title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Fingerprinting-Based Positioning in Distributed Massive MIMO Systems | Location awareness in wireless networks may enable many applications such as
emergency services, autonomous driving and geographic routing. Although there
are many available positioning techniques, none of them is adapted to work with
massive multiple-in-multiple-out (MIMO) systems, which represent a leading 5G
technology candidate. In this paper, we discuss possible solutions for
positioning of mobile stations using a vector of signals at the base station,
equipped with many antennas distributed over deployment area. Our main proposal
is to use fingerprinting techniques based on a vector of received signal
strengths. This kind of methods are able to work in highly-cluttered multipath
environments, and require just one base station, in contrast to standard
range-based and angle-based techniques. We also provide a solution for
fingerprinting-based positioning based on Gaussian process regression, and
discuss main applications and challenges.
|
Fast Randomized Singular Value Thresholding for Low-rank Optimization | Rank minimization can be converted into tractable surrogate problems, such as
Nuclear Norm Minimization (NNM) and Weighted NNM (WNNM). The problems related
to NNM, or WNNM, can be solved iteratively by applying a closed-form proximal
operator, called Singular Value Thresholding (SVT), or Weighted SVT, but they
suffer from high computational cost of Singular Value Decomposition (SVD) at
each iteration. We propose a fast and accurate approximation method for SVT,
that we call fast randomized SVT (FRSVT), with which we avoid direct
computation of SVD. The key idea is to extract an approximate basis for the
range of the matrix from its compressed matrix. Given the basis, we compute
partial singular values of the original matrix from the small factored matrix.
In addition, by developping a range propagation method, our method further
speeds up the extraction of approximate basis at each iteration. Our
theoretical analysis shows the relationship between the approximation bound of
SVD and its effect to NNM via SVT. Along with the analysis, our empirical
results quantitatively and qualitatively show that our approximation rarely
harms the convergence of the host algorithms. We assess the efficiency and
accuracy of the proposed method on various computer vision problems, e.g.,
subspace clustering, weather artifact removal, and simultaneous multi-image
alignment and rectification.
|
Sensor-Type Classification in Buildings | Many sensors/meters are deployed in commercial buildings to monitor and
optimize their performance. However, because sensor metadata is inconsistent
across buildings, software-based solutions are tightly coupled to the sensor
metadata conventions (i.e. schemas and naming) for each building. Running the
same software across buildings requires significant integration effort.
Metadata normalization is critical for scaling the deployment process and
allows us to decouple building-specific conventions from the code written for
building applications. It also allows us to deal with missing metadata. One
important aspect of normalization is to differentiate sensors by the typeof
phenomena being observed. In this paper, we propose a general, simple, yet
effective classification scheme to differentiate sensors in buildings by type.
We perform ensemble learning on data collected from over 2000 sensor streams in
two buildings. Our approach is able to achieve more than 92% accuracy for
classification within buildings and more than 82% accuracy for across
buildings. We also introduce a method for identifying potential misclassified
streams. This is important because it allows us to identify opportunities to
attain more input from experts -- input that could help improve classification
accuracy when ground truth is unavailable. We show that by adjusting a
threshold value we are able to identify at least 30% of the misclassified
instances.
|
Importance Weighted Autoencoders | The variational autoencoder (VAE; Kingma, Welling (2014)) is a recently
proposed generative model pairing a top-down generative network with a
bottom-up recognition network which approximates posterior inference. It
typically makes strong assumptions about posterior inference, for instance that
the posterior distribution is approximately factorial, and that its parameters
can be approximated with nonlinear regression from the observations. As we show
empirically, the VAE objective can lead to overly simplified representations
which fail to use the network's entire modeling capacity. We present the
importance weighted autoencoder (IWAE), a generative model with the same
architecture as the VAE, but which uses a strictly tighter log-likelihood lower
bound derived from importance weighting. In the IWAE, the recognition network
uses multiple samples to approximate the posterior, giving it increased
flexibility to model complex posteriors which do not fit the VAE modeling
assumptions. We show empirically that IWAEs learn richer latent space
representations than VAEs, leading to improved test log-likelihood on density
estimation benchmarks.
|
Discovery of Web Usage Profiles Using Various Clustering Techniques | The explosive growth of World Wide Web (WWW) has necessitated the development
of Web personalization systems in order to understand the user preferences to
dynamically serve customized content to individual users. To reveal information
about user preferences from Web usage data, Web Usage Mining (WUM) techniques
are extensively being applied to the Web log data. Clustering techniques are
widely used in WUM to capture similar interests and trends among users
accessing a Web site. Clustering aims to divide a data set into groups or
clusters where inter-cluster similarities are minimized while the intra cluster
similarities are maximized. This paper reviews four of the popularly used
clustering techniques: k-Means, k-Medoids, Leader and DBSCAN. These techniques
are implemented and tested against the Web user navigational data. Performance
and validity results of each technique are presented and compared.
|
Heavy-tailed Independent Component Analysis | Independent component analysis (ICA) is the problem of efficiently recovering
a matrix $A \in \mathbb{R}^{n\times n}$ from i.i.d. observations of $X=AS$
where $S \in \mathbb{R}^n$ is a random vector with mutually independent
coordinates. This problem has been intensively studied, but all existing
efficient algorithms with provable guarantees require that the coordinates
$S_i$ have finite fourth moments. We consider the heavy-tailed ICA problem
where we do not make this assumption, about the second moment. This problem
also has received considerable attention in the applied literature. In the
present work, we first give a provably efficient algorithm that works under the
assumption that for constant $\gamma > 0$, each $S_i$ has finite
$(1+\gamma)$-moment, thus substantially weakening the moment requirement
condition for the ICA problem to be solvable. We then give an algorithm that
works under the assumption that matrix $A$ has orthogonal columns but requires
no moment assumptions. Our techniques draw ideas from convex geometry and
exploit standard properties of the multivariate spherical Gaussian distribution
in a novel way.
|
A DEEP analysis of the META-DES framework for dynamic selection of
ensemble of classifiers | Dynamic ensemble selection (DES) techniques work by estimating the level of
competence of each classifier from a pool of classifiers. Only the most
competent ones are selected to classify a given test sample. Hence, the key
issue in DES is the criterion used to estimate the level of competence of the
classifiers in predicting the label of a given test sample. In order to perform
a more robust ensemble selection, we proposed the META-DES framework using
meta-learning, where multiple criteria are encoded as meta-features and are
passed down to a meta-classifier that is trained to estimate the competence
level of a given classifier. In this technical report, we present a
step-by-step analysis of each phase of the framework during training and test.
We show how each set of meta-features is extracted as well as their impact on
the estimation of the competence level of the base classifier. Moreover, an
analysis of the impact of several factors in the system performance, such as
the number of classifiers in the pool, the use of different linear base
classifiers, as well as the size of the validation data. We show that using the
dynamic selection of linear classifiers through the META-DES framework, we can
solve complex non-linear classification problems where other combination
techniques such as AdaBoost cannot.
|
What to talk about and how? Selective Generation using LSTMs with
Coarse-to-Fine Alignment | We propose an end-to-end, domain-independent neural encoder-aligner-decoder
model for selective generation, i.e., the joint task of content selection and
surface realization. Our model first encodes a full set of over-determined
database event records via an LSTM-based recurrent neural network, then
utilizes a novel coarse-to-fine aligner to identify the small subset of salient
records to talk about, and finally employs a decoder to generate free-form
descriptions of the aligned, selected records. Our model achieves the best
selection and generation results reported to-date (with 59% relative
improvement in generation) on the benchmark WeatherGov dataset, despite using
no specialized features or linguistic resources. Using an improved k-nearest
neighbor beam filter helps further. We also perform a series of ablations and
visualizations to elucidate the contributions of our key model components.
Lastly, we evaluate the generalizability of our model on the RoboCup dataset,
and get results that are competitive with or better than the state-of-the-art,
despite being severely data-starved.
|
On-the-Fly Learning in a Perpetual Learning Machine | Despite the promise of brain-inspired machine learning, deep neural networks
(DNN) have frustratingly failed to bridge the deceptively large gap between
learning and memory. Here, we introduce a Perpetual Learning Machine; a new
type of DNN that is capable of brain-like dynamic 'on the fly' learning because
it exists in a self-supervised state of Perpetual Stochastic Gradient Descent.
Thus, we provide the means to unify learning and memory within a machine
learning framework. We also explore the elegant duality of abstraction and
synthesis: the Yin and Yang of deep learning.
|
Bayesian Masking: Sparse Bayesian Estimation with Weaker Shrinkage Bias | A common strategy for sparse linear regression is to introduce
regularization, which eliminates irrelevant features by letting the
corresponding weights be zeros. However, regularization often shrinks the
estimator for relevant features, which leads to incorrect feature selection.
Motivated by the above-mentioned issue, we propose Bayesian masking (BM), a
sparse estimation method which imposes no regularization on the weights. The
key concept of BM is to introduce binary latent variables that randomly mask
features. Estimating the masking rates determines the relevance of the features
automatically. We derive a variational Bayesian inference algorithm that
maximizes the lower bound of the factorized information criterion (FIC), which
is a recently developed asymptotic criterion for evaluating the marginal
log-likelihood. In addition, we propose reparametrization to accelerate the
convergence of the derived algorithm. Finally, we show that BM outperforms
Lasso and automatic relevance determination (ARD) in terms of the
sparsity-shrinkage trade-off.
|
Training a Restricted Boltzmann Machine for Classification by Labeling
Model Samples | We propose an alternative method for training a classification model. Using
the MNIST set of handwritten digits and Restricted Boltzmann Machines, it is
possible to reach a classification performance competitive to semi-supervised
learning if we first train a model in an unsupervised fashion on unlabeled data
only, and then manually add labels to model samples instead of training data
samples with the help of a GUI. This approach can benefit from the fact that
model samples can be presented to the human labeler in a video-like fashion,
resulting in a higher number of labeled examples. Also, after some initial
training, hard-to-classify examples can be distinguished from easy ones
automatically, saving manual work.
|
A tree-based kernel for graphs with continuous attributes | The availability of graph data with node attributes that can be either
discrete or real-valued is constantly increasing. While existing kernel methods
are effective techniques for dealing with graphs having discrete node labels,
their adaptation to non-discrete or continuous node attributes has been
limited, mainly for computational issues. Recently, a few kernels especially
tailored for this domain, and that trade predictive performance for
computational efficiency, have been proposed. In this paper, we propose a graph
kernel for complex and continuous nodes' attributes, whose features are tree
structures extracted from specific graph visits. The kernel manages to keep the
same complexity of state-of-the-art kernels while implicitly using a larger
feature space. We further present an approximated variant of the kernel which
reduces its complexity significantly. Experimental results obtained on six
real-world datasets show that the kernel is the best performing one on most of
them. Moreover, in most cases the approximated version reaches comparable
performances to current state-of-the-art kernels in terms of classification
accuracy while greatly shortening the running times.
|
Semi-described and semi-supervised learning with Gaussian processes | Propagating input uncertainty through non-linear Gaussian process (GP)
mappings is intractable. This hinders the task of training GPs using uncertain
and partially observed inputs. In this paper we refer to this task as
"semi-described learning". We then introduce a GP framework that solves both,
the semi-described and the semi-supervised learning problems (where missing
values occur in the outputs). Auto-regressive state space simulation is also
recognised as a special case of semi-described learning. To achieve our goal we
develop variational methods for handling semi-described inputs in GPs, and
couple them with algorithms that allow for imputing the missing values while
treating the uncertainty in a principled, Bayesian manner. Extensive
experiments on simulated and real-world data study the problems of iterative
forecasting and regression/classification with missing values. The results
suggest that the principled propagation of uncertainty stemming from our
framework can significantly improve performance in these tasks.
|
Fast Clustering and Topic Modeling Based on Rank-2 Nonnegative Matrix
Factorization | The importance of unsupervised clustering and topic modeling is well
recognized with ever-increasing volumes of text data. In this paper, we propose
a fast method for hierarchical clustering and topic modeling called HierNMF2.
Our method is based on fast Rank-2 nonnegative matrix factorization (NMF) that
performs binary clustering and an efficient node splitting rule. Further
utilizing the final leaf nodes generated in HierNMF2 and the idea of
nonnegative least squares fitting, we propose a new clustering/topic modeling
method called FlatNMF2 that recovers a flat clustering/topic modeling result in
a very simple yet significantly more effective way than any other existing
methods. We implement highly optimized open source software in C++ for both
HierNMF2 and FlatNMF2 for hierarchical and partitional clustering/topic
modeling of document data sets.
Substantial experimental tests are presented that illustrate significant
improvements both in computational time as well as quality of solutions. We
compare our methods to other clustering methods including K-means, standard
NMF, and CLUTO, and also topic modeling methods including latent Dirichlet
allocation (LDA) and recently proposed algorithms for NMF with separability
constraints. Overall, we present efficient tools for analyzing large-scale data
sets, and techniques that can be generalized to many other data analytics
problem domains.
|
Train faster, generalize better: Stability of stochastic gradient
descent | We show that parametric models trained by a stochastic gradient method (SGM)
with few iterations have vanishing generalization error. We prove our results
by arguing that SGM is algorithmically stable in the sense of Bousquet and
Elisseeff. Our analysis only employs elementary tools from convex and
continuous optimization. We derive stability bounds for both convex and
non-convex optimization under standard Lipschitz and smoothness assumptions.
Applying our results to the convex case, we provide new insights for why
multiple epochs of stochastic gradient methods generalize well in practice. In
the non-convex case, we give a new interpretation of common practices in neural
networks, and formally show that popular techniques for training large deep
models are indeed stability-promoting. Our findings conceptually underscore the
importance of reducing training time beyond its obvious benefit.
|
Machine Learning Methods to Analyze Arabidopsis Thaliana Plant Root
Growth | One of the challenging problems in biology is to classify plants based on
their reaction on genetic mutation. Arabidopsis Thaliana is a plant that is so
interesting, because its genetic structure has some similarities with that of
human beings. Biologists classify the type of this plant to mutated and not
mutated (wild) types. Phenotypic analysis of these types is a time-consuming
and costly effort by individuals. In this paper, we propose a modified feature
extraction step by using velocity and acceleration of root growth. In the
second step, for plant classification, we employed different Support Vector
Machine (SVM) kernels and two hybrid systems of neural networks. Gated Negative
Correlation Learning (GNCL) and Mixture of Negatively Correlated Experts (MNCE)
are two ensemble methods based on complementary feature of classical
classifiers; Mixture of Expert (ME) and Negative Correlation Learning (NCL).
The hybrid systems conserve of advantages and decrease the effects of
disadvantages of NCL and ME. Our Experimental shows that MNCE and GNCL improve
the efficiency of classical classifiers, however, some SVM kernels function has
better performance than classifiers based on neural network ensemble method.
Moreover, kernels consume less time to obtain a classification rate.
|
Probabilistic Neural Network Training for Semi-Supervised Classifiers | In this paper, we propose another version of help-training approach by
employing a Probabilistic Neural Network (PNN) that improves the performance of
the main discriminative classifier in the semi-supervised strategy. We
introduce the PNN-training algorithm and use it for training the support vector
machine (SVM) with a few numbers of labeled data and a large number of
unlabeled data. We try to find the best labels for unlabeled data and then use
SVM to enhance the classification rate. We test our method on two famous
benchmarks and show the efficiency of our method in comparison with pervious
methods.
|
Incremental Active Opinion Learning Over a Stream of Opinionated
Documents | Applications that learn from opinionated documents, like tweets or product
reviews, face two challenges. First, the opinionated documents constitute an
evolving stream, where both the author's attitude and the vocabulary itself may
change. Second, labels of documents are scarce and labels of words are
unreliable, because the sentiment of a word depends on the (unknown) context in
the author's mind. Most of the research on mining over opinionated streams
focuses on the first aspect of the problem, whereas for the second a continuous
supply of labels from the stream is assumed. Such an assumption though is
utopian as the stream is infinite and the labeling cost is prohibitive. To this
end, we investigate the potential of active stream learning algorithms that ask
for labels on demand. Our proposed ACOSTREAM 1 approach works with limited
labels: it uses an initial seed of labeled documents, occasionally requests
additional labels for documents from the human expert and incrementally adapts
to the underlying stream while exploiting the available labeled documents. In
its core, ACOSTREAM consists of a MNB classifier coupled with "sampling"
strategies for requesting class labels for new unlabeled documents. In the
experiments, we evaluate the classifier performance over time by varying: (a)
the class distribution of the opinionated stream, while assuming that the set
of the words in the vocabulary is fixed but their polarities may change with
the class distribution; and (b) the number of unknown words arriving at each
moment, while the class polarity may also change. Our results show that active
learning on a stream of opinionated documents, delivers good performance while
requiring a small selection of labels
|
l1-norm Penalized Orthogonal Forward Regression | A l1-norm penalized orthogonal forward regression (l1-POFR) algorithm is
proposed based on the concept of leaveone- out mean square error (LOOMSE).
Firstly, a new l1-norm penalized cost function is defined in the constructed
orthogonal space, and each orthogonal basis is associated with an individually
tunable regularization parameter. Secondly, due to orthogonal computation, the
LOOMSE can be analytically computed without actually splitting the data set,
and moreover a closed form of the optimal regularization parameter in terms of
minimal LOOMSE is derived. Thirdly, a lower bound for regularization parameters
is proposed, which can be used for robust LOOMSE estimation by adaptively
detecting and removing regressors to an inactive set so that the computational
cost of the algorithm is significantly reduced. Illustrative examples are
included to demonstrate the effectiveness of this new l1-POFR approach.
|
Deep Broad Learning - Big Models for Big Data | Deep learning has demonstrated the power of detailed modeling of complex
high-order (multivariate) interactions in data. For some learning tasks there
is power in learning models that are not only Deep but also Broad. By Broad, we
mean models that incorporate evidence from large numbers of features. This is
of especial value in applications where many different features and
combinations of features all carry small amounts of information about the
class. The most accurate models will integrate all that information. In this
paper, we propose an algorithm for Deep Broad Learning called DBL. The proposed
algorithm has a tunable parameter $n$, that specifies the depth of the model.
It provides straightforward paths towards out-of-core learning for large data.
We demonstrate that DBL learns models from large quantities of data with
accuracy that is highly competitive with the state-of-the-art.
|
Parallel and Distributed Approaches for Graph Based Semi-supervised
Learning | Two approaches for graph based semi-supervised learning are proposed. The
firstapproach is based on iteration of an affine map. A key element of the
affine map iteration is sparsematrix-vector multiplication, which has several
very efficient parallel implementations. The secondapproach belongs to the
class of Markov Chain Monte Carlo (MCMC) algorithms. It is based onsampling of
nodes by performing a random walk on the graph. The latter approach is
distributedby its nature and can be easily implemented on several processors or
over the network. Boththeoretical and practical evaluations are provided. It is
found that the nodes are classified intotheir class with very small error. The
sampling algorithm's ability to track new incoming nodesand to classify them is
also demonstrated.
|
Diffusion-KLMS Algorithm and its Performance Analysis for Non-Linear
Distributed Networks | In a distributed network environment, the diffusion-least mean squares (LMS)
algorithm gives faster convergence than the original LMS algorithm. It has also
been observed that, the diffusion-LMS generally outperforms other distributed
LMS algorithms like spatial LMS and incremental LMS. However, both the original
LMS and diffusion-LMS are not applicable in non-linear environments where data
may not be linearly separable. A variant of LMS called kernel-LMS (KLMS) has
been proposed in the literature for such non-linearities. In this paper, we
propose kernelised version of diffusion-LMS for non-linear distributed
environments. Simulations show that the proposed approach has superior
convergence as compared to algorithms of the same genre. We also introduce a
technique to predict the transient and steady-state behaviour of the proposed
algorithm. The techniques proposed in this work (or algorithms of same genre)
can be easily extended to distributed parameter estimation applications like
cooperative spectrum sensing and massive multiple input multiple output (MIMO)
receiver design which are potential components for 5G communication systems.
|
CNN Based Hashing for Image Retrieval | Along with data on the web increasing dramatically, hashing is becoming more
and more popular as a method of approximate nearest neighbor search. Previous
supervised hashing methods utilized similarity/dissimilarity matrix to get
semantic information. But the matrix is not easy to construct for a new
dataset. Rather than to reconstruct the matrix, we proposed a straightforward
CNN-based hashing method, i.e. binarilizing the activations of a fully
connected layer with threshold 0 and taking the binary result as hash codes.
This method achieved the best performance on CIFAR-10 and was comparable with
the state-of-the-art on MNIST. And our experiments on CIFAR-10 suggested that
the signs of activations may carry more information than the relative values of
activations between samples, and that the co-adaption between feature extractor
and hash functions is important for hashing.
|
Predicting SLA Violations in Real Time using Online Machine Learning | Detecting faults and SLA violations in a timely manner is critical for
telecom providers, in order to avoid loss in business, revenue and reputation.
At the same time predicting SLA violations for user services in telecom
environments is difficult, due to time-varying user demands and infrastructure
load conditions.
In this paper, we propose a service-agnostic online learning approach,
whereby the behavior of the system is learned on the fly, in order to predict
client-side SLA violations. The approach uses device-level metrics, which are
collected in a streaming fashion on the server side.
Our results show that the approach can produce highly accurate predictions
(>90% classification accuracy and < 10% false alarm rate) in scenarios where
SLA violations are predicted for a video-on-demand service under changing load
patterns. The paper also highlight the limitations of traditional offline
learning methods, which perform significantly worse in many of the considered
scenarios.
|
Coordinate Descent Methods for Symmetric Nonnegative Matrix
Factorization | Given a symmetric nonnegative matrix $A$, symmetric nonnegative matrix
factorization (symNMF) is the problem of finding a nonnegative matrix $H$,
usually with much fewer columns than $A$, such that $A \approx HH^T$. SymNMF
can be used for data analysis and in particular for various clustering tasks.
In this paper, we propose simple and very efficient coordinate descent schemes
to solve this problem, and that can handle large and sparse input matrices. The
effectiveness of our methods is illustrated on synthetic and real-world data
sets, and we show that they perform favorably compared to recent
state-of-the-art methods.
|
Quantization based Fast Inner Product Search | We propose a quantization based approach for fast approximate Maximum Inner
Product Search (MIPS). Each database vector is quantized in multiple subspaces
via a set of codebooks, learned directly by minimizing the inner product
quantization error. Then, the inner product of a query to a database vector is
approximated as the sum of inner products with the subspace quantizers.
Different from recently proposed LSH approaches to MIPS, the database vectors
and queries do not need to be augmented in a higher dimensional feature space.
We also provide a theoretical analysis of the proposed approach, consisting of
the concentration results under mild assumptions. Furthermore, if a small
sample of example queries is given at the training time, we propose a modified
codebook learning procedure which further improves the accuracy. Experimental
results on a variety of datasets including those arising from deep neural
networks show that the proposed approach significantly outperforms the existing
state-of-the-art.
|
EM Algorithms for Weighted-Data Clustering with Application to
Audio-Visual Scene Analysis | Data clustering has received a lot of attention and numerous methods,
algorithms and software packages are available. Among these techniques,
parametric finite-mixture models play a central role due to their interesting
mathematical properties and to the existence of maximum-likelihood estimators
based on expectation-maximization (EM). In this paper we propose a new mixture
model that associates a weight with each observed point. We introduce the
weighted-data Gaussian mixture and we derive two EM algorithms. The first one
considers a fixed weight for each observation. The second one treats each
weight as a random variable following a gamma distribution. We propose a model
selection method based on a minimum message length criterion, provide a weight
initialization strategy, and validate the proposed algorithms by comparing them
with several state of the art parametric and non-parametric clustering
techniques. We also demonstrate the effectiveness and robustness of the
proposed clustering technique in the presence of heterogeneous data, namely
audio-visual scene analysis.
|
Minimum Spectral Connectivity Projection Pursuit | We study the problem of determining the optimal low dimensional projection
for maximising the separability of a binary partition of an unlabelled dataset,
as measured by spectral graph theory. This is achieved by finding projections
which minimise the second eigenvalue of the graph Laplacian of the projected
data, which corresponds to a non-convex, non-smooth optimisation problem. We
show that the optimal univariate projection based on spectral connectivity
converges to the vector normal to the maximum margin hyperplane through the
data, as the scaling parameter is reduced to zero. This establishes a
connection between connectivity as measured by spectral graph theory and
maximal Euclidean separation. The computational cost associated with each
eigen-problem is quadratic in the number of data. To mitigate this issue, we
propose an approximation method using microclusters with provable approximation
error bounds. Combining multiple binary partitions within a divisive
hierarchical model allows us to construct clustering solutions admitting
clusters with varying scales and lying within different subspaces. We evaluate
the performance of the proposed method on a large collection of benchmark
datasets and find that it compares favourably with existing methods for
projection pursuit and dimension reduction for data clustering.
|
Giraffe: Using Deep Reinforcement Learning to Play Chess | This report presents Giraffe, a chess engine that uses self-play to discover
all its domain-specific knowledge, with minimal hand-crafted knowledge given by
the programmer. Unlike previous attempts using machine learning only to perform
parameter-tuning on hand-crafted evaluation functions, Giraffe's learning
system also performs automatic feature extraction and pattern recognition. The
trained evaluation function performs comparably to the evaluation functions of
state-of-the-art chess engines - all of which containing thousands of lines of
carefully hand-crafted pattern recognizers, tuned over many years by both
computer chess experts and human chess masters. Giraffe is the most successful
attempt thus far at using end-to-end machine learning to play chess.
|
Efficient Sampling for k-Determinantal Point Processes | Determinantal Point Processes (DPPs) are elegant probabilistic models of
repulsion and diversity over discrete sets of items. But their applicability to
large sets is hindered by expensive cubic-complexity matrix operations for
basic tasks such as sampling. In light of this, we propose a new method for
approximate sampling from discrete $k$-DPPs. Our method takes advantage of the
diversity property of subsets sampled from a DPP, and proceeds in two stages:
first it constructs coresets for the ground set of items; thereafter, it
efficiently samples subsets based on the constructed coresets. As opposed to
previous approaches, our algorithm aims to minimize the total variation
distance to the original distribution. Experiments on both synthetic and real
datasets indicate that our sampling algorithm works efficiently on large data
sets, and yields more accurate samples than previous approaches.
|
Character-level Convolutional Networks for Text Classification | This article offers an empirical exploration on the use of character-level
convolutional networks (ConvNets) for text classification. We constructed
several large-scale datasets to show that character-level convolutional
networks could achieve state-of-the-art or competitive results. Comparisons are
offered against traditional models such as bag of words, n-grams and their
TFIDF variants, and deep learning models such as word-based ConvNets and
recurrent neural networks.
|
Reinforcement Learning with Parameterized Actions | We introduce a model-free algorithm for learning in Markov decision processes
with parameterized actions-discrete actions with continuous parameters. At each
step the agent must select both which action to use and which parameters to use
with that action. We introduce the Q-PAMDP algorithm for learning in these
domains, show that it converges to a local optimum, and compare it to direct
policy search in the goal-scoring and Platform domains.
|
Gravitational Clustering | The downfall of many supervised learning algorithms, such as neural networks,
is the inherent need for a large amount of training data. Although there is a
lot of buzz about big data, there is still the problem of doing classification
from a small dataset. Other methods such as support vector machines, although
capable of dealing with few samples, are inherently binary classifiers, and are
in need of learning strategies such as One vs All in the case of
multi-classification. In the presence of a large number of classes this can
become problematic. In this paper we present, a novel approach to supervised
learning through the method of clustering. Unlike traditional methods such as
K-Means, Gravitational Clustering does not require the initial number of
clusters, and automatically builds the clusters, individual samples can be
arbitrarily weighted and it requires only few samples while staying resilient
to over-fitting.
|
HAMSI: A Parallel Incremental Optimization Algorithm Using Quadratic
Approximations for Solving Partially Separable Problems | We propose HAMSI (Hessian Approximated Multiple Subsets Iteration), which is
a provably convergent, second order incremental algorithm for solving
large-scale partially separable optimization problems. The algorithm is based
on a local quadratic approximation, and hence, allows incorporating curvature
information to speed-up the convergence. HAMSI is inherently parallel and it
scales nicely with the number of processors. Combined with techniques for
effectively utilizing modern parallel computer architectures, we illustrate
that the proposed method converges more rapidly than a parallel stochastic
gradient descent when both methods are used to solve large-scale matrix
factorization problems. This performance gain comes only at the expense of
using memory that scales linearly with the total size of the optimization
variables. We conclude that HAMSI may be considered as a viable alternative in
many large scale problems, where first order methods based on variants of
stochastic gradient descent are applicable.
|
Theoretic Analysis and Extremely Easy Algorithms for Domain Adaptive
Feature Learning | Domain adaptation problems arise in a variety of applications, where a
training dataset from the \textit{source} domain and a test dataset from the
\textit{target} domain typically follow different distributions. The primary
difficulty in designing effective learning models to solve such problems lies
in how to bridge the gap between the source and target distributions. In this
paper, we provide comprehensive analysis of feature learning algorithms used in
conjunction with linear classifiers for domain adaptation. Our analysis shows
that in order to achieve good adaptation performance, the second moments of the
source domain distribution and target domain distribution should be similar.
Based on our new analysis, a novel extremely easy feature learning algorithm
for domain adaptation is proposed. Furthermore, our algorithm is extended by
leveraging multiple layers, leading to a deep linear model. We evaluate the
effectiveness of the proposed algorithms in terms of domain adaptation tasks on
the Amazon review dataset and the spam dataset from the ECML/PKDD 2006
discovery challenge.
|
Theoretical and Experimental Analyses of Tensor-Based Regression and
Classification | We theoretically and experimentally investigate tensor-based regression and
classification. Our focus is regularization with various tensor norms,
including the overlapped trace norm, the latent trace norm, and the scaled
latent trace norm. We first give dual optimization methods using the
alternating direction method of multipliers, which is computationally efficient
when the number of training samples is moderate. We then theoretically derive
an excess risk bound for each tensor norm and clarify their behavior. Finally,
we perform extensive experiments using simulated and real data and demonstrate
the superiority of tensor-based learning methods over vector- and matrix-based
learning methods.
|
Sampled Weighted Min-Hashing for Large-Scale Topic Mining | We present Sampled Weighted Min-Hashing (SWMH), a randomized approach to
automatically mine topics from large-scale corpora. SWMH generates multiple
random partitions of the corpus vocabulary based on term co-occurrence and
agglomerates highly overlapping inter-partition cells to produce the mined
topics. While other approaches define a topic as a probabilistic distribution
over a vocabulary, SWMH topics are ordered subsets of such vocabulary.
Interestingly, the topics mined by SWMH underlie themes from the corpus at
different levels of granularity. We extensively evaluate the meaningfulness of
the mined topics both qualitatively and quantitatively on the NIPS (1.7 K
documents), 20 Newsgroups (20 K), Reuters (800 K) and Wikipedia (4 M) corpora.
Additionally, we compare the quality of SWMH with Online LDA topics for
document representation in classification.
|
Research: Analysis of Transport Model that Approximates Decision Taker's
Preferences | Paper provides a method for solving the reverse Monge-Kantorovich transport
problem (TP). It allows to accumulate positive decision-taking experience made
by decision-taker in situations that can be presented in the form of TP. The
initial data for the solution of the inverse TP is the information on orders,
inventories and effective decisions take by decision-taker. The result of
solving the inverse TP contains evaluations of the TPs payoff matrix elements.
It can be used in new situations to select the solution corresponding to the
preferences of the decision-taker. The method allows to gain decision-taker
experience, so it can be used by others. The method allows to build the model
of decision-taker preferences in a specific application area. The model can be
updated regularly to ensure its relevance and adequacy to the decision-taker
system of preferences. This model is adaptive to the current preferences of the
decision taker.
|
On collapsed representation of hierarchical Completely Random Measures | The aim of the paper is to provide an exact approach for generating a Poisson
process sampled from a hierarchical CRM, without having to instantiate the
infinitely many atoms of the random measures. We use completely random
measures~(CRM) and hierarchical CRM to define a prior for Poisson processes. We
derive the marginal distribution of the resultant point process, when the
underlying CRM is marginalized out. Using well known properties unique to
Poisson processes, we were able to derive an exact approach for instantiating a
Poisson process with a hierarchical CRM prior. Furthermore, we derive Gibbs
sampling strategies for hierarchical CRM models based on Chinese restaurant
franchise sampling scheme. As an example, we present the sum of generalized
gamma process (SGGP), and show its application in topic-modelling. We show that
one can determine the power-law behaviour of the topics and words in a Bayesian
fashion, by defining a prior on the parameters of SGGP.
|
Deep Online Convex Optimization by Putting Forecaster to Sleep | Methods from convex optimization such as accelerated gradient descent are
widely used as building blocks for deep learning algorithms. However, the
reasons for their empirical success are unclear, since neural networks are not
convex and standard guarantees do not apply. This paper develops the first
rigorous link between online convex optimization and error backpropagation on
convolutional networks. The first step is to introduce circadian games, a mild
generalization of convex games with similar convergence properties. The main
result is that error backpropagation on a convolutional network is equivalent
to playing out a circadian game. It follows immediately that the waking-regret
of players in the game (the units in the neural network) controls the overall
rate of convergence of the network. Finally, we explore some implications of
the results: (i) we describe the representations learned by a neural network
game-theoretically, (ii) propose a learning setting at the level of individual
units that can be plugged into deep architectures, and (iii) propose a new
approach to adaptive model selection by applying bandit algorithms to choose
which players to wake on each round.
|
Hierarchical Deep Learning Architecture For 10K Objects Classification | Evolution of visual object recognition architectures based on Convolutional
Neural Networks & Convolutional Deep Belief Networks paradigms has
revolutionized artificial Vision Science. These architectures extract & learn
the real world hierarchical visual features utilizing supervised & unsupervised
learning approaches respectively. Both the approaches yet cannot scale up
realistically to provide recognition for a very large number of objects as high
as 10K. We propose a two level hierarchical deep learning architecture inspired
by divide & conquer principle that decomposes the large scale recognition
architecture into root & leaf level model architectures. Each of the root &
leaf level models is trained exclusively to provide superior results than
possible by any 1-level deep learning architecture prevalent today. The
proposed architecture classifies objects in two steps. In the first step the
root level model classifies the object in a high level category. In the second
step, the leaf level recognition model for the recognized high level category
is selected among all the leaf models. This leaf level model is presented with
the same input object image which classifies it in a specific category. Also we
propose a blend of leaf level models trained with either supervised or
unsupervised learning approaches. Unsupervised learning is suitable whenever
labelled data is scarce for the specific leaf level models. Currently the
training of leaf level models is in progress; where we have trained 25 out of
the total 47 leaf level models as of now. We have trained the leaf models with
the best case top-5 error rate of 3.2% on the validation data set for the
particular leaf models. Also we demonstrate that the validation error of the
leaf level models saturates towards the above mentioned accuracy as the number
of epochs are increased to more than sixty.
|
Automated Analysis of Behavioural Variability and Filial Imprinting of
Chicks (G. gallus), using Autonomous Robots | Inter-individual variability has various impacts in animal social behaviour.
This implies that not only collective behaviours have to be studied but also
the behavioural variability of each member composing the groups. To understand
those effects on group behaviour, we develop a quantitative methodology based
on automated ethograms and autonomous robots to study the inter-individual
variability among social animals. We choose chicks of \textit{Gallus gallus
domesticus} as a classic social animal model system for their suitability in
laboratory and controlled experimentation. Moreover, even domesticated chicken
present social structures implying forms or leadership and filial imprinting.
We develop an imprinting methodology on autonomous robots to study individual
and social behaviour of free moving animals. This allows to quantify the
behaviours of large number of animals. We develop an automated experimental
methodology that allows to make relatively fast controlled experiments and
efficient data analysis. Our analysis are based on high-throughput data
allowing a fine quantification of individual behavioural traits. We quantify
the efficiency of various state-of-the-art algorithms to automate data analysis
and produce automated ethograms. We show that the use of robots allows to
provide controlled and quantified stimuli to the animals in absence of human
intervention. We quantify the individual behaviour of 205 chicks obtained from
hatching after synchronized fecundation. Our results show a high variability of
individual behaviours and of imprinting quality and success. Three classes of
chicks are observed with various level of imprinting. Our study shows that the
concomitant use of autonomous robots and automated ethograms allows detailed
and quantitative analysis of behavioural patterns of animals in controlled
laboratory experiments.
|
Data-selective Transfer Learning for Multi-Domain Speech Recognition | Negative transfer in training of acoustic models for automatic speech
recognition has been reported in several contexts such as domain change or
speaker characteristics. This paper proposes a novel technique to overcome
negative transfer by efficient selection of speech data for acoustic model
training. Here data is chosen on relevance for a specific target. A submodular
function based on likelihood ratios is used to determine how acoustically
similar each training utterance is to a target test set. The approach is
evaluated on a wide-domain data set, covering speech from radio and TV
broadcasts, telephone conversations, meetings, lectures and read speech.
Experiments demonstrate that the proposed technique both finds relevant data
and limits negative transfer. Results on a 6--hour test set show a relative
improvement of 4% with data selection over using all data in PLP based models,
and 2% with DNN features.
|
Improved Twitter Sentiment Prediction through Cluster-then-Predict Model | Over the past decade humans have experienced exponential growth in the use of
online resources, in particular social media and microblogging websites such as
Facebook, Twitter, YouTube and also mobile applications such as WhatsApp, Line,
etc. Many companies have identified these resources as a rich mine of marketing
knowledge. This knowledge provides valuable feedback which allows them to
further develop the next generation of their product. In this paper, sentiment
analysis of a product is performed by extracting tweets about that product and
classifying the tweets showing it as positive and negative sentiment. The
authors propose a hybrid approach which combines unsupervised learning in the
form of K-means clustering to cluster the tweets and then performing supervised
learning methods such as Decision Trees and Support Vector Machines for
classification.
|
A Behavior Analysis-Based Game Bot Detection Approach Considering
Various Play Styles | An approach for game bot detection in MMORPGs is proposed based on the
analysis of game playing behavior. Since MMORPGs are large scale games, users
can play in various ways. This variety in playing behavior makes it hard to
detect game bots based on play behaviors. In order to cope with this problem,
the proposed approach observes game playing behaviors of users and groups them
by their behavioral similarities. Then, it develops a local bot detection model
for each player group. Since the locally optimized models can more accurately
detect game bots within each player group, the combination of those models
brings about overall improvement. For a practical purpose of reducing the
workloads of the game servers in service, the game data is collected at a low
resolution in time. Behavioral features are selected and developed to
accurately detect game bots with the low resolution data, considering common
aspects of MMORPG playing. Through the experiment with the real data from a
game currently in service, it is shown that the proposed local model approach
yields more accurate results.
|
Deep Attributes from Context-Aware Regional Neural Codes | Recently, many researches employ middle-layer output of convolutional neural
network models (CNN) as features for different visual recognition tasks.
Although promising results have been achieved in some empirical studies, such
type of representations still suffer from the well-known issue of semantic gap.
This paper proposes so-called deep attribute framework to alleviate this issue
from three aspects. First, we introduce object region proposals as intermedia
to represent target images, and extract features from region proposals. Second,
we study aggregating features from different CNN layers for all region
proposals. The aggregation yields a holistic yet compact representation of
input images. Results show that cross-region max-pooling of soft-max layer
output outperform all other layers. As soft-max layer directly corresponds to
semantic concepts, this representation is named "deep attributes". Third, we
observe that only a small portion of generated regions by object proposals
algorithm are correlated to classification target. Therefore, we introduce
context-aware region refining algorithm to pick out contextual regions and
build context-aware classifiers.
We apply the proposed deep attributes framework for various vision tasks.
Extensive experiments are conducted on standard benchmarks for three visual
recognition tasks, i.e., image classification, fine-grained recognition and
visual instance retrieval. Results show that deep attribute approaches achieve
state-of-the-art results, and outperforms existing peer methods with a
significant margin, even though some benchmarks have little overlap of concepts
with the pre-trained CNN models.
|
Optimizing Static and Adaptive Probing Schedules for Rapid Event
Detection | We formulate and study a fundamental search and detection problem, Schedule
Optimization, motivated by a variety of real-world applications, ranging from
monitoring content changes on the web, social networks, and user activities to
detecting failure on large systems with many individual machines.
We consider a large system consists of many nodes, where each node has its
own rate of generating new events, or items. A monitoring application can probe
a small number of nodes at each step, and our goal is to compute a probing
schedule that minimizes the expected number of undiscovered items at the
system, or equivalently, minimizes the expected time to discover a new item in
the system.
We study the Schedule Optimization problem both for deterministic and
randomized memoryless algorithms. We provide lower bounds on the cost of an
optimal schedule and construct close to optimal schedules with rigorous
mathematical guarantees. Finally, we present an adaptive algorithm that starts
with no prior information on the system and converges to the optimal memoryless
algorithms by adapting to observed data.
|
DeepCough: A Deep Convolutional Neural Network in A Wearable Cough
Detection System | In this paper, we present a system that employs a wearable acoustic sensor
and a deep convolutional neural network for detecting coughs. We evaluate the
performance of our system on 14 healthy volunteers and compare it to that of
other cough detection systems that have been reported in the literature.
Experimental results show that our system achieves a classification sensitivity
of 95.1% and a specificity of 99.5%.
|
Asynchronous Distributed ADMM for Large-Scale Optimization- Part I:
Algorithm and Convergence Analysis | Aiming at solving large-scale learning problems, this paper studies
distributed optimization methods based on the alternating direction method of
multipliers (ADMM). By formulating the learning problem as a consensus problem,
the ADMM can be used to solve the consensus problem in a fully parallel fashion
over a computer network with a star topology. However, traditional synchronized
computation does not scale well with the problem size, as the speed of the
algorithm is limited by the slowest workers. This is particularly true in a
heterogeneous network where the computing nodes experience different
computation and communication delays. In this paper, we propose an asynchronous
distributed ADMM (AD-AMM) which can effectively improve the time efficiency of
distributed optimization. Our main interest lies in analyzing the convergence
conditions of the AD-ADMM, under the popular partially asynchronous model,
which is defined based on a maximum tolerable delay of the network.
Specifically, by considering general and possibly non-convex cost functions, we
show that the AD-ADMM is guaranteed to converge to the set of
Karush-Kuhn-Tucker (KKT) points as long as the algorithm parameters are chosen
appropriately according to the network delay. We further illustrate that the
asynchrony of the ADMM has to be handled with care, as slightly modifying the
implementation of the AD-ADMM can jeopardize the algorithm convergence, even
under a standard convex setting.
|
Asynchronous Distributed ADMM for Large-Scale Optimization- Part II:
Linear Convergence Analysis and Numerical Performance | The alternating direction method of multipliers (ADMM) has been recognized as
a versatile approach for solving modern large-scale machine learning and signal
processing problems efficiently. When the data size and/or the problem
dimension is large, a distributed version of ADMM can be used, which is capable
of distributing the computation load and the data set to a network of computing
nodes. Unfortunately, a direct synchronous implementation of such algorithm
does not scale well with the problem size, as the algorithm speed is limited by
the slowest computing nodes. To address this issue, in a companion paper, we
have proposed an asynchronous distributed ADMM (AD-ADMM) and studied its
worst-case convergence conditions. In this paper, we further the study by
characterizing the conditions under which the AD-ADMM achieves linear
convergence. Our conditions as well as the resulting linear rates reveal the
impact that various algorithm parameters, network delay and network size have
on the algorithm performance. To demonstrate the superior time efficiency of
the proposed AD-ADMM, we test the AD-ADMM on a high-performance computer
cluster by solving a large-scale logistic regression problem.
|
Finite Dictionary Variants of the Diffusion KLMS Algorithm | The diffusion based distributed learning approaches have been found to be a
viable solution for learning over linearly separable datasets over a network.
However, approaches till date are suitable for linearly separable datasets and
need to be extended to scenarios in which we need to learn a non-linearity. In
such scenarios, the recently proposed diffusion kernel least mean squares
(KLMS) has been found to be performing better than diffusion least mean squares
(LMS). The drawback of diffusion KLMS is that it requires infinite storage for
observations (also called dictionary). This paper formulates the diffusion KLMS
in a fixed budget setting such that the storage requirement is curtailed while
maintaining appreciable performance in terms of convergence. Simulations have
been carried out to validate the two newly proposed algorithms named as
quantised diffusion KLMS (QDKLMS) and fixed budget diffusion KLMS (FBDKLMS)
against KLMS, which indicate that both the proposed algorithms deliver better
performance as compared to the KLMS while reducing the dictionary size storage
requirement.
|
Clustering by Hierarchical Nearest Neighbor Descent (H-NND) | Previously in 2014, we proposed the Nearest Descent (ND) method, capable of
generating an efficient Graph, called the in-tree (IT). Due to some beautiful
and effective features, this IT structure proves well suited for data
clustering. Although there exist some redundant edges in IT, they usually have
salient features and thus it is not hard to remove them.
Subsequently, in order to prevent the seemingly redundant edges from
occurring, we proposed the Nearest Neighbor Descent (NND) by adding the
"Neighborhood" constraint on ND. Consequently, clusters automatically emerged,
without the additional requirement of removing the redundant edges. However,
NND proved still not perfect, since it brought in a new yet worse problem, the
"over-partitioning" problem.
Now, in this paper, we propose a method, called the Hierarchical Nearest
Neighbor Descent (H-NND), which overcomes the over-partitioning problem of NND
via using the hierarchical strategy. Specifically, H-NND uses ND to effectively
merge the over-segmented sub-graphs or clusters that NND produces. Like ND,
H-NND also generates the IT structure, in which the redundant edges once again
appear. This seemingly comes back to the situation that ND faces. However,
compared with ND, the redundant edges in the IT structure generated by H-NND
generally become more salient, thus being much easier and more reliable to be
identified even by the simplest edge-removing method which takes the edge
length as the only measure. In other words, the IT structure constructed by
H-NND becomes more fitted for data clustering. We prove this on several
clustering datasets of varying shapes, dimensions and attributes. Besides,
compared with ND, H-NND generally takes less computation time to construct the
IT data structure for the input data.
|
Statistical Inference, Learning and Models in Big Data | The need for new methods to deal with big data is a common theme in most
scientific fields, although its definition tends to vary with the context.
Statistical ideas are an essential part of this, and as a partial response, a
thematic program on statistical inference, learning, and models in big data was
held in 2015 in Canada, under the general direction of the Canadian Statistical
Sciences Institute, with major funding from, and most activities located at,
the Fields Institute for Research in Mathematical Sciences. This paper gives an
overview of the topics covered, describing challenges and strategies that seem
common to many different areas of application, and including some examples of
applications to make these challenges and strategies more concrete.
|
Sensor Selection by Linear Programming | We learn sensor trees from training data to minimize sensor acquisition costs
during test time. Our system adaptively selects sensors at each stage if
necessary to make a confident classification. We pose the problem as empirical
risk minimization over the choice of trees and node decision rules. We
decompose the problem, which is known to be intractable, into combinatorial
(tree structures) and continuous parts (node decision rules) and propose to
solve them separately. Using training data we greedily solve for the
combinatorial tree structures and for the continuous part, which is a
non-convex multilinear objective function, we derive convex surrogate loss
functions that are piecewise linear. The resulting problem can be cast as a
linear program and has the advantage of guaranteed convergence, global
optimality, repeatability and computational efficiency. We show that our
proposed approach outperforms the state-of-art on a number of benchmark
datasets.
|
Continuous control with deep reinforcement learning | We adapt the ideas underlying the success of Deep Q-Learning to the
continuous action domain. We present an actor-critic, model-free algorithm
based on the deterministic policy gradient that can operate over continuous
action spaces. Using the same learning algorithm, network architecture and
hyper-parameters, our algorithm robustly solves more than 20 simulated physics
tasks, including classic problems such as cartpole swing-up, dexterous
manipulation, legged locomotion and car driving. Our algorithm is able to find
policies whose performance is competitive with those found by a planning
algorithm with full access to the dynamics of the domain and its derivatives.
We further demonstrate that for many of the tasks the algorithm can learn
policies end-to-end: directly from raw pixel inputs.
|
Compatible Value Gradients for Reinforcement Learning of Continuous Deep
Policies | This paper proposes GProp, a deep reinforcement learning algorithm for
continuous policies with compatible function approximation. The algorithm is
based on two innovations. Firstly, we present a temporal-difference based
method for learning the gradient of the value-function. Secondly, we present
the deviator-actor-critic (DAC) model, which comprises three neural networks
that estimate the value function, its gradient, and determine the actor's
policy respectively. We evaluate GProp on two challenging tasks: a contextual
bandit problem constructed from nonparametric regression datasets that is
designed to probe the ability of reinforcement learning algorithms to
accurately estimate gradients; and the octopus arm, a challenging reinforcement
learning benchmark. GProp is competitive with fully supervised methods on the
bandit task and achieves the best performance to date on the octopus arm.
|
Fast low-rank estimation by projected gradient descent: General
statistical and algorithmic guarantees | Optimization problems with rank constraints arise in many applications,
including matrix regression, structured PCA, matrix completion and matrix
decomposition problems. An attractive heuristic for solving such problems is to
factorize the low-rank matrix, and to run projected gradient descent on the
nonconvex factorized optimization problem. The goal of this problem is to
provide a general theoretical framework for understanding when such methods
work well, and to characterize the nature of the resulting fixed point. We
provide a simple set of conditions under which projected gradient descent, when
given a suitable initialization, converges geometrically to a statistically
useful solution. Our results are applicable even when the initial solution is
outside any region of local convexity, and even when the problem is globally
concave. Working in a non-asymptotic framework, we show that our conditions are
satisfied for a wide range of concrete models, including matrix regression,
structured PCA, matrix completion with real and quantized observations, matrix
decomposition, and graph clustering problems. Simulation results show excellent
agreement with the theoretical predictions.
|
Recurrent Reinforcement Learning: A Hybrid Approach | Successful applications of reinforcement learning in real-world problems
often require dealing with partially observable states. It is in general very
challenging to construct and infer hidden states as they often depend on the
agent's entire interaction history and may require substantial domain
knowledge. In this work, we investigate a deep-learning approach to learning
the representation of states in partially observable tasks, with minimal prior
knowledge of the domain. In particular, we propose a new family of hybrid
models that combines the strength of both supervised learning (SL) and
reinforcement learning (RL), trained in a joint fashion: The SL component can
be a recurrent neural networks (RNN) or its long short-term memory (LSTM)
version, which is equipped with the desired property of being able to capture
long-term dependency on history, thus providing an effective way of learning
the representation of hidden states. The RL component is a deep Q-network (DQN)
that learns to optimize the control for maximizing long-term rewards. Extensive
experiments in a direct mailing campaign problem demonstrate the effectiveness
and advantages of the proposed approach, which performs the best among a set of
previous state-of-the-art methods.
|
Use it or Lose it: Selective Memory and Forgetting in a Perpetual
Learning Machine | In a recent article we described a new type of deep neural network - a
Perpetual Learning Machine (PLM) - which is capable of learning 'on the fly'
like a brain by existing in a state of Perpetual Stochastic Gradient Descent
(PSGD). Here, by simulating the process of practice, we demonstrate both
selective memory and selective forgetting when we introduce statistical recall
biases during PSGD. Frequently recalled memories are remembered, whilst
memories recalled rarely are forgotten. This results in a 'use it or lose it'
stimulus driven memory process that is similar to human memory.
|
A new Initial Centroid finding Method based on Dissimilarity Tree for
K-means Algorithm | Cluster analysis is one of the primary data analysis technique in data mining
and K-means is one of the commonly used partitioning clustering algorithm. In
K-means algorithm, resulting set of clusters depend on the choice of initial
centroids. If we can find initial centroids which are coherent with the
arrangement of data, the better set of clusters can be obtained. This paper
proposes a method based on the Dissimilarity Tree to find, the better initial
centroid as well as every bit more accurate cluster with less computational
time. Theory analysis and experimental results indicate that the proposed
method can effectively improve the accuracy of clusters and reduce the
computational complexity of the K-means algorithm.
|
Gibbs Sampling Strategies for Semantic Perception of Streaming Video
Data | Topic modeling of streaming sensor data can be used for high level perception
of the environment by a mobile robot. In this paper we compare various Gibbs
sampling strategies for topic modeling of streaming spatiotemporal data, such
as video captured by a mobile robot. Compared to previous work on online topic
modeling, such as o-LDA and incremental LDA, we show that the proposed
technique results in lower online and final perplexity, given the realtime
constraints.
|
A deep matrix factorization method for learning attribute
representations | Semi-Non-negative Matrix Factorization is a technique that learns a
low-dimensional representation of a dataset that lends itself to a clustering
interpretation. It is possible that the mapping between this new representation
and our original data matrix contains rather complex hierarchical information
with implicit lower-level hidden attributes, that classical one level
clustering methodologies can not interpret. In this work we propose a novel
model, Deep Semi-NMF, that is able to learn such hidden representations that
allow themselves to an interpretation of clustering according to different,
unknown attributes of a given dataset. We also present a semi-supervised
version of the algorithm, named Deep WSF, that allows the use of (partial)
prior information for each of the known attributes of a dataset, that allows
the model to be used on datasets with mixed attribute knowledge. Finally, we
show that our models are able to learn low-dimensional representations that are
better suited for clustering, but also classification, outperforming
Semi-Non-negative Matrix Factorization, but also other state-of-the-art
methodologies variants.
|
Performance Bounds for Pairwise Entity Resolution | One significant challenge to scaling entity resolution algorithms to massive
datasets is understanding how performance changes after moving beyond the realm
of small, manually labeled reference datasets. Unlike traditional machine
learning tasks, when an entity resolution algorithm performs well on small
hold-out datasets, there is no guarantee this performance holds on larger
hold-out datasets. We prove simple bounding properties between the performance
of a match function on a small validation set and the performance of a pairwise
entity resolution algorithm on arbitrarily sized datasets. Thus, our approach
enables optimization of pairwise entity resolution algorithms for large
datasets, using a small set of labeled data.
|
Hessian-free Optimization for Learning Deep Multidimensional Recurrent
Neural Networks | Multidimensional recurrent neural networks (MDRNNs) have shown a remarkable
performance in the area of speech and handwriting recognition. The performance
of an MDRNN is improved by further increasing its depth, and the difficulty of
learning the deeper network is overcome by using Hessian-free (HF)
optimization. Given that connectionist temporal classification (CTC) is
utilized as an objective of learning an MDRNN for sequence labeling, the
non-convexity of CTC poses a problem when applying HF to the network. As a
solution, a convex approximation of CTC is formulated and its relationship with
the EM algorithm and the Fisher information matrix is discussed. An MDRNN up to
a depth of 15 layers is successfully trained using HF, resulting in an improved
performance for sequence labeling.
|
Hardness of Online Sleeping Combinatorial Optimization Problems | We show that several online combinatorial optimization problems that admit
efficient no-regret algorithms become computationally hard in the sleeping
setting where a subset of actions becomes unavailable in each round.
Specifically, we show that the sleeping versions of these problems are at least
as hard as PAC learning DNF expressions, a long standing open problem. We show
hardness for the sleeping versions of Online Shortest Paths, Online Minimum
Spanning Tree, Online $k$-Subsets, Online $k$-Truncated Permutations, Online
Minimum Cut, and Online Bipartite Matching. The hardness result for the
sleeping version of the Online Shortest Paths problem resolves an open problem
presented at COLT 2015 (Koolen et al., 2015).
|
Toward better feature weighting algorithms: a focus on Relief | Feature weighting algorithms try to solve a problem of great importance
nowadays in machine learning: The search of a relevance measure for the
features of a given domain. This relevance is primarily used for feature
selection as feature weighting can be seen as a generalization of it, but it is
also useful to better understand a problem's domain or to guide an inductor in
its learning process. Relief family of algorithms are proven to be very
effective in this task. Some other feature weighting methods are reviewed in
order to give some context and then the different existing extensions to the
original algorithm are explained.
One of Relief's known issues is the performance degradation of its estimates
when redundant features are present. A novel theoretical definition of
redundancy level is given in order to guide the work towards an extension of
the algorithm that is more robust against redundancy. A new extension is
presented that aims for improving the algorithms performance. Some experiments
were driven to test this new extension against the existing ones with a set of
artificial and real datasets and denoted that in certain cases it improves the
weight's estimation accuracy.
|
Dropping Convexity for Faster Semi-definite Optimization | We study the minimization of a convex function $f(X)$ over the set of
$n\times n$ positive semi-definite matrices, but when the problem is recast as
$\min_U g(U) := f(UU^\top)$, with $U \in \mathbb{R}^{n \times r}$ and $r \leq
n$. We study the performance of gradient descent on $g$---which we refer to as
Factored Gradient Descent (FGD)---under standard assumptions on the original
function $f$.
We provide a rule for selecting the step size and, with this choice, show
that the local convergence rate of FGD mirrors that of standard gradient
descent on the original $f$: i.e., after $k$ steps, the error is $O(1/k)$ for
smooth $f$, and exponentially small in $k$ when $f$ is (restricted) strongly
convex. In addition, we provide a procedure to initialize FGD for (restricted)
strongly convex objectives and when one only has access to $f$ via a
first-order oracle; for several problem instances, such proper initialization
leads to global convergence guarantees.
FGD and similar procedures are widely used in practice for problems that can
be posed as matrix factorization. To the best of our knowledge, this is the
first paper to provide precise convergence rate guarantees for general convex
functions under standard convex assumptions.
|
Parametric Maxflows for Structured Sparse Learning with Convex
Relaxations of Submodular Functions | The proximal problem for structured penalties obtained via convex relaxations
of submodular functions is known to be equivalent to minimizing separable
convex functions over the corresponding submodular polyhedra. In this paper, we
reveal a comprehensive class of structured penalties for which penalties this
problem can be solved via an efficiently solvable class of parametric maxflow
optimization. We then show that the parametric maxflow algorithm proposed by
Gallo et al. and its variants, which runs, in the worst-case, at the cost of
only a constant factor of a single computation of the corresponding maxflow
optimization, can be adapted to solve the proximal problems for those
penalties. Several existing structured penalties satisfy these conditions;
thus, regularized learning with these penalties is solvable quickly using the
parametric maxflow algorithm. We also investigate the empirical runtime
performance of the proposed framework.
|
Optimization of anemia treatment in hemodialysis patients via
reinforcement learning | Objective: Anemia is a frequent comorbidity in hemodialysis patients that can
be successfully treated by administering erythropoiesis-stimulating agents
(ESAs). ESAs dosing is currently based on clinical protocols that often do not
account for the high inter- and intra-individual variability in the patient's
response. As a result, the hemoglobin level of some patients oscillates around
the target range, which is associated with multiple risks and side-effects.
This work proposes a methodology based on reinforcement learning (RL) to
optimize ESA therapy.
Methods: RL is a data-driven approach for solving sequential decision-making
problems that are formulated as Markov decision processes (MDPs). Computing
optimal drug administration strategies for chronic diseases is a sequential
decision-making problem in which the goal is to find the best sequence of drug
doses. MDPs are particularly suitable for modeling these problems due to their
ability to capture the uncertainty associated with the outcome of the treatment
and the stochastic nature of the underlying process. The RL algorithm employed
in the proposed methodology is fitted Q iteration, which stands out for its
ability to make an efficient use of data.
Results: The experiments reported here are based on a computational model
that describes the effect of ESAs on the hemoglobin level. The performance of
the proposed method is evaluated and compared with the well-known Q-learning
algorithm and with a standard protocol. Simulation results show that the
performance of Q-learning is substantially lower than FQI and the protocol.
Conclusion: Although prospective validation is required, promising results
demonstrate the potential of RL to become an alternative to current protocols.
|
Fame for sale: efficient detection of fake Twitter followers | $\textit{Fake followers}$ are those Twitter accounts specifically created to
inflate the number of followers of a target account. Fake followers are
dangerous for the social platform and beyond, since they may alter concepts
like popularity and influence in the Twittersphere - hence impacting on
economy, politics, and society. In this paper, we contribute along different
dimensions. First, we review some of the most relevant existing features and
rules (proposed by Academia and Media) for anomalous Twitter accounts
detection. Second, we create a baseline dataset of verified human and fake
follower accounts. Such baseline dataset is publicly available to the
scientific community. Then, we exploit the baseline dataset to train a set of
machine-learning classifiers built over the reviewed rules and features. Our
results show that most of the rules proposed by Media provide unsatisfactory
performance in revealing fake followers, while features proposed in the past by
Academia for spam detection provide good results. Building on the most
promising features, we revise the classifiers both in terms of reduction of
overfitting and cost for gathering the data needed to compute the features. The
final result is a novel $\textit{Class A}$ classifier, general enough to thwart
overfitting, lightweight thanks to the usage of the less costly features, and
still able to correctly classify more than 95% of the accounts of the original
training set. We ultimately perform an information fusion-based sensitivity
analysis, to assess the global sensitivity of each of the features employed by
the classifier. The findings reported in this paper, other than being supported
by a thorough experimental methodology and interesting on their own, also pave
the way for further investigation on the novel issue of fake Twitter followers.
|
Model Accuracy and Runtime Tradeoff in Distributed Deep Learning:A
Systematic Study | This paper presents Rudra, a parameter server based distributed computing
framework tuned for training large-scale deep neural networks. Using variants
of the asynchronous stochastic gradient descent algorithm we study the impact
of synchronization protocol, stale gradient updates, minibatch size, learning
rates, and number of learners on runtime performance and model accuracy. We
introduce a new learning rate modulation strategy to counter the effect of
stale gradients and propose a new synchronization protocol that can effectively
bound the staleness in gradients, improve runtime performance and achieve good
model accuracy. Our empirical investigation reveals a principled approach for
distributed training of neural networks: the mini-batch size per learner should
be reduced as more learners are added to the system to preserve the model
accuracy. We validate this approach using commonly-used image classification
benchmarks: CIFAR10 and ImageNet.
|
Double Relief with progressive weighting function | Feature weighting algorithms try to solve a problem of great importance
nowadays in machine learning: The search of a relevance measure for the
features of a given domain. This relevance is primarily used for feature
selection as feature weighting can be seen as a generalization of it, but it is
also useful to better understand a problem's domain or to guide an inductor in
its learning process. Relief family of algorithms are proven to be very
effective in this task.
On previous work, a new extension was proposed that aimed for improving the
algorithm's performance and it was shown that in certain cases it improved the
weights' estimation accuracy. However, it also seemed to be sensible to some
characteristics of the data. An improvement of that previously presented
extension is presented in this work that aims to make it more robust to problem
specific characteristics. An experimental design is proposed to test its
performance. Results of the tests prove that it indeed increase the robustness
of the previously proposed extension.
|
Voted Kernel Regularization | This paper presents an algorithm, Voted Kernel Regularization , that provides
the flexibility of using potentially very complex kernel functions such as
predictors based on much higher-degree polynomial kernels, while benefitting
from strong learning guarantees. The success of our algorithm arises from
derived bounds that suggest a new regularization penalty in terms of the
Rademacher complexities of the corresponding families of kernel maps. In a
series of experiments we demonstrate the improved performance of our algorithm
as compared to baselines. Furthermore, the algorithm enjoys several favorable
properties. The optimization problem is convex, it allows for learning with
non-PDS kernels, and the solutions are highly sparse, resulting in improved
classification speed and memory requirements.
|
Towards Making High Dimensional Distance Metric Learning Practical | In this work, we study distance metric learning (DML) for high dimensional
data. A typical approach for DML with high dimensional data is to perform the
dimensionality reduction first before learning the distance metric. The main
shortcoming of this approach is that it may result in a suboptimal solution due
to the subspace removed by the dimensionality reduction method. In this work,
we present a dual random projection frame for DML with high dimensional data
that explicitly addresses the limitation of dimensionality reduction for DML.
The key idea is to first project all the data points into a low dimensional
space by random projection, and compute the dual variables using the projected
vectors. It then reconstructs the distance metric in the original space using
the estimated dual variables. The proposed method, on one hand, enjoys the
light computation of random projection, and on the other hand, alleviates the
limitation of most dimensionality reduction methods. We verify both empirically
and theoretically the effectiveness of the proposed algorithm for high
dimensional DML.
|
Precise Phase Transition of Total Variation Minimization | Characterizing the phase transitions of convex optimizations in recovering
structured signals or data is of central importance in compressed sensing,
machine learning and statistics. The phase transitions of many convex
optimization signal recovery methods such as $\ell_1$ minimization and nuclear
norm minimization are well understood through recent years' research. However,
rigorously characterizing the phase transition of total variation (TV)
minimization in recovering sparse-gradient signal is still open. In this paper,
we fully characterize the phase transition curve of the TV minimization. Our
proof builds on Donoho, Johnstone and Montanari's conjectured phase transition
curve for the TV approximate message passing algorithm (AMP), together with the
linkage between the minmax Mean Square Error of a denoising problem and the
high-dimensional convex geometry for TV minimization.
|
Exponential Family Matrix Completion under Structural Constraints | We consider the matrix completion problem of recovering a structured matrix
from noisy and partial measurements. Recent works have proposed tractable
estimators with strong statistical guarantees for the case where the underlying
matrix is low--rank, and the measurements consist of a subset, either of the
exact individual entries, or of the entries perturbed by additive Gaussian
noise, which is thus implicitly suited for thin--tailed continuous data.
Arguably, common applications of matrix completion require estimators for (a)
heterogeneous data--types, such as skewed--continuous, count, binary, etc., (b)
for heterogeneous noise models (beyond Gaussian), which capture varied
uncertainty in the measurements, and (c) heterogeneous structural constraints
beyond low--rank, such as block--sparsity, or a superposition structure of
low--rank plus elementwise sparseness, among others. In this paper, we provide
a vastly unified framework for generalized matrix completion by considering a
matrix completion setting wherein the matrix entries are sampled from any
member of the rich family of exponential family distributions; and impose
general structural constraints on the underlying matrix, as captured by a
general regularizer $\mathcal{R}(.)$. We propose a simple convex regularized
$M$--estimator for the generalized framework, and provide a unified and novel
statistical analysis for this general class of estimators. We finally
corroborate our theoretical results on simulated datasets.
|
Adapting Resilient Propagation for Deep Learning | The Resilient Propagation (Rprop) algorithm has been very popular for
backpropagation training of multilayer feed-forward neural networks in various
applications. The standard Rprop however encounters difficulties in the context
of deep neural networks as typically happens with gradient-based learning
algorithms. In this paper, we propose a modification of the Rprop that combines
standard Rprop steps with a special drop out technique. We apply the method for
training Deep Neural Networks as standalone components and in ensemble
formulations. Results on the MNIST dataset show that the proposed modification
alleviates standard Rprop's problems demonstrating improved learning speed and
accuracy.
|
Dynamic Poisson Factorization | Models for recommender systems use latent factors to explain the preferences
and behaviors of users with respect to a set of items (e.g., movies, books,
academic papers). Typically, the latent factors are assumed to be static and,
given these factors, the observed preferences and behaviors of users are
assumed to be generated without order. These assumptions limit the explorative
and predictive capabilities of such models, since users' interests and item
popularity may evolve over time. To address this, we propose dPF, a dynamic
matrix factorization model based on the recent Poisson factorization model for
recommendations. dPF models the time evolving latent factors with a Kalman
filter and the actions with Poisson distributions. We derive a scalable
variational inference algorithm to infer the latent factors. Finally, we
demonstrate dPF on 10 years of user click data from arXiv.org, one of the
largest repository of scientific papers and a formidable source of information
about the behavior of scientists. Empirically we show performance improvement
over both static and, more recently proposed, dynamic recommendation models. We
also provide a thorough exploration of the inferred posteriors over the latent
variables.
|
Forecasting Method for Grouped Time Series with the Use of k-Means
Algorithm | The paper is focused on the forecasting method for time series groups with
the use of algorithms for cluster analysis. $K$-means algorithm is suggested to
be a basic one for clustering. The coordinates of the centers of clusters have
been put in correspondence with summarizing time series data the centroids of
the clusters. A description of time series, the centroids of the clusters, is
implemented with the use of forecasting models. They are based on strict binary
trees and a modified clonal selection algorithm. With the help of such
forecasting models, the possibility of forming analytic dependences is shown.
It is suggested to use a common forecasting model, which is constructed for
time series the centroid of the cluster, in forecasting the private
(individual) time series in the cluster. The promising application of the
suggested method for grouped time series forecasting is demonstrated.
|
On the Expressive Power of Deep Learning: A Tensor Analysis | It has long been conjectured that hypotheses spaces suitable for data that is
compositional in nature, such as text or images, may be more efficiently
represented with deep hierarchical networks than with shallow ones. Despite the
vast empirical evidence supporting this belief, theoretical justifications to
date are limited. In particular, they do not account for the locality, sharing
and pooling constructs of convolutional networks, the most successful deep
learning architecture to date. In this work we derive a deep network
architecture based on arithmetic circuits that inherently employs locality,
sharing and pooling. An equivalence between the networks and hierarchical
tensor factorizations is established. We show that a shallow network
corresponds to CP (rank-1) decomposition, whereas a deep network corresponds to
Hierarchical Tucker decomposition. Using tools from measure theory and matrix
algebra, we prove that besides a negligible set, all functions that can be
implemented by a deep network of polynomial size, require exponential size in
order to be realized (or even approximated) by a shallow network. Since
log-space computation transforms our networks into SimNets, the result applies
directly to a deep learning architecture demonstrating promising empirical
performance. The construction and theory developed in this paper shed new light
on various practices and ideas employed by the deep learning community.
|
Fast Sequence Component Analysis for Attack Detection in Synchrophasor
Networks | Modern power systems have begun integrating synchrophasor technologies into
part of daily operations. Given the amount of solutions offered and the
maturity rate of application development it is not a matter of "if" but a
matter of "when" in regards to these technologies becoming ubiquitous in
control centers around the world. While the benefits are numerous, the
functionality of operator-level applications can easily be nullified by
injection of deceptive data signals disguised as genuine measurements. Such
deceptive action is a common precursor to nefarious, often malicious activity.
A correlation coefficient characterization and machine learning methodology are
proposed to detect and identify injection of spoofed data signals. The proposed
method utilizes statistical relationships intrinsic to power system parameters,
which are quantified and presented. Several spoofing schemes have been
developed to qualitatively and quantitatively demonstrate detection
capabilities.
|
Revealed Preference at Scale: Learning Personalized Preferences from
Assortment Choices | We consider the problem of learning the preferences of a heterogeneous
population by observing choices from an assortment of products, ads, or other
offerings. Our observation model takes a form common in assortment planning
applications: each arriving customer is offered an assortment consisting of a
subset of all possible offerings; we observe only the assortment and the
customer's single choice.
In this paper we propose a mixture choice model with a natural underlying
low-dimensional structure, and show how to estimate its parameters. In our
model, the preferences of each customer or segment follow a separate parametric
choice model, but the underlying structure of these parameters over all the
models has low dimension. We show that a nuclear-norm regularized maximum
likelihood estimator can learn the preferences of all customers using a number
of observations much smaller than the number of item-customer combinations.
This result shows the potential for structural assumptions to speed up learning
and improve revenues in assortment planning and customization. We provide a
specialized factored gradient descent algorithm and study the success of the
approach empirically.
|
Fast Gaussian Process Regression for Big Data | Gaussian Processes are widely used for regression tasks. A known limitation
in the application of Gaussian Processes to regression tasks is that the
computation of the solution requires performing a matrix inversion. The
solution also requires the storage of a large matrix in memory. These factors
restrict the application of Gaussian Process regression to small and moderate
size data sets. We present an algorithm that combines estimates from models
developed using subsets of the data obtained in a manner similar to the
bootstrap. The sample size is a critical parameter for this algorithm.
Guidelines for reasonable choices of algorithm parameters, based on detailed
experimental study, are provided. Various techniques have been proposed to
scale Gaussian Processes to large scale regression tasks. The most appropriate
choice depends on the problem context. The proposed method is most appropriate
for problems where an additive model works well and the response depends on a
small number of features. The minimax rate of convergence for such problems is
attractive and we can build effective models with a small subset of the data.
The Stochastic Variational Gaussian Process and the Sparse Gaussian Process are
also appropriate choices for such problems. These methods pick a subset of data
based on theoretical considerations. The proposed algorithm uses bagging and
random sampling. Results from experiments conducted as part of this study
indicate that the algorithm presented in this work can be as effective as these
methods. Model stacking can be used to combine the model developed with the
proposed method with models from other methods for large scale regression such
as Gradient Boosted Trees. This can yield performance gains.
|
Generalized Emphatic Temporal Difference Learning: Bias-Variance
Analysis | We consider the off-policy evaluation problem in Markov decision processes
with function approximation. We propose a generalization of the recently
introduced \emph{emphatic temporal differences} (ETD) algorithm
\citep{SuttonMW15}, which encompasses the original ETD($\lambda$), as well as
several other off-policy evaluation algorithms as special cases. We call this
framework \ETD, where our introduced parameter $\beta$ controls the decay rate
of an importance-sampling term. We study conditions under which the projected
fixed-point equation underlying \ETD\ involves a contraction operator, allowing
us to present the first asymptotic error bounds (bias) for \ETD. Our results
show that the original ETD algorithm always involves a contraction operator,
and its bias is bounded. Moreover, by controlling $\beta$, our proposed
generalization allows trading-off bias for variance reduction, thereby
achieving a lower total error.
|
Taming the ReLU with Parallel Dither in a Deep Neural Network | Rectified Linear Units (ReLU) seem to have displaced traditional 'smooth'
nonlinearities as activation-function-du-jour in many - but not all - deep
neural network (DNN) applications. However, nobody seems to know why. In this
article, we argue that ReLU are useful because they are ideal demodulators -
this helps them perform fast abstract learning. However, this fast learning
comes at the expense of serious nonlinear distortion products - decoy features.
We show that Parallel Dither acts to suppress the decoy features, preventing
overfitting and leaving the true features cleanly demodulated for rapid,
reliable learning.
|
(Blue) Taxi Destination and Trip Time Prediction from Partial
Trajectories | Real-time estimation of destination and travel time for taxis is of great
importance for existing electronic dispatch systems. We present an approach
based on trip matching and ensemble learning, in which we leverage the patterns
observed in a dataset of roughly 1.7 million taxi journeys to predict the
corresponding final destination and travel time for ongoing taxi trips, as a
solution for the ECML/PKDD Discovery Challenge 2015 competition. The results of
our empirical evaluation show that our approach is effective and very robust,
which led our team -- BlueTaxi -- to the 3rd and 7th position of the final
rankings for the trip time and destination prediction tasks, respectively.
Given the fact that the final rankings were computed using a very small test
set (with only 320 trips) we believe that our approach is one of the most
robust solutions for the challenge based on the consistency of our good results
across the test sets.
|
DeXpression: Deep Convolutional Neural Network for Expression
Recognition | We propose a convolutional neural network (CNN) architecture for facial
expression recognition. The proposed architecture is independent of any
hand-crafted feature extraction and performs better than the earlier proposed
convolutional neural network based approaches. We visualize the automatically
extracted features which have been learned by the network in order to provide a
better understanding. The standard datasets, i.e. Extended Cohn-Kanade (CKP)
and MMI Facial Expression Databse are used for the quantitative evaluation. On
the CKP set the current state of the art approach, using CNNs, achieves an
accuracy of 99.2%. For the MMI dataset, currently the best accuracy for emotion
recognition is 93.33%. The proposed architecture achieves 99.6% for CKP and
98.63% for MMI, therefore performing better than the state of the art using
CNNs. Automatic facial expression recognition has a broad spectrum of
applications such as human-computer interaction and safety systems. This is due
to the fact that non-verbal cues are important forms of communication and play
a pivotal role in interpersonal communication. The performance of the proposed
architecture endorses the efficacy and reliable usage of the proposed work for
real world applications.
|
Learning to Hash for Indexing Big Data - A Survey | The explosive growth in big data has attracted much attention in designing
efficient indexing and search methods recently. In many critical applications
such as large-scale search and pattern matching, finding the nearest neighbors
to a query is a fundamental research problem. However, the straightforward
solution using exhaustive comparison is infeasible due to the prohibitive
computational complexity and memory requirement. In response, Approximate
Nearest Neighbor (ANN) search based on hashing techniques has become popular
due to its promising performance in both efficiency and accuracy. Prior
randomized hashing methods, e.g., Locality-Sensitive Hashing (LSH), explore
data-independent hash functions with random projections or permutations.
Although having elegant theoretic guarantees on the search quality in certain
metric spaces, performance of randomized hashing has been shown insufficient in
many real-world applications. As a remedy, new approaches incorporating
data-driven learning methods in development of advanced hash functions have
emerged. Such learning to hash methods exploit information such as data
distributions or class labels when optimizing the hash codes or functions.
Importantly, the learned hash codes are able to preserve the proximity of
neighboring data in the original feature spaces in the hash code spaces. The
goal of this paper is to provide readers with systematic understanding of
insights, pros and cons of the emerging techniques. We provide a comprehensive
survey of the learning to hash framework and representative techniques of
various types, including unsupervised, semi-supervised, and supervised. In
addition, we also summarize recent hashing approaches utilizing the deep
learning models. Finally, we discuss the future direction and trends of
research in this area.
|
Algorithmic statistics, prediction and machine learning | Algorithmic statistics considers the following problem: given a binary string
$x$ (e.g., some experimental data), find a "good" explanation of this data. It
uses algorithmic information theory to define formally what is a good
explanation. In this paper we extend this framework in two directions.
First, the explanations are not only interesting in themselves but also used
for prediction: we want to know what kind of data we may reasonably expect in
similar situations (repeating the same experiment). We show that some kind of
hierarchy can be constructed both in terms of algorithmic statistics and using
the notion of a priori probability, and these two approaches turn out to be
equivalent.
Second, a more realistic approach that goes back to machine learning theory,
assumes that we have not a single data string $x$ but some set of "positive
examples" $x_1,\ldots,x_l$ that all belong to some unknown set $A$, a property
that we want to learn. We want this set $A$ to contain all positive examples
and to be as small and simple as possible. We show how algorithmic statistic
can be extended to cover this situation.
|
Fast and Simple PCA via Convex Optimization | The problem of principle component analysis (PCA) is traditionally solved by
spectral or algebraic methods. We show how computing the leading principal
component could be reduced to solving a \textit{small} number of
well-conditioned {\it convex} optimization problems. This gives rise to a new
efficient method for PCA based on recent advances in stochastic methods for
convex optimization.
In particular we show that given a $d\times d$ matrix $\X =
\frac{1}{n}\sum_{i=1}^n\x_i\x_i^{\top}$ with top eigenvector $\u$ and top
eigenvalue $\lambda_1$ it is possible to: \begin{itemize} \item compute a unit
vector $\w$ such that $(\w^{\top}\u)^2 \geq 1-\epsilon$ in
$\tilde{O}\left({\frac{d}{\delta^2}+N}\right)$ time, where $\delta = \lambda_1
- \lambda_2$ and $N$ is the total number of non-zero entries in
$\x_1,...,\x_n$,
\item compute a unit vector $\w$ such that $\w^{\top}\X\w \geq
\lambda_1-\epsilon$ in $\tilde{O}(d/\epsilon^2)$ time. \end{itemize} To the
best of our knowledge, these bounds are the fastest to date for a wide regime
of parameters. These results could be further accelerated when $\delta$ (in the
first case) and $\epsilon$ (in the second case) are smaller than $\sqrt{d/N}$.
|
Accelerating Optimization via Adaptive Prediction | We present a powerful general framework for designing data-dependent
optimization algorithms, building upon and unifying recent techniques in
adaptive regularization, optimistic gradient predictions, and problem-dependent
randomization. We first present a series of new regret guarantees that hold at
any time and under very minimal assumptions, and then show how different
relaxations recover existing algorithms, both basic as well as more recent
sophisticated ones. Finally, we show how combining adaptivity, optimism, and
problem-dependent randomization can guide the design of algorithms that benefit
from more favorable guarantees than recent state-of-the-art methods.
|
"Oddball SGD": Novelty Driven Stochastic Gradient Descent for Training
Deep Neural Networks | Stochastic Gradient Descent (SGD) is arguably the most popular of the machine
learning methods applied to training deep neural networks (DNN) today. It has
recently been demonstrated that SGD can be statistically biased so that certain
elements of the training set are learned more rapidly than others. In this
article, we place SGD into a feedback loop whereby the probability of selection
is proportional to error magnitude. This provides a novelty-driven oddball SGD
process that learns more rapidly than traditional SGD by prioritising those
elements of the training set with the largest novelty (error). In our DNN
example, oddball SGD trains some 50x faster than regular SGD.
|
BLC: Private Matrix Factorization Recommenders via Automatic Group
Learning | We propose a privacy-enhanced matrix factorization recommender that exploits
the fact that users can often be grouped together by interest. This allows a
form of "hiding in the crowd" privacy. We introduce a novel matrix
factorization approach suited to making recommendations in a shared group (or
nym) setting and the BLC algorithm for carrying out this matrix factorization
in a privacy-enhanced manner. We demonstrate that the increased privacy does
not come at the cost of reduced recommendation accuracy.
|
Word, graph and manifold embedding from Markov processes | Continuous vector representations of words and objects appear to carry
surprisingly rich semantic content. In this paper, we advance both the
conceptual and theoretical understanding of word embeddings in three ways.
First, we ground embeddings in semantic spaces studied in
cognitive-psychometric literature and introduce new evaluation tasks. Second,
in contrast to prior work, we take metric recovery as the key object of study,
unify existing algorithms as consistent metric recovery methods based on
co-occurrence counts from simple Markov random walks, and propose a new
recovery algorithm. Third, we generalize metric recovery to graphs and
manifolds, relating co-occurence counts on random walks in graphs and random
processes on manifolds to the underlying metric to be recovered, thereby
reconciling manifold estimation and embedding algorithms. We compare embedding
algorithms across a range of tasks, from nonlinear dimensionality reduction to
three semantic language tasks, including analogies, sequence completion, and
classification.
|
STDP as presynaptic activity times rate of change of postsynaptic
activity | We introduce a weight update formula that is expressed only in terms of
firing rates and their derivatives and that results in changes consistent with
those associated with spike-timing dependent plasticity (STDP) rules and
biological observations, even though the explicit timing of spikes is not
needed. The new rule changes a synaptic weight in proportion to the product of
the presynaptic firing rate and the temporal rate of change of activity on the
postsynaptic side. These quantities are interesting for studying theoretical
explanation for synaptic changes from a machine learning perspective. In
particular, if neural dynamics moved neural activity towards reducing some
objective function, then this STDP rule would correspond to stochastic gradient
descent on that objective function.
|
Telugu OCR Framework using Deep Learning | In this paper, we address the task of Optical Character Recognition(OCR) for
the Telugu script. We present an end-to-end framework that segments the text
image, classifies the characters and extracts lines using a language model. The
segmentation is based on mathematical morphology. The classification module,
which is the most challenging task of the three, is a deep convolutional neural
network. The language is modelled as a third degree markov chain at the glyph
level. Telugu script is a complex alphasyllabary and the language is
agglutinative, making the problem hard. In this paper we apply the latest
advances in neural networks to achieve state-of-the-art error rates. We also
review convolutional neural networks in great detail and expound the
statistical justification behind the many tricks needed to make Deep Learning
work.
|
Denoising without access to clean data using a partitioned autoencoder | Training a denoising autoencoder neural network requires access to truly
clean data, a requirement which is often impractical. To remedy this, we
introduce a method to train an autoencoder using only noisy data, having
examples with and without the signal class of interest. The autoencoder learns
a partitioned representation of signal and noise, learning to reconstruct each
separately. We illustrate the method by denoising birdsong audio (available
abundantly in uncontrolled noisy datasets) using a convolutional autoencoder.
|
Robust Image Sentiment Analysis Using Progressively Trained and Domain
Transferred Deep Networks | Sentiment analysis of online user generated content is important for many
social media analytics tasks. Researchers have largely relied on textual
sentiment analysis to develop systems to predict political elections, measure
economic indicators, and so on. Recently, social media users are increasingly
using images and videos to express their opinions and share their experiences.
Sentiment analysis of such large scale visual content can help better extract
user sentiments toward events or topics, such as those in image tweets, so that
prediction of sentiment from visual content is complementary to textual
sentiment analysis. Motivated by the needs in leveraging large scale yet noisy
training data to solve the extremely challenging problem of image sentiment
analysis, we employ Convolutional Neural Networks (CNN). We first design a
suitable CNN architecture for image sentiment analysis. We obtain half a
million training samples by using a baseline sentiment algorithm to label
Flickr images. To make use of such noisy machine labeled data, we employ a
progressive strategy to fine-tune the deep network. Furthermore, we improve the
performance on Twitter images by inducing domain transfer with a small number
of manually labeled Twitter images. We have conducted extensive experiments on
manually labeled Twitter images. The results show that the proposed CNN can
achieve better performance in image sentiment analysis than competing
algorithms.
|
Significance Analysis of High-Dimensional, Low-Sample Size Partially
Labeled Data | Classification and clustering are both important topics in statistical
learning. A natural question herein is whether predefined classes are really
different from one another, or whether clusters are really there. Specifically,
we may be interested in knowing whether the two classes defined by some class
labels (when they are provided), or the two clusters tagged by a clustering
algorithm (where class labels are not provided), are from the same underlying
distribution. Although both are challenging questions for the high-dimensional,
low-sample size data, there has been some recent development for both. However,
when it is costly to manually place labels on observations, it is often that
only a small portion of the class labels is available. In this article, we
propose a significance analysis approach for such type of data, namely
partially labeled data. Our method makes use of the whole data and tries to
test the class difference as if all the labels were observed. Compared to a
testing method that ignores the label information, our method provides a
greater power, meanwhile, maintaining the size, illustrated by a comprehensive
simulation study. Theoretical properties of the proposed method are studied
with emphasis on the high-dimensional, low-sample size setting. Our simulated
examples help to understand when and how the information extracted from the
labeled data can be effective. A real data example further illustrates the
usefulness of the proposed method.
|
Multilayer bootstrap network for unsupervised speaker recognition | We apply multilayer bootstrap network (MBN), a recent proposed unsupervised
learning method, to unsupervised speaker recognition. The proposed method first
extracts supervectors from an unsupervised universal background model, then
reduces the dimension of the high-dimensional supervectors by multilayer
bootstrap network, and finally conducts unsupervised speaker recognition by
clustering the low-dimensional data. The comparison results with 2 unsupervised
and 1 supervised speaker recognition techniques demonstrate the effectiveness
and robustness of the proposed method.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.