, we transform
the candidate solution into a new one that better approximates the discrete
one. Finally, those three subtasks are integrated into a unified framework,
with each subtask iteratively boosted by using the results of the others
towards an overall optimal solution. It is known that the performance of a
kernel method is largely determined by the choice of kernels. To tackle this
practical problem of how to select the most suitable kernel for a particular
data set, we further extend our model to incorporate multiple kernel learning
ability. Extensive experiments demonstrate the superiority of our proposed
method as compared to existing clustering approaches.
|
Scale out for large minibatch SGD: Residual network training on
ImageNet-1K with improved accuracy and reduced time to train | For the past 5 years, the ILSVRC competition and the ImageNet dataset have
attracted a lot of interest from the Computer Vision community, allowing for
state-of-the-art accuracy to grow tremendously. This should be credited to the
use of deep artificial neural network designs. As these became more complex,
the storage, bandwidth, and compute requirements increased. This means that
with a non-distributed approach, even when using the most high-density server
available, the training process may take weeks, making it prohibitive.
Furthermore, as datasets grow, the representation learning potential of deep
networks grows as well by using more complex models. This synchronicity
triggers a sharp increase in the computational requirements and motivates us to
explore the scaling behaviour on petaflop scale supercomputers. In this paper
we will describe the challenges and novel solutions needed in order to train
ResNet-50 in this large scale environment. We demonstrate above 90\% scaling
efficiency and a training time of 28 minutes using up to 104K x86 cores. This
is supported by software tools from Intel's ecosystem. Moreover, we show that
with regular 90 - 120 epoch train runs we can achieve a top-1 accuracy as high
as 77\% for the unmodified ResNet-50 topology. We also introduce the novel
Collapsed Ensemble (CE) technique that allows us to obtain a 77.5\% top-1
accuracy, similar to that of a ResNet-152, while training a unmodified
ResNet-50 topology for the same fixed training budget. All ResNet-50 models as
well as the scripts needed to replicate them will be posted shortly.
|
On the ERM Principle with Networked Data | Networked data, in which every training example involves two objects and may
share some common objects with others, is used in many machine learning tasks
such as learning to rank and link prediction. A challenge of learning from
networked examples is that target values are not known for some pairs of
objects. In this case, neither the classical i.i.d.\ assumption nor techniques
based on complete U-statistics can be used. Most existing theoretical results
of this problem only deal with the classical empirical risk minimization (ERM)
principle that always weights every example equally, but this strategy leads to
unsatisfactory bounds. We consider general weighted ERM and show new universal
risk bounds for this problem. These new bounds naturally define an optimization
problem which leads to appropriate weights for networked examples. Though this
optimization problem is not convex in general, we devise a new fully
polynomial-time approximation scheme (FPTAS) to solve it.
|
Semi-Supervised Learning via New Deep Network Inversion | We exploit a recently derived inversion scheme for arbitrary deep neural
networks to develop a new semi-supervised learning framework that applies to a
wide range of systems and problems. The approach outperforms current
state-of-the-art methods on MNIST reaching $99.14\%$ of test set accuracy while
using $5$ labeled examples per class. Experiments with one-dimensional signals
highlight the generality of the method. Importantly, our approach is simple,
efficient, and requires no change in the deep network architecture.
|
A machine learning approach for efficient uncertainty quantification
using multiscale methods | Several multiscale methods account for sub-grid scale features using coarse
scale basis functions. For example, in the Multiscale Finite Volume method the
coarse scale basis functions are obtained by solving a set of local problems
over dual-grid cells. We introduce a data-driven approach for the estimation of
these coarse scale basis functions. Specifically, we employ a neural network
predictor fitted using a set of solution samples from which it learns to
generate subsequent basis functions at a lower computational cost than solving
the local problems. The computational advantage of this approach is realized
for uncertainty quantification tasks where a large number of realizations has
to be evaluated. We attribute the ability to learn these basis functions to the
modularity of the local problems and the redundancy of the permeability patches
between samples. The proposed method is evaluated on elliptic problems yielding
very promising results.
|
High-Order Attention Models for Visual Question Answering | The quest for algorithms that enable cognitive abilities is an important part
of machine learning. A common trait in many recently investigated
cognitive-like tasks is that they take into account different data modalities,
such as visual and textual input. In this paper we propose a novel and
generally applicable form of attention mechanism that learns high-order
correlations between various data modalities. We show that high-order
correlations effectively direct the appropriate attention to the relevant
elements in the different data modalities that are required to solve the joint
task. We demonstrate the effectiveness of our high-order attention mechanism on
the task of visual question answering (VQA), where we achieve state-of-the-art
performance on the standard VQA dataset.
|
Extremely Large Minibatch SGD: Training ResNet-50 on ImageNet in 15
Minutes | We demonstrate that training ResNet-50 on ImageNet for 90 epochs can be
achieved in 15 minutes with 1024 Tesla P100 GPUs. This was made possible by
using a large minibatch size of 32k. To maintain accuracy with this large
minibatch size, we employed several techniques such as RMSprop warm-up, batch
normalization without moving averages, and a slow-start learning rate schedule.
This paper also describes the details of the hardware and software of the
system used to achieve the above performance.
|
Medical Diagnosis From Laboratory Tests by Combining Generative and
Discriminative Learning | A primary goal of computational phenotype research is to conduct medical
diagnosis. In hospital, physicians rely on massive clinical data to make
diagnosis decisions, among which laboratory tests are one of the most important
resources. However, the longitudinal and incomplete nature of laboratory test
data casts a significant challenge on its interpretation and usage, which may
result in harmful decisions by both human physicians and automatic diagnosis
systems. In this work, we take advantage of deep generative models to deal with
the complex laboratory tests. Specifically, we propose an end-to-end
architecture that involves a deep generative variational recurrent neural
networks (VRNN) to learn robust and generalizable features, and a
discriminative neural network (NN) model to learn diagnosis decision making,
and the two models are trained jointly. Our experiments are conducted on a
dataset involving 46,252 patients, and the 50 most frequent tests are used to
predict the 50 most common diagnoses. The results show that our model, VRNN+NN,
significantly (p<0.001) outperforms other baseline models. Moreover, we
demonstrate that the representations learned by the joint training are more
informative than those learned by pure generative models. Finally, we find that
our model offers a surprisingly good imputation for missing values.
|
Data Augmentation Generative Adversarial Networks | Effective training of neural networks requires much data. In the low-data
regime, parameters are underdetermined, and learnt networks generalise poorly.
Data Augmentation alleviates this by using existing data more effectively.
However standard data augmentation produces only limited plausible alternative
data. Given there is potential to generate a much broader set of augmentations,
we design and train a generative model to do data augmentation. The model,
based on image conditional Generative Adversarial Networks, takes data from a
source domain and learns to take any data item and generalise it to generate
other within-class data items. As this generative process does not depend on
the classes themselves, it can be applied to novel unseen classes of data. We
show that a Data Augmentation Generative Adversarial Network (DAGAN) augments
standard vanilla classifiers well. We also show a DAGAN can enhance few-shot
learning systems such as Matching Networks. We demonstrate these approaches on
Omniglot, on EMNIST having learnt the DAGAN on Omniglot, and VGG-Face data. In
our experiments we can see over 13% increase in accuracy in the low-data regime
experiments in Omniglot (from 69% to 82%), EMNIST (73.9% to 76%) and VGG-Face
(4.5% to 12%); in Matching Networks for Omniglot we observe an increase of 0.5%
(from 96.9% to 97.4%) and an increase of 1.8% in EMNIST (from 59.5% to 61.3%).
|
Alpha-Divergences in Variational Dropout | We investigate the use of alternative divergences to Kullback-Leibler (KL) in
variational inference(VI), based on the Variational Dropout \cite{kingma2015}.
Stochastic gradient variational Bayes (SGVB) \cite{aevb} is a general framework
for estimating the evidence lower bound (ELBO) in Variational Bayes. In this
work, we extend the SGVB estimator with using Alpha-Divergences, which are
alternative to divergences to VI' KL objective. The Gaussian dropout can be
seen as a local reparametrization trick of the SGVB objective. We extend the
Variational Dropout to use alpha divergences for variational inference. Our
results compare $\alpha$-divergence variational dropout with standard
variational dropout with correlated and uncorrelated weight noise. We show that
the $\alpha$-divergence with $\alpha \rightarrow 1$ (or KL divergence) is still
a good measure for use in variational inference, in spite of the efficient use
of Alpha-divergences for Dropout VI \cite{Li17}. $\alpha \rightarrow 1$ can
yield the lowest training error, and optimizes a good lower bound for the
evidence lower bound (ELBO) among all values of the parameter $\alpha \in
[0,\infty)$.
|
Parameter Estimation in Finite Mixture Models by Regularized Optimal
Transport: A Unified Framework for Hard and Soft Clustering | In this short paper, we formulate parameter estimation for finite mixture
models in the context of discrete optimal transportation with convex
regularization. The proposed framework unifies hard and soft clustering methods
for general mixture models. It also generalizes the celebrated
$k$\nobreakdash-means and expectation-maximization algorithms in relation to
associated Bregman divergences when applied to exponential family mixture
models.
|
Machine vs Machine: Minimax-Optimal Defense Against Adversarial Examples | Recently, researchers have discovered that the state-of-the-art object
classifiers can be fooled easily by small perturbations in the input
unnoticeable to human eyes. It is also known that an attacker can generate
strong adversarial examples if she knows the classifier parameters. Conversely,
a defender can robustify the classifier by retraining if she has access to the
adversarial examples. We explain and formulate this adversarial example problem
as a two-player continuous zero-sum game, and demonstrate the fallacy of
evaluating a defense or an attack as a static problem. To find the best
worst-case defense against whitebox attacks, we propose a continuous minimax
optimization algorithm. We demonstrate the minimax defense with two types of
attack classes -- gradient-based and neural network-based attacks. Experiments
with the MNIST and the CIFAR-10 datasets demonstrate that the defense found by
numerical minimax optimization is indeed more robust than non-minimax defenses.
We discuss directions for improving the result toward achieving robustness
against multiple types of attack classes.
|
Sparsification of the Alignment Path Search Space in Dynamic Time
Warping | Temporal data are naturally everywhere, especially in the digital era that
sees the advent of big data and internet of things. One major challenge that
arises during temporal data analysis and mining is the comparison of time
series or sequences, which requires to determine a proper distance or
(dis)similarity measure. In this context, the Dynamic Time Warping (DTW) has
enjoyed success in many domains, due to its 'temporal elasticity', a property
particularly useful when matching temporal data. Unfortunately this
dissimilarity measure suffers from a quadratic computational cost, which
prohibits its use for large scale applications. This work addresses the
sparsification of the alignment path search space for DTW-like measures,
essentially to lower their computational cost without loosing on the quality of
the measure. As a result of our sparsification approach, two new
(dis)similarity measures, namely SP-DTW (Sparsified-Paths search space DTW) and
its kernelization SP-K rdtw (Sparsified-Paths search space K rdtw kernel) are
proposed for time series comparison. A wide range of public datasets is used to
evaluate the efficiency (estimated in term of speed-up ratio and classification
accuracy) of the proposed (dis)similarity measures on the 1-Nearest Neighbor
(1-NN) and the Support Vector Machine (SVM) classification algorithms. Our
experiment shows that our proposed measures provide a significant speed-up
without loosing on accuracy. Furthermore, at the cost of a slight reduction of
the speedup they significantly outperform on the accuracy criteria the old but
well known Sakoe-Chiba approach that reduces the DTW path search space using a
symmetric corridor.
|
A Parallel Best-Response Algorithm with Exact Line Search for Nonconvex
Sparsity-Regularized Rank Minimization | In this paper, we propose a convergent parallel best-response algorithm with
the exact line search for the nondifferentiable nonconvex sparsity-regularized
rank minimization problem. On the one hand, it exhibits a faster convergence
than subgradient algorithms and block coordinate descent algorithms. On the
other hand, its convergence to a stationary point is guaranteed, while ADMM
algorithms only converge for convex problems. Furthermore, the exact line
search procedure in the proposed algorithm is performed efficiently in
closed-form to avoid the meticulous choice of stepsizes, which is however a
common bottleneck in subgradient algorithms and successive convex approximation
algorithms. Finally, the proposed algorithm is numerically tested.
|
A Supervised Learning Concept for Reducing User Interaction in Passenger
Cars | In this article an automation system for human-machine-interfaces (HMI) for
setpoint adjustment using supervised learning is presented. We use HMIs of
multi-modal thermal conditioning systems in passenger cars as example for a
complex setpoint selection system. The goal is the reduction of interaction
complexity up to full automation. The approach is not limited to climate
control applications but can be extended to other setpoint-based HMIs.
|
Simple And Efficient Architecture Search for Convolutional Neural
Networks | Neural networks have recently had a lot of success for many tasks. However,
neural network architectures that perform well are still typically designed
manually by experts in a cumbersome trial-and-error process. We propose a new
method to automatically search for well-performing CNN architectures based on a
simple hill climbing procedure whose operators apply network morphisms,
followed by short optimization runs by cosine annealing. Surprisingly, this
simple method yields competitive results, despite only requiring resources in
the same order of magnitude as training a single network. E.g., on CIFAR-10,
our method designs and trains networks with an error rate below 6% in only 12
hours on a single GPU; training for one day reduces this error further, to
almost 5%.
|
Not all bytes are equal: Neural byte sieve for fuzzing | Fuzzing is a popular dynamic program analysis technique used to find
vulnerabilities in complex software. Fuzzing involves presenting a target
program with crafted malicious input designed to cause crashes, buffer
overflows, memory errors, and exceptions. Crafting malicious inputs in an
efficient manner is a difficult open problem and often the best approach to
generating such inputs is through applying uniform random mutations to
pre-existing valid inputs (seed files). We present a learning technique that
uses neural networks to learn patterns in the input files from past fuzzing
explorations to guide future fuzzing explorations. In particular, the neural
models learn a function to predict good (and bad) locations in input files to
perform fuzzing mutations based on the past mutations and corresponding code
coverage information. We implement several neural models including LSTMs and
sequence-to-sequence models that can encode variable length input files. We
incorporate our models in the state-of-the-art AFL (American Fuzzy Lop) fuzzer
and show significant improvements in terms of code coverage, unique code paths,
and crashes for various input formats including ELF, PNG, PDF, and XML.
|
Provably efficient neural network representation for image
classification | The state-of-the-art approaches for image classification are based on neural
networks. Mathematically, the task of classifying images is equivalent to
finding the function that maps an image to the label it is associated with. To
rigorously establish the success of neural network methods, we should first
prove that the function has an efficient neural network representation, and
then design provably efficient training algorithms to find such a
representation. Here, we achieve the first goal based on a set of assumptions
about the patterns in the images. The validity of these assumptions is very
intuitive in many image classification problems, including but not limited to,
recognizing handwritten digits.
|
Three Factors Influencing Minima in SGD | We investigate the dynamical and convergent properties of stochastic gradient
descent (SGD) applied to Deep Neural Networks (DNNs). Characterizing the
relation between learning rate, batch size and the properties of the final
minima, such as width or generalization, remains an open question. In order to
tackle this problem we investigate the previously proposed approximation of SGD
by a stochastic differential equation (SDE). We theoretically argue that three
factors - learning rate, batch size and gradient covariance - influence the
minima found by SGD. In particular we find that the ratio of learning rate to
batch size is a key determinant of SGD dynamics and of the width of the final
minima, and that higher values of the ratio lead to wider minima and often
better generalization. We confirm these findings experimentally. Further, we
include experiments which show that learning rate schedules can be replaced
with batch size schedules and that the ratio of learning rate to batch size is
an important factor influencing the memorization process.
|
Attention-based Information Fusion using Multi-Encoder-Decoder Recurrent
Neural Networks | With the rising number of interconnected devices and sensors, modeling
distributed sensor networks is of increasing interest. Recurrent neural
networks (RNN) are considered particularly well suited for modeling sensory and
streaming data. When predicting future behavior, incorporating information from
neighboring sensor stations is often beneficial. We propose a new RNN based
architecture for context specific information fusion across multiple spatially
distributed sensor stations. Hereby, latent representations of multiple local
models, each modeling one sensor station, are jointed and weighted, according
to their importance for the prediction. The particular importance is assessed
depending on the current context using a separate attention function. We
demonstrate the effectiveness of our model on three different real-world sensor
network datasets.
|
Tensor Decompositions for Modeling Inverse Dynamics | Modeling inverse dynamics is crucial for accurate feedforward robot control.
The model computes the necessary joint torques, to perform a desired movement.
The highly non-linear inverse function of the dynamical system can be
approximated using regression techniques. We propose as regression method a
tensor decomposition model that exploits the inherent three-way interaction of
positions x velocities x accelerations. Most work in tensor factorization has
addressed the decomposition of dense tensors. In this paper, we build upon the
decomposition of sparse tensors, with only small amounts of nonzero entries.
The decomposition of sparse tensors has successfully been used in relational
learning, e.g., the modeling of large knowledge graphs. Recently, the approach
has been extended to multi-class classification with discrete input variables.
Representing the data in high dimensional sparse tensors enables the
approximation of complex highly non-linear functions. In this paper we show how
the decomposition of sparse tensors can be applied to regression problems.
Furthermore, we extend the method to continuous inputs, by learning a mapping
from the continuous inputs to the latent representations of the tensor
decomposition, using basis functions. We evaluate our proposed model on a
dataset with trajectories from a seven degrees of freedom SARCOS robot arm. Our
experimental results show superior performance of the proposed functional
tensor model, compared to challenging state-of-the art methods.
|
Weightless: Lossy Weight Encoding For Deep Neural Network Compression | The large memory requirements of deep neural networks limit their deployment
and adoption on many devices. Model compression methods effectively reduce the
memory requirements of these models, usually through applying transformations
such as weight pruning or quantization. In this paper, we present a novel
scheme for lossy weight encoding which complements conventional compression
techniques. The encoding is based on the Bloomier filter, a probabilistic data
structure that can save space at the cost of introducing random errors.
Leveraging the ability of neural networks to tolerate these imperfections and
by re-training around the errors, the proposed technique, Weightless, can
compress DNN weights by up to 496x with the same model accuracy. This results
in up to a 1.51x improvement over the state-of-the-art.
|
Person Recognition using Smartphones' Accelerometer Data | Smartphones have become quite pervasive in various aspects of our daily
lives. They have become important links to a host of important data and
applications, which if compromised, can lead to disastrous results. Due to
this, today's smartphones are equipped with multiple layers of authentication
modules. However, there still lies the need for a viable and unobtrusive layer
of security which can perform the task of user authentication using resources
which are cost-efficient and widely available on smartphones. In this work, we
propose a method to recognize users using data from a phone's embedded
accelerometer sensors. Features encapsulating information from both time and
frequency domains are extracted from walking data samples, and are used to
build a Random Forest ensemble classification model. Based on the experimental
results, the resultant model delivers an accuracy of 0.9679 and Area under
Curve (AUC) of 0.9822.
|
Machine Learning for the Geosciences: Challenges and Opportunities | Geosciences is a field of great societal relevance that requires solutions to
several urgent problems facing our humanity and the planet. As geosciences
enters the era of big data, machine learning (ML) -- that has been widely
successful in commercial domains -- offers immense potential to contribute to
problems in geosciences. However, problems in geosciences have several unique
challenges that are seldom found in traditional applications, requiring novel
problem formulations and methodologies in machine learning. This article
introduces researchers in the machine learning (ML) community to these
challenges offered by geoscience problems and the opportunities that exist for
advancing both machine learning and geosciences. We first highlight typical
sources of geoscience data and describe their properties that make it
challenging to use traditional machine learning techniques. We then describe
some of the common categories of geoscience problems where machine learning can
play a role, and discuss some of the existing efforts and promising directions
for methodological development in machine learning. We conclude by discussing
some of the emerging research themes in machine learning that are applicable
across all problems in the geosciences, and the importance of a deep
collaboration between machine learning and geosciences for synergistic
advancements in both disciplines.
|
Spatio-Temporal Data Mining: A Survey of Problems and Methods | Large volumes of spatio-temporal data are increasingly collected and studied
in diverse domains including, climate science, social sciences, neuroscience,
epidemiology, transportation, mobile health, and Earth sciences.
Spatio-temporal data differs from relational data for which computational
approaches are developed in the data mining community for multiple decades, in
that both spatial and temporal attributes are available in addition to the
actual measurements/attributes. The presence of these attributes introduces
additional challenges that needs to be dealt with. Approaches for mining
spatio-temporal data have been studied for over a decade in the data mining
community. In this article we present a broad survey of this relatively young
field of spatio-temporal data mining. We discuss different types of
spatio-temporal data and the relevant data mining questions that arise in the
context of analyzing each of these datasets. Based on the nature of the data
mining problem studied, we classify literature on spatio-temporal data mining
into six major categories: clustering, predictive learning, change detection,
frequent pattern mining, anomaly detection, and relationship mining. We discuss
the various forms of spatio-temporal data mining problems in each of these
categories.
|
ADaPTION: Toolbox and Benchmark for Training Convolutional Neural
Networks with Reduced Numerical Precision Weights and Activation | Deep Neural Networks (DNNs) and Convolutional Neural Networks (CNNs) are
useful for many practical tasks in machine learning. Synaptic weights, as well
as neuron activation functions within the deep network are typically stored
with high-precision formats, e.g. 32 bit floating point. However, since storage
capacity is limited and each memory access consumes power, both storage
capacity and memory access are two crucial factors in these networks. Here we
present a method and present the ADaPTION toolbox to extend the popular deep
learning library Caffe to support training of deep CNNs with reduced numerical
precision of weights and activations using fixed point notation. ADaPTION
includes tools to measure the dynamic range of weights and activations. Using
the ADaPTION tools, we quantized several CNNs including VGG16 down to 16-bit
weights and activations with only 0.8% drop in Top-1 accuracy. The
quantization, especially of the activations, leads to increase of up to 50% of
sparsity especially in early and intermediate layers, which we exploit to skip
multiplications with zero, thus performing faster and computationally cheaper
inference.
|
Resurrecting the sigmoid in deep learning through dynamical isometry:
theory and practice | It is well known that the initialization of weights in deep neural networks
can have a dramatic impact on learning speed. For example, ensuring the mean
squared singular value of a network's input-output Jacobian is $O(1)$ is
essential for avoiding the exponential vanishing or explosion of gradients. The
stronger condition that all singular values of the Jacobian concentrate near
$1$ is a property known as dynamical isometry. For deep linear networks,
dynamical isometry can be achieved through orthogonal weight initialization and
has been shown to dramatically speed up learning; however, it has remained
unclear how to extend these results to the nonlinear setting. We address this
question by employing powerful tools from free probability theory to compute
analytically the entire singular value distribution of a deep network's
input-output Jacobian. We explore the dependence of the singular value
distribution on the depth of the network, the weight initialization, and the
choice of nonlinearity. Intriguingly, we find that ReLU networks are incapable
of dynamical isometry. On the other hand, sigmoidal networks can achieve
isometry, but only with orthogonal weight initialization. Moreover, we
demonstrate empirically that deep nonlinear networks achieving dynamical
isometry learn orders of magnitude faster than networks that do not. Indeed, we
show that properly-initialized deep sigmoidal networks consistently outperform
deep ReLU networks. Overall, our analysis reveals that controlling the entire
distribution of Jacobian singular values is an important design consideration
in deep learning.
|
ACtuAL: Actor-Critic Under Adversarial Learning | Generative Adversarial Networks (GANs) are a powerful framework for deep
generative modeling. Posed as a two-player minimax problem, GANs are typically
trained end-to-end on real-valued data and can be used to train a generator of
high-dimensional and realistic images. However, a major limitation of GANs is
that training relies on passing gradients from the discriminator through the
generator via back-propagation. This makes it fundamentally difficult to train
GANs with discrete data, as generation in this case typically involves a
non-differentiable function. These difficulties extend to the reinforcement
learning setting when the action space is composed of discrete decisions. We
address these issues by reframing the GAN framework so that the generator is no
longer trained using gradients through the discriminator, but is instead
trained using a learned critic in the actor-critic framework with a Temporal
Difference (TD) objective. This is a natural fit for sequence modeling and we
use it to achieve improvements on language modeling tasks over the standard
Teacher-Forcing methods.
|
"Found in Translation": Predicting Outcomes of Complex Organic Chemistry
Reactions using Neural Sequence-to-Sequence Models | There is an intuitive analogy of an organic chemist's understanding of a
compound and a language speaker's understanding of a word. Consequently, it is
possible to introduce the basic concepts and analyze potential impacts of
linguistic analysis to the world of organic chemistry. In this work, we cast
the reaction prediction task as a translation problem by introducing a
template-free sequence-to-sequence model, trained end-to-end and fully
data-driven. We propose a novel way of tokenization, which is arbitrarily
extensible with reaction information. With this approach, we demonstrate
results superior to the state-of-the-art solution by a significant margin on
the top-1 accuracy. Specifically, our approach achieves an accuracy of 80.1%
without relying on auxiliary knowledge such as reaction templates. Also, 66.4%
accuracy is reached on a larger and noisier dataset.
|
Improving Factor-Based Quantitative Investing by Forecasting Company
Fundamentals | On a periodic basis, publicly traded companies are required to report
fundamentals: financial data such as revenue, operating income, debt, among
others. These data points provide some insight into the financial health of a
company. Academic research has identified some factors, i.e. computed features
of the reported data, that are known through retrospective analysis to
outperform the market average. Two popular factors are the book value
normalized by market capitalization (book-to-market) and the operating income
normalized by the enterprise value (EBIT/EV). In this paper: we first show
through simulation that if we could (clairvoyantly) select stocks using factors
calculated on future fundamentals (via oracle), then our portfolios would far
outperform a standard factor approach. Motivated by this analysis, we train
deep neural networks to forecast future fundamentals based on a trailing
5-years window. Quantitative analysis demonstrates a significant improvement in
MSE over a naive strategy. Moreover, in retrospective analysis using an
industry-grade stock portfolio simulator (backtester), we show an improvement
in compounded annual return to 17.1% (MLP) vs 14.4% for a standard factor
model.
|
Invariances and Data Augmentation for Supervised Music Transcription | This paper explores a variety of models for frame-based music transcription,
with an emphasis on the methods needed to reach state-of-the-art on human
recordings. The translation-invariant network discussed in this paper, which
combines a traditional filterbank with a convolutional neural network, was the
top-performing model in the 2017 MIREX Multiple Fundamental Frequency
Estimation evaluation. This class of models shares parameters in the
log-frequency domain, which exploits the frequency invariance of music to
reduce the number of model parameters and avoid overfitting to the training
data. All models in this paper were trained with supervision by labeled data
from the MusicNet dataset, augmented by random label-preserving pitch-shift
transformations.
|
Learning and Visualizing Localized Geometric Features Using 3D-CNN: An
Application to Manufacturability Analysis of Drilled Holes | 3D Convolutional Neural Networks (3D-CNN) have been used for object
recognition based on the voxelized shape of an object. However, interpreting
the decision making process of these 3D-CNNs is still an infeasible task. In
this paper, we present a unique 3D-CNN based Gradient-weighted Class Activation
Mapping method (3D-GradCAM) for visual explanations of the distinct local
geometric features of interest within an object. To enable efficient learning
of 3D geometries, we augment the voxel data with surface normals of the object
boundary. We then train a 3D-CNN with this augmented data and identify the
local features critical for decision-making using 3D GradCAM. An application of
this feature identification framework is to recognize difficult-to-manufacture
drilled hole features in a complex CAD geometry. The framework can be extended
to identify difficult-to-manufacture features at multiple spatial scales
leading to a real-time design for manufacturability decision support system.
|
Modeling Human Categorization of Natural Images Using Deep Feature
Representations | Over the last few decades, psychologists have developed sophisticated formal
models of human categorization using simple artificial stimuli. In this paper,
we use modern machine learning methods to extend this work into the realm of
naturalistic stimuli, enabling human categorization to be studied over the
complex visual domain in which it evolved and developed. We show that
representations derived from a convolutional neural network can be used to
model behavior over a database of >300,000 human natural image classifications,
and find that a group of models based on these representations perform well,
near the reliability of human judgments. Interestingly, this group includes
both exemplar and prototype models, contrasting with the dominance of exemplar
models in previous work. We are able to improve the performance of the
remaining models by preprocessing neural network representations to more
closely capture human similarity judgments.
|
STARK: Structured Dictionary Learning Through Rank-one Tensor Recovery | In recent years, a class of dictionaries have been proposed for
multidimensional (tensor) data representation that exploit the structure of
tensor data by imposing a Kronecker structure on the dictionary underlying the
data. In this work, a novel algorithm called "STARK" is provided to learn
Kronecker structured dictionaries that can represent tensors of any order. By
establishing that the Kronecker product of any number of matrices can be
rearranged to form a rank-1 tensor, we show that Kronecker structure can be
enforced on the dictionary by solving a rank-1 tensor recovery problem. Because
rank-1 tensor recovery is a challenging nonconvex problem, we resort to solving
a convex relaxation of this problem. Empirical experiments on synthetic and
real data show promising results for our proposed algorithm.
|
Sobolev GAN | We propose a new Integral Probability Metric (IPM) between distributions: the
Sobolev IPM. The Sobolev IPM compares the mean discrepancy of two distributions
for functions (critic) restricted to a Sobolev ball defined with respect to a
dominant measure $\mu$. We show that the Sobolev IPM compares two distributions
in high dimensions based on weighted conditional Cumulative Distribution
Functions (CDF) of each coordinate on a leave one out basis. The Dominant
measure $\mu$ plays a crucial role as it defines the support on which
conditional CDFs are compared. Sobolev IPM can be seen as an extension of the
one dimensional Von-Mises Cram\'er statistics to high dimensional
distributions. We show how Sobolev IPM can be used to train Generative
Adversarial Networks (GANs). We then exploit the intrinsic conditioning implied
by Sobolev IPM in text generation. Finally we show that a variant of Sobolev
GAN achieves competitive results in semi-supervised learning on CIFAR-10,
thanks to the smoothness enforced on the critic by Sobolev GAN which relates to
Laplacian regularization.
|
Robust Multilingual Part-of-Speech Tagging via Adversarial Training | Adversarial training (AT) is a powerful regularization method for neural
networks, aiming to achieve robustness to input perturbations. Yet, the
specific effects of the robustness obtained from AT are still unclear in the
context of natural language processing. In this paper, we propose and analyze a
neural POS tagging model that exploits AT. In our experiments on the Penn
Treebank WSJ corpus and the Universal Dependencies (UD) dataset (27 languages),
we find that AT not only improves the overall tagging accuracy, but also 1)
prevents over-fitting well in low resource languages and 2) boosts tagging
accuracy for rare / unseen words. We also demonstrate that 3) the improved
tagging performance by AT contributes to the downstream task of dependency
parsing, and that 4) AT helps the model to learn cleaner word representations.
5) The proposed AT model is generally effective in different sequence labeling
tasks. These positive results motivate further use of AT for natural language
tasks.
|
pyLEMMINGS: Large Margin Multiple Instance Classification and Ranking
for Bioinformatics Applications | Motivation: A major challenge in the development of machine learning based
methods in computational biology is that data may not be accurately labeled due
to the time and resources required for experimentally annotating properties of
proteins and DNA sequences. Standard supervised learning algorithms assume
accurate instance-level labeling of training data. Multiple instance learning
is a paradigm for handling such labeling ambiguities. However, the widely used
large-margin classification methods for multiple instance learning are
heuristic in nature with high computational requirements. In this paper, we
present stochastic sub-gradient optimization large margin algorithms for
multiple instance classification and ranking, and provide them in a software
suite called pyLEMMINGS.
Results: We have tested pyLEMMINGS on a number of bioinformatics problems as
well as benchmark datasets. pyLEMMINGS has successfully been able to identify
functionally important segments of proteins: binding sites in Calmodulin
binding proteins, prion forming regions, and amyloid cores. pyLEMMINGS achieves
state-of-the-art performance in all these tasks, demonstrating the value of
multiple instance learning. Furthermore, our method has shown more than
100-fold improvement in terms of running time as compared to heuristic
solutions with improved accuracy over benchmark datasets.
Availability and Implementation: pyLEMMINGS python package is available for
download at: http://faculty.pieas.edu.pk/fayyaz/software.html#pylemmings.
|
Adversarial Symmetric Variational Autoencoder | A new form of variational autoencoder (VAE) is developed, in which the joint
distribution of data and codes is considered in two (symmetric) forms: ($i$)
from observed data fed through the encoder to yield codes, and ($ii$) from
latent codes drawn from a simple prior and propagated through the decoder to
manifest data. Lower bounds are learned for marginal log-likelihood fits
observed data and latent codes. When learning with the variational bound, one
seeks to minimize the symmetric Kullback-Leibler divergence of joint density
functions from ($i$) and ($ii$), while simultaneously seeking to maximize the
two marginal log-likelihoods. To facilitate learning, a new form of adversarial
training is developed. An extensive set of experiments is performed, in which
we demonstrate state-of-the-art data reconstruction and generation on several
image benchmark datasets.
|
Scalable Peaceman-Rachford Splitting Method with Proximal Terms | Along with developing of Peaceman-Rachford Splittling Method (PRSM), many
batch algorithms based on it have been studied very deeply. But almost no
algorithm focused on the performance of stochastic version of PRSM. In this
paper, we propose a new stochastic algorithm based on PRSM, prove its
convergence rate in ergodic sense, and test its performance on both artificial
and real data. We show that our proposed algorithm, Stochastic Scalable PRSM
(SS-PRSM), enjoys the $O(1/K)$ convergence rate, which is the same as those
newest stochastic algorithms that based on ADMM but faster than general
Stochastic ADMM (which is $O(1/\sqrt{K})$). Our algorithm also owns wide
flexibility, outperforms many state-of-the-art stochastic algorithms coming
from ADMM, and has low memory cost in large-scale splitting optimization
problems.
|
Near-optimal sample complexity for convex tensor completion | We analyze low rank tensor completion (TC) using noisy measurements of a
subset of the tensor. Assuming a rank-$r$, order-$d$, $N \times N \times \cdots
\times N$ tensor where $r=O(1)$, the best sampling complexity that was achieved
is $O(N^{\frac{d}{2}})$, which is obtained by solving a tensor nuclear-norm
minimization problem. However, this bound is significantly larger than the
number of free variables in a low rank tensor which is $O(dN)$. In this paper,
we show that by using an atomic-norm whose atoms are rank-$1$ sign tensors, one
can obtain a sample complexity of $O(dN)$. Moreover, we generalize the matrix
max-norm definition to tensors, which results in a max-quasi-norm (max-qnorm)
whose unit ball has small Rademacher complexity. We prove that solving a
constrained least squares estimation using either the convex atomic-norm or the
nonconvex max-qnorm results in optimal sample complexity for the problem of
low-rank tensor completion. Furthermore, we show that these bounds are nearly
minimax rate-optimal. We also provide promising numerical results for max-qnorm
constrained tensor completion, showing improved recovery results compared to
matricization and alternating least squares.
|
Straggler Mitigation in Distributed Optimization Through Data Encoding | Slow running or straggler tasks can significantly reduce computation speed in
distributed computation. Recently, coding-theory-inspired approaches have been
applied to mitigate the effect of straggling, through embedding redundancy in
certain linear computational steps of the optimization algorithm, thus
completing the computation without waiting for the stragglers. In this paper,
we propose an alternate approach where we embed the redundancy directly in the
data itself, and allow the computation to proceed completely oblivious to
encoding. We propose several encoding schemes, and demonstrate that popular
batch algorithms, such as gradient descent and L-BFGS, applied in a
coding-oblivious manner, deterministically achieve sample path linear
convergence to an approximate solution of the original problem, using an
arbitrarily varying subset of the nodes at each iteration. Moreover, this
approximation can be controlled by the amount of redundancy and the number of
nodes used in each iteration. We provide experimental results demonstrating the
advantage of the approach over uncoded and data replication strategies.
|
Feature importance scores and lossless feature pruning using Banzhaf
power indices | Understanding the influence of features in machine learning is crucial to
interpreting models and selecting the best features for classification. In this
work we propose the use of principles from coalitional game theory to reason
about importance of features. In particular, we propose the use of the Banzhaf
power index as a measure of influence of features on the outcome of a
classifier. We show that features having Banzhaf power index of zero can be
losslessly pruned without damage to classifier accuracy. Computing the power
indices does not require having access to data samples. However, if samples are
available, the indices can be empirically estimated. We compute Banzhaf power
indices for a neural network classifier on real-life data, and compare the
results with gradient-based feature saliency, and coefficients of a logistic
regression model with $L_1$ regularization.
|
Multiple-Source Adaptation for Regression Problems | We present a detailed theoretical analysis of the problem of multiple-source
adaptation in the general stochastic scenario, extending known results that
assume a single target labeling function. Our results cover a more realistic
scenario and show the existence of a single robust predictor accurate for
\emph{any} target mixture of the source distributions. Moreover, we present an
efficient and practical optimization solution to determine the robust predictor
in the important case of squared loss, by casting the problem as an instance of
DC-programming. We report the results of experiments with both an artificial
task and a sentiment analysis task. We find that our algorithm outperforms
competing approaches by producing a single robust model that performs well on
any target mixture distribution.
|
Dynamic Principal Projection for Cost-Sensitive Online Multi-Label
Classification | We study multi-label classification (MLC) with three important real-world
issues: online updating, label space dimensional reduction (LSDR), and
cost-sensitivity. Current MLC algorithms have not been designed to address
these three issues simultaneously. In this paper, we propose a novel algorithm,
cost-sensitive dynamic principal projection (CS-DPP) that resolves all three
issues. The foundation of CS-DPP is an online LSDR framework derived from a
leading LSDR algorithm. In particular, CS-DPP is equipped with an efficient
online dimension reducer motivated by matrix stochastic gradient, and
establishes its theoretical backbone when coupled with a carefully-designed
online regression learner. In addition, CS-DPP embeds the cost information into
label weights to achieve cost-sensitivity along with theoretical guarantees.
Experimental results verify that CS-DPP achieves better practical performance
than current MLC algorithms across different evaluation criteria, and
demonstrate the importance of resolving the three issues simultaneously.
|
Robust Matrix Elastic Net based Canonical Correlation Analysis: An
Effective Algorithm for Multi-View Unsupervised Learning | This paper presents a robust matrix elastic net based canonical correlation
analysis (RMEN-CCA) for multiple view unsupervised learning problems, which
emphasizes the combination of CCA and the robust matrix elastic net (RMEN) used
as coupled feature selection. The RMEN-CCA leverages the strength of the RMEN
to distill naturally meaningful features without any prior assumption and to
measure effectively correlations between different 'views'. We can further
employ directly the kernel trick to extend the RMEN-CCA to the kernel scenario
with theoretical guarantees, which takes advantage of the kernel trick for
highly complicated nonlinear feature learning. Rather than simply incorporating
existing regularization minimization terms into CCA, this paper provides a new
learning paradigm for CCA and is the first to derive a coupled feature
selection based CCA algorithm that guarantees convergence. More significantly,
for CCA, the newly-derived RMEN-CCA bridges the gap between measurement of
relevance and coupled feature selection. Moreover, it is nontrivial to tackle
directly the RMEN-CCA by previous optimization approaches derived from its
sophisticated model architecture. Therefore, this paper further offers a bridge
between a new optimization problem and an existing efficient iterative
approach. As a consequence, the RMEN-CCA can overcome the limitation of CCA and
address large-scale and streaming data problems. Experimental results on four
popular competing datasets illustrate that the RMEN-CCA performs more
effectively and efficiently than do state-of-the-art approaches.
|
TripletGAN: Training Generative Model with Triplet Loss | As an effective way of metric learning, triplet loss has been widely used in
many deep learning tasks, including face recognition and person-ReID, leading
to many states of the arts. The main innovation of triplet loss is using
feature map to replace softmax in the classification task. Inspired by this
concept, we propose here a new adversarial modeling method by substituting the
classification loss of discriminator with triplet loss. Theoretical proof based
on IPM (Integral probability metric) demonstrates that such setting will help
the generator converge to the given distribution theoretically under some
conditions. Moreover, since triplet loss requires the generator to maximize
distance within a class, we justify tripletGAN is also helpful to prevent mode
collapse through both theory and experiment.
|
Overcoming data scarcity with transfer learning | Despite increasing focus on data publication and discovery in materials
science and related fields, the global view of materials data is highly sparse.
This sparsity encourages training models on the union of multiple datasets, but
simple unions can prove problematic as (ostensibly) equivalent properties may
be measured or computed differently depending on the data source. These hidden
contextual differences introduce irreducible errors into analyses,
fundamentally limiting their accuracy. Transfer learning, where information
from one dataset is used to inform a model on another, can be an effective tool
for bridging sparse data while preserving the contextual differences in the
underlying measurements. Here, we describe and compare three techniques for
transfer learning: multi-task, difference, and explicit latent variable
architectures. We show that difference architectures are most accurate in the
multi-fidelity case of mixed DFT and experimental band gaps, while multi-task
most improves classification performance of color with band gaps. For
activation energies of steps in NO reduction, the explicit latent variable
method is not only the most accurate, but also enjoys cancellation of errors in
functions that depend on multiple tasks. These results motivate the publication
of high quality materials datasets that encode transferable information,
independent of industrial or academic interest in the particular labels, and
encourage further development and application of transfer learning methods to
materials informatics problems.
|
Decoupled Weight Decay Regularization | L$_2$ regularization and weight decay regularization are equivalent for
standard stochastic gradient descent (when rescaled by the learning rate), but
as we demonstrate this is \emph{not} the case for adaptive gradient algorithms,
such as Adam. While common implementations of these algorithms employ L$_2$
regularization (often calling it "weight decay" in what may be misleading due
to the inequivalence we expose), we propose a simple modification to recover
the original formulation of weight decay regularization by \emph{decoupling}
the weight decay from the optimization steps taken w.r.t. the loss function. We
provide empirical evidence that our proposed modification (i) decouples the
optimal choice of weight decay factor from the setting of the learning rate for
both standard SGD and Adam and (ii) substantially improves Adam's
generalization performance, allowing it to compete with SGD with momentum on
image classification datasets (on which it was previously typically
outperformed by the latter). Our proposed decoupled weight decay has already
been adopted by many researchers, and the community has implemented it in
TensorFlow and PyTorch; the complete source code for our experiments is
available at https://github.com/loshchil/AdamW-and-SGDW
|
The Multi-layer Information Bottleneck Problem | The muti-layer information bottleneck (IB) problem, where information is
propagated (or successively refined) from layer to layer, is considered. Based
on information forwarded by the preceding layer, each stage of the network is
required to preserve a certain level of relevance with regards to a specific
hidden variable, quantified by the mutual information. The hidden variables and
the source can be arbitrarily correlated. The optimal trade-off between rates
of relevance and compression (or complexity) is obtained through a
single-letter characterization, referred to as the rate-relevance region.
Conditions of successive refinabilty are given. Binary source with BSC hidden
variables and binary source with BSC/BEC mixed hidden variables are both proved
to be successively refinable. We further extend our result to Guassian models.
A counterexample of successive refinability is also provided.
|
Deep Rewiring: Training very sparse deep networks | Neuromorphic hardware tends to pose limits on the connectivity of deep
networks that one can run on them. But also generic hardware and software
implementations of deep learning run more efficiently for sparse networks.
Several methods exist for pruning connections of a neural network after it was
trained without connectivity constraints. We present an algorithm, DEEP R, that
enables us to train directly a sparsely connected neural network. DEEP R
automatically rewires the network during supervised training so that
connections are there where they are most needed for the task, while its total
number is all the time strictly bounded. We demonstrate that DEEP R can be used
to train very sparse feedforward and recurrent neural networks on standard
benchmark tasks with just a minor loss in performance. DEEP R is based on a
rigorous theoretical foundation that views rewiring as stochastic sampling of
network configurations from a posterior.
|
Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup
Fairness | The most prevalent notions of fairness in machine learning are statistical
definitions: they fix a small collection of pre-defined groups, and then ask
for parity of some statistic of the classifier across these groups. Constraints
of this form are susceptible to intentional or inadvertent "fairness
gerrymandering", in which a classifier appears to be fair on each individual
group, but badly violates the fairness constraint on one or more structured
subgroups defined over the protected attributes. We propose instead to demand
statistical notions of fairness across exponentially (or infinitely) many
subgroups, defined by a structured class of functions over the protected
attributes. This interpolates between statistical definitions of fairness and
recently proposed individual notions of fairness, but raises several
computational challenges. It is no longer clear how to audit a fixed classifier
to see if it satisfies such a strong definition of fairness. We prove that the
computational problem of auditing subgroup fairness for both equality of false
positive rates and statistical parity is equivalent to the problem of weak
agnostic learning, which means it is computationally hard in the worst case,
even for simple structured subclasses.
We then derive two algorithms that provably converge to the best fair
classifier, given access to oracles which can solve the agnostic learning
problem. The algorithms are based on a formulation of subgroup fairness as a
two-player zero-sum game between a Learner and an Auditor. Our first algorithm
provably converges in a polynomial number of steps. Our second algorithm enjoys
only provably asymptotic convergence, but has the merit of simplicity and
faster per-step computation. We implement the simpler algorithm using linear
regression as a heuristic oracle, and show that we can effectively both audit
and learn fair classifiers on real datasets.
|
On Extending Neural Networks with Loss Ensembles for Text Classification | Ensemble techniques are powerful approaches that combine several weak
learners to build a stronger one. As a meta learning framework, ensemble
techniques can easily be applied to many machine learning techniques. In this
paper we propose a neural network extended with an ensemble loss function for
text classification. The weight of each weak loss function is tuned within the
training phase through the gradient propagation optimization method of the
neural network. The approach is evaluated on several text classification
datasets. We also evaluate its performance in various environments with several
degrees of label noise. Experimental results indicate an improvement of the
results and strong resilience against label noise in comparison with other
methods.
|
Near-Optimal Discrete Optimization for Experimental Design: A Regret
Minimization Approach | The experimental design problem concerns the selection of k points from a
potentially large design pool of p-dimensional vectors, so as to maximize the
statistical efficiency regressed on the selected k design points. Statistical
efficiency is measured by optimality criteria, including A(verage),
D(eterminant), T(race), E(igen), V(ariance) and G-optimality. Except for the
T-optimality, exact optimization is NP-hard.
We propose a polynomial-time regret minimization framework to achieve a
$(1+\varepsilon)$ approximation with only $O(p/\varepsilon^2)$ design points,
for all the optimality criteria above.
In contrast, to the best of our knowledge, before our work, no
polynomial-time algorithm achieves $(1+\varepsilon)$ approximations for
D/E/G-optimality, and the best poly-time algorithm achieving
$(1+\varepsilon)$-approximation for A/V-optimality requires $k =
\Omega(p^2/\varepsilon)$ design points.
|
CryptoDL: Deep Neural Networks over Encrypted Data | Machine learning algorithms based on deep neural networks have achieved
remarkable results and are being extensively used in different domains.
However, the machine learning algorithms requires access to raw data which is
often privacy sensitive. To address this issue, we develop new techniques to
provide solutions for running deep neural networks over encrypted data. In this
paper, we develop new techniques to adopt deep neural networks within the
practical limitation of current homomorphic encryption schemes. More
specifically, we focus on classification of the well-known convolutional neural
networks (CNN). First, we design methods for approximation of the activation
functions commonly used in CNNs (i.e. ReLU, Sigmoid, and Tanh) with low degree
polynomials which is essential for efficient homomorphic encryption schemes.
Then, we train convolutional neural networks with the approximation polynomials
instead of original activation functions and analyze the performance of the
models. Finally, we implement convolutional neural networks over encrypted data
and measure performance of the models. Our experimental results validate the
soundness of our approach with several convolutional neural networks with
varying number of layers and structures. When applied to the MNIST optical
character recognition tasks, our approach achieves 99.52\% accuracy which
significantly outperforms the state-of-the-art solutions and is very close to
the accuracy of the best non-private version, 99.77\%. Also, it can make close
to 164000 predictions per hour. We also applied our approach to CIFAR-10, which
is much more complex compared to MNIST, and were able to achieve 91.5\%
accuracy with approximation polynomials used as activation functions. These
results show that CryptoDL provides efficient, accurate and scalable
privacy-preserving predictions.
|
A learning problem that is independent of the set theory ZFC axioms | We consider the following statistical estimation problem: given a family F of
real valued functions over some domain X and an i.i.d. sample drawn from an
unknown distribution P over X, find h in F such that the expectation of h
w.r.t. P is probably approximately equal to the supremum over expectations on
members of F. This Expectation Maximization (EMX) problem captures many well
studied learning problems; in fact, it is equivalent to Vapnik's general
setting of learning.
Surprisingly, we show that the EMX learnability, as well as the learning
rates of some basic class F, depend on the cardinality of the continuum and is
therefore independent of the set theory ZFC axioms (that are widely accepted as
a formalization of the notion of a mathematical proof).
We focus on the case where the functions in F are Boolean, which generalizes
classification problems. We study the interaction between the statistical
sample complexity of F and its combinatorial structure. We introduce a new
version of sample compression schemes and show that it characterizes EMX
learnability for a wide family of classes. However, we show that for the class
of finite subsets of the real line, the existence of such compression schemes
is independent of set theory. We conclude that the learnability of that class
with respect to the family of probability distributions of countable support is
independent of the set theory ZFC axioms.
We also explore the existence of a "VC-dimension-like" parameter that
captures learnability in this setting. Our results imply that that there exist
no "finitary" combinatorial parameter that characterizes EMX learnability in a
way similar to the VC-dimension based characterization of binary valued
classification problems.
|
Joint Gaussian Processes for Biophysical Parameter Retrieval | Solving inverse problems is central to geosciences and remote sensing.
Radiative transfer models (RTMs) represent mathematically the physical laws
which govern the phenomena in remote sensing applications (forward models). The
numerical inversion of the RTM equations is a challenging and computationally
demanding problem, and for this reason, often the application of a nonlinear
statistical regression is preferred. In general, regression models predict the
biophysical parameter of interest from the corresponding received radiance.
However, this approach does not employ the physical information encoded in the
RTMs. An alternative strategy, which attempts to include the physical
knowledge, consists in learning a regression model trained using data simulated
by an RTM code. In this work, we introduce a nonlinear nonparametric regression
model which combines the benefits of the two aforementioned approaches. The
inversion is performed taking into account jointly both real observations and
RTM-simulated data. The proposed Joint Gaussian Process (JGP) provides a solid
framework for exploiting the regularities between the two types of data. The
JGP automatically detects the relative quality of the simulated and real data,
and combines them accordingly. This occurs by learning an additional
hyper-parameter w.r.t. a standard GP model, and fitting parameters through
maximizing the pseudo-likelihood of the real observations. The resulting scheme
is both simple and robust, i.e., capable of adapting to different scenarios.
The advantages of the JGP method compared to benchmark strategies are shown
considering RTM-simulated and real observations in different experiments.
Specifically, we consider leaf area index (LAI) retrieval from Landsat data
combined with simulated data generated by the PROSAIL model.
|
CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep
Learning | We develop an algorithm that can detect pneumonia from chest X-rays at a
level exceeding practicing radiologists. Our algorithm, CheXNet, is a 121-layer
convolutional neural network trained on ChestX-ray14, currently the largest
publicly available chest X-ray dataset, containing over 100,000 frontal-view
X-ray images with 14 diseases. Four practicing academic radiologists annotate a
test set, on which we compare the performance of CheXNet to that of
radiologists. We find that CheXNet exceeds average radiologist performance on
the F1 metric. We extend CheXNet to detect all 14 diseases in ChestX-ray14 and
achieve state of the art results on all 14 diseases.
|
Weakly-supervised Semantic Parsing with Abstract Examples | Training semantic parsers from weak supervision (denotations) rather than
strong supervision (programs) complicates training in two ways. First, a large
search space of potential programs needs to be explored at training time to
find a correct program. Second, spurious programs that accidentally lead to a
correct denotation add noise to training. In this work we propose that in
closed worlds with clear semantic types, one can substantially alleviate these
problems by utilizing an abstract representation, where tokens in both the
language utterance and program are lifted to an abstract form. We show that
these abstractions can be defined with a handful of lexical rules and that they
result in sharing between different examples that alleviates the difficulties
in training. To test our approach, we develop the first semantic parser for
CNLVR, a challenging visual reasoning dataset, where the search space is large
and overcoming spuriousness is critical, because denotations are either TRUE or
FALSE, and thus random programs are likely to lead to a correct denotation. Our
method substantially improves performance, and reaches 82.5% accuracy, a 14.7%
absolute accuracy improvement compared to the best reported accuracy so far.
|
Loss Functions for Multiset Prediction | We study the problem of multiset prediction. The goal of multiset prediction
is to train a predictor that maps an input to a multiset consisting of multiple
items. Unlike existing problems in supervised learning, such as classification,
ranking and sequence generation, there is no known order among items in a
target multiset, and each item in the multiset may appear more than once,
making this problem extremely challenging. In this paper, we propose a novel
multiset loss function by viewing this problem from the perspective of
sequential decision making. The proposed multiset loss function is empirically
evaluated on two families of datasets, one synthetic and the other real, with
varying levels of difficulty, against various baseline loss functions including
reinforcement learning, sequence, and aggregated distribution matching loss
functions. The experiments reveal the effectiveness of the proposed loss
function over the others.
|
Learning Image-Conditioned Dynamics Models for Control of Under-actuated
Legged Millirobots | Millirobots are a promising robotic platform for many applications due to
their small size and low manufacturing costs. Legged millirobots, in
particular, can provide increased mobility in complex environments and improved
scaling of obstacles. However, controlling these small, highly dynamic, and
underactuated legged systems is difficult. Hand-engineered controllers can
sometimes control these legged millirobots, but they have difficulties with
dynamic maneuvers and complex terrains. We present an approach for controlling
a real-world legged millirobot that is based on learned neural network models.
Using less than 17 minutes of data, our method can learn a predictive model of
the robot's dynamics that can enable effective gaits to be synthesized on the
fly for following user-specified waypoints on a given terrain. Furthermore, by
leveraging expressive, high-capacity neural network models, our approach allows
for these predictions to be directly conditioned on camera images, endowing the
robot with the ability to predict how different terrains might affect its
dynamics. This enables sample-efficient and effective learning for locomotion
of a dynamic legged millirobot on various terrains, including gravel, turf,
carpet, and styrofoam. Experiment videos can be found at
https://sites.google.com/view/imageconddyn
|
Deep-ESN: A Multiple Projection-encoding Hierarchical Reservoir
Computing Framework | As an efficient recurrent neural network (RNN) model, reservoir computing
(RC) models, such as Echo State Networks, have attracted widespread attention
in the last decade. However, while they have had great success with time series
data [1], [2], many time series have a multiscale structure, which a
single-hidden-layer RC model may have difficulty capturing. In this paper, we
propose a novel hierarchical reservoir computing framework we call Deep Echo
State Networks (Deep-ESNs). The most distinctive feature of a Deep-ESN is its
ability to deal with time series through hierarchical projections.
Specifically, when an input time series is projected into the high-dimensional
echo-state space of a reservoir, a subsequent encoding layer (e.g., a PCA,
autoencoder, or a random projection) can project the echo-state representations
into a lower-dimensional space. These low-dimensional representations can then
be processed by another ESN. By using projection layers and encoding layers
alternately in the hierarchical framework, a Deep-ESN can not only attenuate
the effects of the collinearity problem in ESNs, but also fully take advantage
of the temporal kernel property of ESNs to explore multiscale dynamics of time
series. To fuse the multiscale representations obtained by each reservoir, we
add connections from each encoding layer to the last output layer. Theoretical
analyses prove that stability of a Deep-ESN is guaranteed by the echo state
property (ESP), and the time complexity is equivalent to a conventional ESN.
Experimental results on some artificial and real world time series demonstrate
that Deep-ESNs can capture multiscale dynamics, and outperform both standard
ESNs and previous hierarchical ESN-based models.
|
An Accelerated Communication-Efficient Primal-Dual Optimization
Framework for Structured Machine Learning | Distributed optimization algorithms are essential for training machine
learning models on very large-scale datasets. However, they often suffer from
communication bottlenecks. Confronting this issue, a communication-efficient
primal-dual coordinate ascent framework (CoCoA) and its improved variant CoCoA+
have been proposed, achieving a convergence rate of $\mathcal{O}(1/t)$ for
solving empirical risk minimization problems with Lipschitz continuous losses.
In this paper, an accelerated variant of CoCoA+ is proposed and shown to
possess a convergence rate of $\mathcal{O}(1/t^2)$ in terms of reducing
suboptimality. The analysis of this rate is also notable in that the
convergence rate bounds involve constants that, except in extreme cases, are
significantly reduced compared to those previously provided for CoCoA+. The
results of numerical experiments are provided to show that acceleration can
lead to significant performance gains.
|
On Optimal Generalizability in Parametric Learning | We consider the parametric learning problem, where the objective of the
learner is determined by a parametric loss function. Employing empirical risk
minimization with possibly regularization, the inferred parameter vector will
be biased toward the training samples. Such bias is measured by the cross
validation procedure in practice where the data set is partitioned into a
training set used for training and a validation set, which is not used in
training and is left to measure the out-of-sample performance. A classical
cross validation strategy is the leave-one-out cross validation (LOOCV) where
one sample is left out for validation and training is done on the rest of the
samples that are presented to the learner, and this process is repeated on all
of the samples. LOOCV is rarely used in practice due to the high computational
complexity. In this paper, we first develop a computationally efficient
approximate LOOCV (ALOOCV) and provide theoretical guarantees for its
performance. Then we use ALOOCV to provide an optimization algorithm for
finding the regularizer in the empirical risk minimization framework. In our
numerical experiments, we illustrate the accuracy and efficiency of ALOOCV as
well as our proposed framework for the optimization of the regularizer.
|
LIUBoost : Locality Informed Underboosting for Imbalanced Data
Classification | The problem of class imbalance along with class-overlapping has become a
major issue in the domain of supervised learning. Most supervised learning
algorithms assume equal cardinality of the classes under consideration while
optimizing the cost function and this assumption does not hold true for
imbalanced datasets which results in sub-optimal classification. Therefore,
various approaches, such as undersampling, oversampling, cost-sensitive
learning and ensemble based methods have been proposed for dealing with
imbalanced datasets. However, undersampling suffers from information loss,
oversampling suffers from increased runtime and potential overfitting while
cost-sensitive methods suffer due to inadequately defined cost assignment
schemes. In this paper, we propose a novel boosting based method called
LIUBoost. LIUBoost uses under sampling for balancing the datasets in every
boosting iteration like RUSBoost while incorporating a cost term for every
instance based on their hardness into the weight update formula minimizing the
information loss introduced by undersampling. LIUBoost has been extensively
evaluated on 18 imbalanced datasets and the results indicate significant
improvement over existing best performing method RUSBoost.
|
Optimizing Kernel Machines using Deep Learning | Building highly non-linear and non-parametric models is central to several
state-of-the-art machine learning systems. Kernel methods form an important
class of techniques that induce a reproducing kernel Hilbert space (RKHS) for
inferring non-linear models through the construction of similarity functions
from data. These methods are particularly preferred in cases where the training
data sizes are limited and when prior knowledge of the data similarities is
available. Despite their usefulness, they are limited by the computational
complexity and their inability to support end-to-end learning with a
task-specific objective. On the other hand, deep neural networks have become
the de facto solution for end-to-end inference in several learning paradigms.
In this article, we explore the idea of using deep architectures to perform
kernel machine optimization, for both computational efficiency and end-to-end
inferencing. To this end, we develop the DKMO (Deep Kernel Machine
Optimization) framework, that creates an ensemble of dense embeddings using
Nystrom kernel approximations and utilizes deep learning to generate
task-specific representations through the fusion of the embeddings.
Intuitively, the filters of the network are trained to fuse information from an
ensemble of linear subspaces in the RKHS. Furthermore, we introduce the kernel
dropout regularization to enable improved training convergence. Finally, we
extend this framework to the multiple kernel case, by coupling a global fusion
layer with pre-trained deep kernel machines for each of the constituent
kernels. Using case studies with limited training data, and lack of explicit
feature sources, we demonstrate the effectiveness of our framework over
conventional model inferencing techniques.
|
Sliced Wasserstein Distance for Learning Gaussian Mixture Models | Gaussian mixture models (GMM) are powerful parametric tools with many
applications in machine learning and computer vision. Expectation maximization
(EM) is the most popular algorithm for estimating the GMM parameters. However,
EM guarantees only convergence to a stationary point of the log-likelihood
function, which could be arbitrarily worse than the optimal solution. Inspired
by the relationship between the negative log-likelihood function and the
Kullback-Leibler (KL) divergence, we propose an alternative formulation for
estimating the GMM parameters using the sliced Wasserstein distance, which
gives rise to a new algorithm. Specifically, we propose minimizing the
sliced-Wasserstein distance between the mixture model and the data distribution
with respect to the GMM parameters. In contrast to the KL-divergence, the
energy landscape for the sliced-Wasserstein distance is more well-behaved and
therefore more suitable for a stochastic gradient descent scheme to obtain the
optimal GMM parameters. We show that our formulation results in parameter
estimates that are more robust to random initializations and demonstrate that
it can estimate high-dimensional data distributions more faithfully than the EM
algorithm.
|
Semiblind subgraph reconstruction in Gaussian graphical models | Consider a social network where only a few nodes (agents) have meaningful
interactions in the sense that the conditional dependency graph over node
attribute variables (behaviors) is sparse. A company that can only observe the
interactions between its own customers will generally not be able to accurately
estimate its customers' dependency subgraph: it is blinded to any external
interactions of its customers and this blindness creates false edges in its
subgraph. In this paper we address the semiblind scenario where the company has
access to a noisy summary of the complementary subgraph connecting external
agents, e.g., provided by a consolidator. The proposed framework applies to
other applications as well, including field estimation from a network of awake
and sleeping sensors and privacy-constrained information sharing over social
subnetworks. We propose a penalized likelihood approach in the context of a
graph signal obeying a Gaussian graphical models (GGM). We use a convex-concave
iterative optimization algorithm to maximize the penalized likelihood.
|
Revisiting Simple Neural Networks for Learning Representations of
Knowledge Graphs | We address the problem of learning vector representations for entities and
relations in Knowledge Graphs (KGs) for Knowledge Base Completion (KBC). This
problem has received significant attention in the past few years and multiple
methods have been proposed. Most of the existing methods in the literature use
a predefined characteristic scoring function for evaluating the correctness of
KG triples. These scoring functions distinguish correct triples (high score)
from incorrect ones (low score). However, their performance vary across
different datasets. In this work, we demonstrate that a simple neural network
based score function can consistently achieve near start-of-the-art performance
on multiple datasets. We also quantitatively demonstrate biases in standard
benchmark datasets, and highlight the need to perform evaluation spanning
various datasets.
|
A Fast and Robust TSVM for Pattern Classification | Twin support vector machine~(TSVM) is a powerful learning algorithm by
solving a pair of smaller SVM-type problems. However, there are still some
specific issues such as low efficiency and weak robustness when it is faced
with some real applications. In this paper, we propose a Fast and Robust
TSVM~(FR-TSVM) to deal with the above issues. In order to alleviate the effects
of noisy inputs, we propose an effective fuzzy membership function and
reformulate the TSVMs such that different input instances can make different
contributions to the learning of the separating hyperplanes. To further speed
up the training procedure, we develop an efficient coordinate descent algorithm
with shirking to solve the involved a pair of quadratic programming problems
(QPPs). Moreover, theoretical foundations of the proposed model are analyzed in
details. The experimental results on several artificial and benchmark datasets
indicate that the FR-TSVM not only obtains a fast learning speed but also shows
a robust classification performance. Code has been made available at:
https://github.com/gaobb/FR-TSVM.
|
MARGIN: Uncovering Deep Neural Networks using Graph Signal Analysis | Interpretability has emerged as a crucial aspect of building trust in machine
learning systems, aimed at providing insights into the working of complex
neural networks that are otherwise opaque to a user. There are a plethora of
existing solutions addressing various aspects of interpretability ranging from
identifying prototypical samples in a dataset to explaining image predictions
or explaining mis-classifications. While all of these diverse techniques
address seemingly different aspects of interpretability, we hypothesize that a
large family of interepretability tasks are variants of the same central
problem which is identifying \emph{relative} change in a model's prediction.
This paper introduces MARGIN, a simple yet general approach to address a large
set of interpretability tasks MARGIN exploits ideas rooted in graph signal
analysis to determine influential nodes in a graph, which are defined as those
nodes that maximally describe a function defined on the graph. By carefully
defining task-specific graphs and functions, we demonstrate that MARGIN
outperforms existing approaches in a number of disparate interpretability
challenges.
|
Z-Forcing: Training Stochastic Recurrent Networks | Many efforts have been devoted to training generative latent variable models
with autoregressive decoders, such as recurrent neural networks (RNN).
Stochastic recurrent models have been successful in capturing the variability
observed in natural sequential data such as speech. We unify successful ideas
from recently proposed architectures into a stochastic recurrent model: each
step in the sequence is associated with a latent variable that is used to
condition the recurrent dynamics for future steps. Training is performed with
amortized variational inference where the approximate posterior is augmented
with a RNN that runs backward through the sequence. In addition to maximizing
the variational lower bound, we ease training of the latent variables by adding
an auxiliary cost which forces them to reconstruct the state of the backward
recurrent network. This provides the latent variables with a task-independent
objective that enhances the performance of the overall model. We found this
strategy to perform better than alternative approaches such as KL annealing.
Although being conceptually simple, our model achieves state-of-the-art results
on standard speech benchmarks such as TIMIT and Blizzard and competitive
performance on sequential MNIST. Finally, we apply our model to language
modeling on the IMDB dataset where the auxiliary cost helps in learning
interpretable latent variables. Source Code:
\url{https://github.com/anirudh9119/zforcing_nips17}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.