title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Alzheimer's Disease Diagnostics by Adaptation of 3D Convolutional
Network | Early diagnosis, playing an important role in preventing progress and
treating the Alzheimer\{'}s disease (AD), is based on classification of
features extracted from brain images. The features have to accurately capture
main AD-related variations of anatomical brain structures, such as, e.g.,
ventricles size, hippocampus shape, cortical thickness, and brain volume. This
paper proposed to predict the AD with a deep 3D convolutional neural network
(3D-CNN), which can learn generic features capturing AD biomarkers and adapt to
different domain datasets. The 3D-CNN is built upon a 3D convolutional
autoencoder, which is pre-trained to capture anatomical shape variations in
structural brain MRI scans. Fully connected upper layers of the 3D-CNN are then
fine-tuned for each task-specific AD classification. Experiments on the
CADDementia MRI dataset with no skull-stripping preprocessing have shown our
3D-CNN outperforms several conventional classifiers by accuracy. Abilities of
the 3D-CNN to generalize the features learnt and adapt to other domains have
been validated on the ADNI dataset.
|
Outlier absorbing based on a Bayesian approach | The presence of outliers is prevalent in machine learning applications and
may produce misleading results. In this paper a new method for dealing with
outliers and anomal samples is proposed. To overcome the outlier issue, the
proposed method combines the global and local views of the samples. By
combination of these views, our algorithm performs in a robust manner. The
experimental results show the capabilities of the proposed method.
|
Adaptive Neighborhood Graph Construction for Inference in
Multi-Relational Networks | A neighborhood graph, which represents the instances as vertices and their
relations as weighted edges, is the basis of many semi-supervised and
relational models for node labeling and link prediction. Most methods employ a
sequential process to construct the neighborhood graph. This process often
consists of generating a candidate graph, pruning the candidate graph to make a
neighborhood graph, and then performing inference on the variables (i.e.,
nodes) in the neighborhood graph. In this paper, we propose a framework that
can dynamically adapt the neighborhood graph based on the states of variables
from intermediate inference results, as well as structural properties of the
relations connecting them. A key strength of our framework is its ability to
handle multi-relational data and employ varying amounts of relations for each
instance based on the intermediate inference results. We formulate the link
prediction task as inference on neighborhood graphs, and include preliminary
results illustrating the effects of different strategies in our proposed
framework.
|
Group Sparse Regularization for Deep Neural Networks | In this paper, we consider the joint task of simultaneously optimizing (i)
the weights of a deep neural network, (ii) the number of neurons for each
hidden layer, and (iii) the subset of active input features (i.e., feature
selection). While these problems are generally dealt with separately, we
present a simple regularized formulation allowing to solve all three of them in
parallel, using standard optimization routines. Specifically, we extend the
group Lasso penalty (originated in the linear regression literature) in order
to impose group-level sparsity on the network's connections, where each group
is defined as the set of outgoing weights from a unit. Depending on the
specific case, the weights can be related to an input variable, to a hidden
neuron, or to a bias unit, thus performing simultaneously all the
aforementioned tasks in order to obtain a compact network. We perform an
extensive experimental evaluation, by comparing with classical weight decay and
Lasso penalties. We show that a sparse version of the group Lasso penalty is
able to achieve competitive performances, while at the same time resulting in
extremely compact networks with a smaller number of input features. We evaluate
both on a toy dataset for handwritten digit recognition, and on multiple
realistic large-scale classification problems.
|
Big IoT and social networking data for smart cities: Algorithmic
improvements on Big Data Analysis in the context of RADICAL city applications | In this paper we present a SOA (Service Oriented Architecture)-based
platform, enabling the retrieval and analysis of big datasets stemming from
social networking (SN) sites and Internet of Things (IoT) devices, collected by
smart city applications and socially-aware data aggregation services. A large
set of city applications in the areas of Participating Urbanism, Augmented
Reality and Sound-Mapping throughout participating cities is being applied,
resulting into produced sets of millions of user-generated events and online SN
reports fed into the RADICAL platform. Moreover, we study the application of
data analytics such as sentiment analysis to the combined IoT and SN data saved
into an SQL database, further investigating algorithmic and configurations to
minimize delays in dataset processing and results retrieval.
|
Approximate Joint Matrix Triangularization | We consider the problem of approximate joint triangularization of a set of
noisy jointly diagonalizable real matrices. Approximate joint triangularizers
are commonly used in the estimation of the joint eigenstructure of a set of
matrices, with applications in signal processing, linear algebra, and tensor
decomposition. By assuming the input matrices to be perturbations of
noise-free, simultaneously diagonalizable ground-truth matrices, the
approximate joint triangularizers are expected to be perturbations of the exact
joint triangularizers of the ground-truth matrices. We provide a priori and a
posteriori perturbation bounds on the `distance' between an approximate joint
triangularizer and its exact counterpart. The a priori bounds are theoretical
inequalities that involve functions of the ground-truth matrices and noise
matrices, whereas the a posteriori bounds are given in terms of observable
quantities that can be computed from the input matrices. From a practical
perspective, the problem of finding the best approximate joint triangularizer
of a set of noisy matrices amounts to solving a nonconvex optimization problem.
We show that, under a condition on the noise level of the input matrices, it is
possible to find a good initial triangularizer such that the solution obtained
by any local descent-type algorithm has certain global guarantees. Finally, we
discuss the application of approximate joint matrix triangularization to
canonical tensor decomposition and we derive novel estimation error bounds.
|
Alzheimer's Disease Diagnostics by a Deeply Supervised Adaptable 3D
Convolutional Network | Early diagnosis, playing an important role in preventing progress and
treating the Alzheimer's disease (AD), is based on classification of features
extracted from brain images. The features have to accurately capture main
AD-related variations of anatomical brain structures, such as, e.g., ventricles
size, hippocampus shape, cortical thickness, and brain volume. This paper
proposes to predict the AD with a deep 3D convolutional neural network
(3D-CNN), which can learn generic features capturing AD biomarkers and adapt to
different domain datasets. The 3D-CNN is built upon a 3D convolutional
autoencoder, which is pre-trained to capture anatomical shape variations in
structural brain MRI scans. Fully connected upper layers of the 3D-CNN are then
fine-tuned for each task-specific AD classification. Experiments on the
\emph{ADNI} MRI dataset with no skull-stripping preprocessing have shown our
3D-CNN outperforms several conventional classifiers by accuracy and robustness.
Abilities of the 3D-CNN to generalize the features learnt and adapt to other
domains have been validated on the \emph{CADDementia} dataset.
|
Rademacher Complexity Bounds for a Penalized Multiclass Semi-Supervised
Algorithm | We propose Rademacher complexity bounds for multiclass classifiers trained
with a two-step semi-supervised model. In the first step, the algorithm
partitions the partially labeled data and then identifies dense clusters
containing $\kappa$ predominant classes using the labeled training examples
such that the proportion of their non-predominant classes is below a fixed
threshold. In the second step, a classifier is trained by minimizing a margin
empirical loss over the labeled training set and a penalization term measuring
the disability of the learner to predict the $\kappa$ predominant classes of
the identified clusters. The resulting data-dependent generalization error
bound involves the margin distribution of the classifier, the stability of the
clustering technique used in the first step and Rademacher complexity terms
corresponding to partially labeled training data. Our theoretical result
exhibit convergence rates extending those proposed in the literature for the
binary case, and experimental results on different multiclass classification
problems show empirical evidence that supports the theory.
|
node2vec: Scalable Feature Learning for Networks | Prediction tasks over nodes and edges in networks require careful effort in
engineering features used by learning algorithms. Recent research in the
broader field of representation learning has led to significant progress in
automating prediction by learning the features themselves. However, present
feature learning approaches are not expressive enough to capture the diversity
of connectivity patterns observed in networks. Here we propose node2vec, an
algorithmic framework for learning continuous feature representations for nodes
in networks. In node2vec, we learn a mapping of nodes to a low-dimensional
space of features that maximizes the likelihood of preserving network
neighborhoods of nodes. We define a flexible notion of a node's network
neighborhood and design a biased random walk procedure, which efficiently
explores diverse neighborhoods. Our algorithm generalizes prior work which is
based on rigid notions of network neighborhoods, and we argue that the added
flexibility in exploring neighborhoods is the key to learning richer
representations. We demonstrate the efficacy of node2vec over existing
state-of-the-art techniques on multi-label classification and link prediction
in several real-world networks from diverse domains. Taken together, our work
represents a new way for efficiently learning state-of-the-art task-independent
representations in complex networks.
|
Unsupervised Learning of 3D Structure from Images | A key goal of computer vision is to recover the underlying 3D structure from
2D observations of the world. In this paper we learn strong deep generative
models of 3D structures, and recover these structures from 3D and 2D images via
probabilistic inference. We demonstrate high-quality samples and report
log-likelihoods on several datasets, including ShapeNet [2], and establish the
first benchmarks in the literature. We also show how these models and their
inference networks can be trained end-to-end from 2D images. This demonstrates
for the first time the feasibility of learning to infer 3D representations of
the world in a purely unsupervised manner.
|
Understanding the Energy and Precision Requirements for Online Learning | It is well-known that the precision of data, hyperparameters, and internal
representations employed in learning systems directly impacts its energy,
throughput, and latency. The precision requirements for the training algorithm
are also important for systems that learn on-the-fly. Prior work has shown that
the data and hyperparameters can be quantized heavily without incurring much
penalty in classification accuracy when compared to floating point
implementations. These works suffer from two key limitations. First, they
assume uniform precision for the classifier and for the training algorithm and
thus miss out on the opportunity to further reduce precision. Second, prior
works are empirical studies. In this article, we overcome both these
limitations by deriving analytical lower bounds on the precision requirements
of the commonly employed stochastic gradient descent (SGD) on-line learning
algorithm in the specific context of a support vector machine (SVM). Lower
bounds on the data precision are derived in terms of the the desired
classification accuracy and precision of the hyperparameters used in the
classifier. Additionally, lower bounds on the hyperparameter precision in the
SGD training algorithm are obtained. These bounds are validated using both
synthetic and the UCI breast cancer dataset. Additionally, the impact of these
precisions on the energy consumption of a fixed-point SVM with on-line training
is studied.
|
Confidence-Weighted Bipartite Ranking | Bipartite ranking is a fundamental machine learning and data mining problem.
It commonly concerns the maximization of the AUC metric. Recently, a number of
studies have proposed online bipartite ranking algorithms to learn from massive
streams of class-imbalanced data. These methods suggest both linear and
kernel-based bipartite ranking algorithms based on first and second-order
online learning. Unlike kernelized ranker, linear ranker is more scalable
learning algorithm. The existing linear online bipartite ranking algorithms
lack either handling non-separable data or constructing adaptive large margin.
These limitations yield unreliable bipartite ranking performance. In this work,
we propose a linear online confidence-weighted bipartite ranking algorithm
(CBR) that adopts soft confidence-weighted learning. The proposed algorithm
leverages the same properties of soft confidence-weighted learning in a
framework for bipartite ranking. We also develop a diagonal variation of the
proposed confidence-weighted bipartite ranking algorithm to deal with
high-dimensional data by maintaining only the diagonal elements of the
covariance matrix. We empirically evaluate the effectiveness of the proposed
algorithms on several benchmark and high-dimensional datasets. The experimental
results validate the reliability of the proposed algorithms. The results also
show that our algorithms outperform or are at least comparable to the competing
online AUC maximization methods.
|
Neighborhood Features Help Detecting Non-Technical Losses in Big Data
Sets | Electricity theft is a major problem around the world in both developed and
developing countries and may range up to 40% of the total electricity
distributed. More generally, electricity theft belongs to non-technical losses
(NTL), which are losses that occur during the distribution of electricity in
power grids. In this paper, we build features from the neighborhood of
customers. We first split the area in which the customers are located into
grids of different sizes. For each grid cell we then compute the proportion of
inspected customers and the proportion of NTL found among the inspected
customers. We then analyze the distributions of features generated and show why
they are useful to predict NTL. In addition, we compute features from the
consumption time series of customers. We also use master data features of
customers, such as their customer class and voltage of their connection. We
compute these features for a Big Data base of 31M meter readings, 700K
customers and 400K inspection results. We then use these features to train four
machine learning algorithms that are particularly suitable for Big Data sets
because of their parallelizable structure: logistic regression, k-nearest
neighbors, linear support vector machine and random forest. Using the
neighborhood features instead of only analyzing the time series has resulted in
appreciable results for Big Data sets for varying NTL proportions of 1%-90%.
This work can therefore be deployed to a wide range of different regions around
the world.
|
Optimal Quantum Sample Complexity of Learning Algorithms | $ \newcommand{\eps}{\varepsilon} $In learning theory, the VC dimension of a
concept class $C$ is the most common way to measure its "richness." In the PAC
model $$ \Theta\Big(\frac{d}{\eps} + \frac{\log(1/\delta)}{\eps}\Big) $$
examples are necessary and sufficient for a learner to output, with probability
$1-\delta$, a hypothesis $h$ that is $\eps$-close to the target concept $c$. In
the related agnostic model, where the samples need not come from a $c\in C$, we
know that $$ \Theta\Big(\frac{d}{\eps^2} + \frac{\log(1/\delta)}{\eps^2}\Big)
$$ examples are necessary and sufficient to output an hypothesis $h\in C$ whose
error is at most $\eps$ worse than the best concept in $C$.
Here we analyze quantum sample complexity, where each example is a coherent
quantum state. This model was introduced by Bshouty and Jackson, who showed
that quantum examples are more powerful than classical examples in some
fixed-distribution settings. However, Atici and Servedio, improved by Zhang,
showed that in the PAC setting, quantum examples cannot be much more powerful:
the required number of quantum examples is $$
\Omega\Big(\frac{d^{1-\eta}}{\eps} + d + \frac{\log(1/\delta)}{\eps}\Big)\mbox{
for all }\eta> 0. $$ Our main result is that quantum and classical sample
complexity are in fact equal up to constant factors in both the PAC and
agnostic models. We give two approaches. The first is a fairly simple
information-theoretic argument that yields the above two classical bounds and
yields the same bounds for quantum sample complexity up to a $\log(d/\eps)$
factor. We then give a second approach that avoids the log-factor loss, based
on analyzing the behavior of the "Pretty Good Measurement" on the quantum state
identification problems that correspond to learning. This shows classical and
quantum sample complexity are equal up to constant factors.
|
Sequence to Backward and Forward Sequences: A Content-Introducing
Approach to Generative Short-Text Conversation | Using neural networks to generate replies in human-computer dialogue systems
is attracting increasing attention over the past few years. However, the
performance is not satisfactory: the neural network tends to generate safe,
universally relevant replies which carry little meaning. In this paper, we
propose a content-introducing approach to neural network-based generative
dialogue systems. We first use pointwise mutual information (PMI) to predict a
noun as a keyword, reflecting the main gist of the reply. We then propose
seq2BF, a "sequence to backward and forward sequences" model, which generates a
reply containing the given keyword. Experimental results show that our approach
significantly outperforms traditional sequence-to-sequence models in terms of
human evaluation and the entropy measure, and that the predicted keyword can
appear at an appropriate position in the reply.
|
Accelerate Stochastic Subgradient Method by Leveraging Local Growth
Condition | In this paper, a new theory is developed for first-order stochastic convex
optimization, showing that the global convergence rate is sufficiently
quantified by a local growth rate of the objective function in a neighborhood
of the optimal solutions. In particular, if the objective function $F(\mathbf
w)$ in the $\epsilon$-sublevel set grows as fast as $\|\mathbf w - \mathbf
w_*\|_2^{1/\theta}$, where $\mathbf w_*$ represents the closest optimal
solution to $\mathbf w$ and $\theta\in(0,1]$ quantifies the local growth rate,
the iteration complexity of first-order stochastic optimization for achieving
an $\epsilon$-optimal solution can be $\widetilde O(1/\epsilon^{2(1-\theta)})$,
which is optimal at most up to a logarithmic factor. To achieve the faster
global convergence, we develop two different accelerated stochastic subgradient
methods by iteratively solving the original problem approximately in a local
region around a historical solution with the size of the local region gradually
decreasing as the solution approaches the optimal set. Besides the theoretical
improvements, this work also includes new contributions towards making the
proposed algorithms practical: (i) we present practical variants of accelerated
stochastic subgradient methods that can run without the knowledge of
multiplicative growth constant and even the growth rate $\theta$; (ii) we
consider a broad family of problems in machine learning to demonstrate that the
proposed algorithms enjoy faster convergence than traditional stochastic
subgradient method. We also characterize the complexity of the proposed
algorithms for ensuring the gradient is small without the smoothness
assumption.
|
Bootstrap Model Aggregation for Distributed Statistical Learning | In distributed, or privacy-preserving learning, we are often given a set of
probabilistic models estimated from different local repositories, and asked to
combine them into a single model that gives efficient statistical estimation. A
simple method is to linearly average the parameters of the local models, which,
however, tends to be degenerate or not applicable on non-convex models, or
models with different parameter dimensions. One more practical strategy is to
generate bootstrap samples from the local models, and then learn a joint model
based on the combined bootstrap set. Unfortunately, the bootstrap procedure
introduces additional noise and can significantly deteriorate the performance.
In this work, we propose two variance reduction methods to correct the
bootstrap noise, including a weighted M-estimator that is both statistically
efficient and practically powerful. Both theoretical and empirical analysis is
provided to demonstrate our methods.
|
Application of Statistical Relational Learning to Hybrid Recommendation
Systems | Recommendation systems usually involve exploiting the relations among known
features and content that describe items (content-based filtering) or the
overlap of similar users who interacted with or rated the target item
(collaborative filtering). To combine these two filtering approaches, current
model-based hybrid recommendation systems typically require extensive feature
engineering to construct a user profile. Statistical Relational Learning (SRL)
provides a straightforward way to combine the two approaches. However, due to
the large scale of the data used in real world recommendation systems, little
research exists on applying SRL models to hybrid recommendation systems, and
essentially none of that research has been applied on real big-data-scale
systems. In this paper, we proposed a way to adapt the state-of-the-art in SRL
learning approaches to construct a real hybrid recommendation system.
Furthermore, in order to satisfy a common requirement in recommendation systems
(i.e. that false positives are more undesirable and therefore penalized more
harshly than false negatives), our approach can also allow tuning the trade-off
between the precision and recall of the system in a principled way. Our
experimental results demonstrate the efficiency of our proposed approach as
well as its improved performance on recommendation precision.
|
AdaNet: Adaptive Structural Learning of Artificial Neural Networks | We present new algorithms for adaptively learning artificial neural networks.
Our algorithms (AdaNet) adaptively learn both the structure of the network and
its weights. They are based on a solid theoretical analysis, including
data-dependent generalization guarantees that we prove and discuss in detail.
We report the results of large-scale experiments with one of our algorithms on
several binary classification tasks extracted from the CIFAR-10 dataset. The
results demonstrate that our algorithm can automatically learn network
structures with very competitive performance accuracies when compared with
those achieved for neural networks found by standard approaches.
|
Minimalist Regression Network with Reinforced Gradients and Weighted
Estimates: a Case Study on Parameters Estimation in Automated Welding | This paper presents a minimalist neural regression network as an aggregate of
independent identical regression blocks that are trained simultaneously.
Moreover, it introduces a new multiplicative parameter, shared by all the
neural units of a given layer, to maintain the quality of its gradients.
Furthermore, it increases its estimation accuracy via learning a weight factor
whose quantity captures the redundancy between the estimated and actual values
at each training iteration. We choose the estimation of the direct weld
parameters of different welding techniques to show a significant improvement in
calculation of these parameters by our model in contrast to state-of-the-arts
techniques in the literature. Furthermore, we demonstrate the ability of our
model to retain its performance when presented with combined data of different
welding techniques. This is a nontrivial result in attaining an scalable model
whose quality of estimation is independent of adopted welding techniques.
|
How to Evaluate the Quality of Unsupervised Anomaly Detection
Algorithms? | When sufficient labeled data are available, classical criteria based on
Receiver Operating Characteristic (ROC) or Precision-Recall (PR) curves can be
used to compare the performance of un-supervised anomaly detection algorithms.
However , in many situations, few or no data are labeled. This calls for
alternative criteria one can compute on non-labeled data. In this paper, two
criteria that do not require labels are empirically shown to discriminate
accurately (w.r.t. ROC or PR based criteria) between algorithms. These criteria
are based on existing Excess-Mass (EM) and Mass-Volume (MV) curves, which
generally cannot be well estimated in large dimension. A methodology based on
feature sub-sampling and aggregating is also described and tested, extending
the use of these criteria to high-dimensional datasets and solving major
drawbacks inherent to standard EM and MV curves.
|
Stochastic Quasi-Newton Methods for Nonconvex Stochastic Optimization | In this paper we study stochastic quasi-Newton methods for nonconvex
stochastic optimization, where we assume that noisy information about the
gradients of the objective function is available via a stochastic first-order
oracle (SFO). We propose a general framework for such methods, for which we
prove almost sure convergence to stationary points and analyze its worst-case
iteration complexity. When a randomly chosen iterate is returned as the output
of such an algorithm, we prove that in the worst-case, the SFO-calls complexity
is $O(\epsilon^{-2})$ to ensure that the expectation of the squared norm of the
gradient is smaller than the given accuracy tolerance $\epsilon$. We also
propose a specific algorithm, namely a stochastic damped L-BFGS (SdLBFGS)
method, that falls under the proposed framework. {Moreover, we incorporate the
SVRG variance reduction technique into the proposed SdLBFGS method, and analyze
its SFO-calls complexity. Numerical results on a nonconvex binary
classification problem using SVM, and a multiclass classification problem using
neural networks are reported.
|
Temporal Topic Analysis with Endogenous and Exogenous Processes | We consider the problem of modeling temporal textual data taking endogenous
and exogenous processes into account. Such text documents arise in real world
applications, including job advertisements and economic news articles, which
are influenced by the fluctuations of the general economy. We propose a
hierarchical Bayesian topic model which imposes a "group-correlated"
hierarchical structure on the evolution of topics over time incorporating both
processes, and show that this model can be estimated from Markov chain Monte
Carlo sampling methods. We further demonstrate that this model captures the
intrinsic relationships between the topic distribution and the time-dependent
factors, and compare its performance with latent Dirichlet allocation (LDA) and
two other related models. The model is applied to two collections of documents
to illustrate its empirical performance: online job advertisements from
DirectEmployers Association and journalists' postings on BusinessInsider.com.
|
Resource Allocation in a MAC with and without security via Game
Theoretic Learning | In this paper a $K$-user fading multiple access channel with and without
security constraints is studied. First we consider a F-MAC without the security
constraints. Under the assumption of individual CSI of users, we propose the
problem of power allocation as a stochastic game when the receiver sends an ACK
or a NACK depending on whether it was able to decode the message or not. We
have used Multiplicative weight no-regret algorithm to obtain a Coarse
Correlated Equilibrium (CCE). Then we consider the case when the users can
decode ACK/NACK of each other. In this scenario we provide an algorithm to
maximize the weighted sum-utility of all the users and obtain a Pareto optimal
point. PP is socially optimal but may be unfair to individual users. Next we
consider the case where the users can cooperate with each other so as to
disagree with the policy which will be unfair to individual user. We then
obtain a Nash bargaining solution, which in addition to being Pareto optimal,
is also fair to each user.
Next we study a $K$-user fading multiple access wiretap Channel with CSI of
Eve available to the users. We use the previous algorithms to obtain a CCE, PP
and a NBS.
Next we consider the case where each user does not know the CSI of Eve but
only its distribution. In that case we use secrecy outage as the criterion for
the receiver to send an ACK or a NACK. Here also we use the previous algorithms
to obtain a CCE, PP or a NBS. Finally we show that our algorithms can be
extended to the case where a user can transmit at different rates. At the end
we provide a few examples to compute different solutions and compare them under
different CSI scenarios.
|
Learning Discriminative Features using Encoder-Decoder type Deep Neural
Nets | As machine learning is applied to an increasing variety of complex problems,
which are defined by high dimensional and complex data sets, the necessity for
task oriented feature learning grows in importance. With the advancement of
Deep Learning algorithms, various successful feature learning techniques have
evolved. In this paper, we present a novel way of learning discriminative
features by training Deep Neural Nets which have Encoder or Decoder type
architecture similar to an Autoencoder. We demonstrate that our approach can
learn discriminative features which can perform better at pattern
classification tasks when the number of training samples is relatively small in
size.
|
An Aggregate and Iterative Disaggregate Algorithm with Proven Optimality
in Machine Learning | We propose a clustering-based iterative algorithm to solve certain
optimization problems in machine learning, where we start the algorithm by
aggregating the original data, solving the problem on aggregated data, and then
in subsequent steps gradually disaggregate the aggregated data. We apply the
algorithm to common machine learning problems such as the least absolute
deviation regression problem, support vector machines, and semi-supervised
support vector machines. We derive model-specific data aggregation and
disaggregation procedures. We also show optimality, convergence, and the
optimality gap of the approximated solution in each iteration. A computational
study is provided.
|
Algorithms for Generalized Cluster-wise Linear Regression | Cluster-wise linear regression (CLR), a clustering problem intertwined with
regression, is to find clusters of entities such that the overall sum of
squared errors from regressions performed over these clusters is minimized,
where each cluster may have different variances. We generalize the CLR problem
by allowing each entity to have more than one observation, and refer to it as
generalized CLR. We propose an exact mathematical programming based approach
relying on column generation, a column generation based heuristic algorithm
that clusters predefined groups of entities, a metaheuristic genetic algorithm
with adapted Lloyd's algorithm for K-means clustering, a two-stage approach,
and a modified algorithm of Sp{\"a}th \cite{Spath1979} for solving generalized
CLR. We examine the performance of our algorithms on a stock keeping unit (SKU)
clustering problem employed in forecasting halo and cannibalization effects in
promotions using real-world retail data from a large supermarket chain. In the
SKU clustering problem, the retailer needs to cluster SKUs based on their
seasonal effects in response to promotions. The seasonal effects are the
results of regressions with predictors being promotion mechanisms and seasonal
dummies performed over clusters generated. We compare the performance of all
proposed algorithms for the SKU problem with real-world and synthetic data.
|
An optimal learning method for developing personalized treatment regimes | A treatment regime is a function that maps individual patient information to
a recommended treatment, hence explicitly incorporating the heterogeneity in
need for treatment across individuals. Patient responses are dichotomous and
can be predicted through an unknown relationship that depends on the patient
information and the selected treatment. The goal is to find the treatments that
lead to the best patient responses on average. Each experiment is expensive,
forcing us to learn the most from each experiment. We adopt a Bayesian approach
both to incorporate possible prior information and to update our treatment
regime continuously as information accrues, with the potential to allow smaller
yet more informative trials and for patients to receive better treatment. By
formulating the problem as contextual bandits, we introduce a knowledge
gradient policy to guide the treatment assignment by maximizing the expected
value of information, for which an approximation method is used to overcome
computational challenges. We provide a detailed study on how to make sequential
medical decisions under uncertainty to reduce health care costs on a real world
knee replacement dataset. We use clustering and LASSO to deal with the
intrinsic sparsity in health datasets. We show experimentally that even though
the problem is sparse, through careful selection of physicians (versus picking
them at random), we can significantly improve the success rates.
|
On Sampling and Greedy MAP Inference of Constrained Determinantal Point
Processes | Subset selection problems ask for a small, diverse yet representative subset
of the given data. When pairwise similarities are captured by a kernel, the
determinants of submatrices provide a measure of diversity or independence of
items within a subset. Matroid theory gives another notion of independence,
thus giving rise to optimization and sampling questions about Determinantal
Point Processes (DPPs) under matroid constraints. Partition constraints, as a
special case, arise naturally when incorporating additional labeling or
clustering information, besides the kernel, in DPPs. Finding the maximum
determinant submatrix under matroid constraints on its row/column indices has
been previously studied. However, the corresponding question of sampling from
DPPs under matroid constraints has been unresolved, beyond the simple
cardinality constrained k-DPPs. We give the first polynomial time algorithm to
sample exactly from DPPs under partition constraints, for any constant number
of partitions. We complement this by a complexity theoretic barrier that rules
out such a result under general matroid constraints. Our experiments indicate
that partition-constrained DPPs offer more flexibility and more diversity than
k-DPPs and their naive extensions, while being reasonably efficient in running
time. We also show that a simple greedy initialization followed by local search
gives improved approximation guarantees for the problem of MAP inference from
k- DPPs on well-conditioned kernels. Our experiments show that this improvement
is significant for larger values of k, supporting our theoretical result.
|
Bagged Boosted Trees for Classification of Ecological Momentary
Assessment Data | Ecological Momentary Assessment (EMA) data is organized in multiple levels
(per-subject, per-day, etc.) and this particular structure should be taken into
account in machine learning algorithms used in EMA like decision trees and its
variants. We propose a new algorithm called BBT (standing for Bagged Boosted
Trees) that is enhanced by a over/under sampling method and can provide better
estimates for the conditional class probability function. Experimental results
on a real-world dataset show that BBT can benefit EMA data classification and
performance.
|
Tensor Decomposition for Signal Processing and Machine Learning | Tensors or {\em multi-way arrays} are functions of three or more indices
$(i,j,k,\cdots)$ -- similar to matrices (two-way arrays), which are functions
of two indices $(r,c)$ for (row,column). Tensors have a rich history,
stretching over almost a century, and touching upon numerous disciplines; but
they have only recently become ubiquitous in signal and data analytics at the
confluence of signal processing, statistics, data mining and machine learning.
This overview article aims to provide a good starting point for researchers and
practitioners interested in learning about and working with tensors. As such,
it focuses on fundamentals and motivation (using various application examples),
aiming to strike an appropriate balance of breadth {\em and depth} that will
enable someone having taken first graduate courses in matrix algebra and
probability to get started doing research and/or developing tensor algorithms
and software. Some background in applied optimization is useful but not
strictly required. The material covered includes tensor rank and rank
decomposition; basic tensor factorization models and their relationships and
properties (including fairly good coverage of identifiability); broad coverage
of algorithms ranging from alternating optimization to stochastic gradient;
statistical performance analysis; and applications ranging from source
separation to collaborative filtering, mixture and topic modeling,
classification, and multilinear subspace learning.
|
A New Hierarchical Redundancy Eliminated Tree Augmented Naive Bayes
Classifier for Coping with Gene Ontology-based Features | The Tree Augmented Naive Bayes classifier is a type of probabilistic
graphical model that can represent some feature dependencies. In this work, we
propose a Hierarchical Redundancy Eliminated Tree Augmented Naive Bayes
(HRE-TAN) algorithm, which considers removing the hierarchical redundancy
during the classifier learning process, when coping with data containing
hierarchically structured features. The experiments showed that HRE-TAN obtains
significantly better predictive performance than the conventional Tree
Augmented Naive Bayes classifier, and enhanced the robustness against
imbalanced class distributions, in aging-related gene datasets with Gene
Ontology terms used as features.
|
Deep CORAL: Correlation Alignment for Deep Domain Adaptation | Deep neural networks are able to learn powerful representations from large
quantities of labeled input data, however they cannot always generalize well
across changes in input distributions. Domain adaptation algorithms have been
proposed to compensate for the degradation in performance due to domain shift.
In this paper, we address the case when the target domain is unlabeled,
requiring unsupervised adaptation. CORAL is a "frustratingly easy" unsupervised
domain adaptation method that aligns the second-order statistics of the source
and target distributions with a linear transformation. Here, we extend CORAL to
learn a nonlinear transformation that aligns correlations of layer activations
in deep neural networks (Deep CORAL). Experiments on standard benchmark
datasets show state-of-the-art performance.
|
Finding Significant Fourier Coefficients: Clarifications,
Simplifications, Applications and Limitations | Ideas from Fourier analysis have been used in cryptography for the last three
decades. Akavia, Goldwasser and Safra unified some of these ideas to give a
complete algorithm that finds significant Fourier coefficients of functions on
any finite abelian group. Their algorithm stimulated a lot of interest in the
cryptography community, especially in the context of `bit security'. This
manuscript attempts to be a friendly and comprehensive guide to the tools and
results in this field. The intended readership is cryptographers who have heard
about these tools and seek an understanding of their mechanics and their
usefulness and limitations. A compact overview of the algorithm is presented
with emphasis on the ideas behind it. We show how these ideas can be extended
to a `modulus-switching' variant of the algorithm. We survey some applications
of this algorithm, and explain that several results should be taken in the
right context. In particular, we point out that some of the most important bit
security problems are still open. Our original contributions include: a
discussion of the limitations on the usefulness of these tools; an answer to an
open question about the modular inversion hidden number problem.
|
Stock trend prediction using news sentiment analysis | Efficient Market Hypothesis is the popular theory about stock prediction.
With its failure much research has been carried in the area of prediction of
stocks. This project is about taking non quantifiable data such as financial
news articles about a company and predicting its future stock trend with news
sentiment classification. Assuming that news articles have impact on stock
market, this is an attempt to study relationship between news and stock trend.
To show this, we created three different classification models which depict
polarity of news articles being positive or negative. Observations show that RF
and SVM perform well in all types of testing. Na\"ive Bayes gives good result
but not compared to the other two. Experiments are conducted to evaluate
various aspects of the proposed model and encouraging results are obtained in
all of the experiments. The accuracy of the prediction model is more than 80%
and in comparison with news random labeling with 50% of accuracy; the model has
increased the accuracy by 30%.
|
Sequence Training and Adaptation of Highway Deep Neural Networks | Highway deep neural network (HDNN) is a type of depth-gated feedforward
neural network, which has shown to be easier to train with more hidden layers
and also generalise better compared to conventional plain deep neural networks
(DNNs). Previously, we investigated a structured HDNN architecture for speech
recognition, in which the two gate functions were tied across all the hidden
layers, and we were able to train a much smaller model without sacrificing the
recognition accuracy. In this paper, we carry on the study of this architecture
with sequence-discriminative training criterion and speaker adaptation
techniques on the AMI meeting speech recognition corpus. We show that these two
techniques improve speech recognition accuracy on top of the model trained with
the cross entropy criterion. Furthermore, we demonstrate that the two gate
functions that are tied across all the hidden layers are able to control the
information flow over the whole network, and we can achieve considerable
improvements by only updating these gate functions in both sequence training
and adaptation experiments.
|
Nesterov's Accelerated Gradient and Momentum as approximations to
Regularised Update Descent | We present a unifying framework for adapting the update direction in
gradient-based iterative optimization methods. As natural special cases we
re-derive classical momentum and Nesterov's accelerated gradient method,
lending a new intuitive interpretation to the latter algorithm. We show that a
new algorithm, which we term Regularised Gradient Descent, can converge more
quickly than either Nesterov's algorithm or the classical momentum algorithm.
|
Mini-Batch Spectral Clustering | The cost of computing the spectrum of Laplacian matrices hinders the
application of spectral clustering to large data sets. While approximations
recover computational tractability, they can potentially affect clustering
performance. This paper proposes a practical approach to learn spectral
clustering based on adaptive stochastic gradient optimization. Crucially, the
proposed approach recovers the exact spectrum of Laplacian matrices in the
limit of the iterations, and the cost of each iteration is linear in the number
of samples. Extensive experimental validation on data sets with up to half a
million samples demonstrate its scalability and its ability to outperform
state-of-the-art approximate methods to learn spectral clustering for a given
computational budget.
|
Artificial neural networks and fuzzy logic for recognizing alphabet
characters and mathematical symbols | Optical Character Recognition software (OCR) are important tools for
obtaining accessible texts. We propose the use of artificial neural networks
(ANN) in order to develop pattern recognition algorithms capable of recognizing
both normal texts and formulae. We present an original improvement of the
backpropagation algorithm. Moreover, we describe a novel image segmentation
algorithm that exploits fuzzy logic for separating touching characters.
|
DeepChrome: Deep-learning for predicting gene expression from histone
modifications | Motivation: Histone modifications are among the most important factors that
control gene regulation. Computational methods that predict gene expression
from histone modification signals are highly desirable for understanding their
combinatorial effects in gene regulation. This knowledge can help in developing
'epigenetic drugs' for diseases like cancer. Previous studies for quantifying
the relationship between histone modifications and gene expression levels
either failed to capture combinatorial effects or relied on multiple methods
that separate predictions and combinatorial analysis. This paper develops a
unified discriminative framework using a deep convolutional neural network to
classify gene expression using histone modification data as input. Our system,
called DeepChrome, allows automatic extraction of complex interactions among
important features. To simultaneously visualize the combinatorial interactions
among histone modifications, we propose a novel optimization-based technique
that generates feature pattern maps from the learnt deep model. This provides
an intuitive description of underlying epigenetic mechanisms that regulate
genes. Results: We show that DeepChrome outperforms state-of-the-art models
like Support Vector Machines and Random Forests for gene expression
classification task on 56 different cell-types from REMC database. The output
of our visualization technique not only validates the previous observations but
also allows novel insights about combinatorial interactions among histone
modification marks, some of which have recently been observed by experimental
studies.
|
Single-Channel Multi-Speaker Separation using Deep Clustering | Deep clustering is a recently introduced deep learning architecture that uses
discriminatively trained embeddings as the basis for clustering. It was
recently applied to spectrogram segmentation, resulting in impressive results
on speaker-independent multi-speaker separation. In this paper we extend the
baseline system with an end-to-end signal approximation objective that greatly
improves performance on a challenging speech separation. We first significantly
improve upon the baseline system performance by incorporating better
regularization, larger temporal context, and a deeper architecture, culminating
in an overall improvement in signal to distortion ratio (SDR) of 10.3 dB
compared to the baseline of 6.0 dB for two-speaker separation, as well as a 7.1
dB SDR improvement for three-speaker separation. We then extend the model to
incorporate an enhancement layer to refine the signal estimates, and perform
end-to-end training through both the clustering and enhancement stages to
maximize signal fidelity. We evaluate the results using automatic speech
recognition. The new signal approximation objective, combined with end-to-end
training, produces unprecedented performance, reducing the word error rate
(WER) from 89.1% down to 30.8%. This represents a major advancement towards
solving the cocktail party problem.
|
Applying Deep Learning to the Newsvendor Problem | The newsvendor problem is one of the most basic and widely applied inventory
models. There are numerous extensions of this problem. If the probability
distribution of the demand is known, the problem can be solved analytically.
However, approximating the probability distribution is not easy and is prone
to error; therefore, the resulting solution to the newsvendor problem may be
not optimal. To address this issue, we propose an algorithm based on deep
learning that optimizes the order quantities for all products based on features
of the demand data. Our algorithm integrates the forecasting and
inventory-optimization steps, rather than solving them separately, as is
typically done, and does not require knowledge of the probability distributions
of the demand. Numerical experiments on real-world data suggest that our
algorithm outperforms other approaches, including data-driven and machine
learning approaches, especially for demands with high volatility. Finally, in
order to show how this approach can be used for other inventory optimization
problems, we provide an extension for (r,Q) policies.
|
Overcoming Challenges in Fixed Point Training of Deep Convolutional
Networks | It is known that training deep neural networks, in particular, deep
convolutional networks, with aggressively reduced numerical precision is
challenging. The stochastic gradient descent algorithm becomes unstable in the
presence of noisy gradient updates resulting from arithmetic with limited
numeric precision. One of the well-accepted solutions facilitating the training
of low precision fixed point networks is stochastic rounding. However, to the
best of our knowledge, the source of the instability in training neural
networks with noisy gradient updates has not been well investigated. This work
is an attempt to draw a theoretical connection between low numerical precision
and training algorithm stability. In doing so, we will also propose and verify
through experiments methods that are able to improve the training performance
of deep convolutional networks in fixed point.
|
CNN-LTE: a Class of 1-X Pooling Convolutional Neural Networks on Label
Tree Embeddings for Audio Scene Recognition | We describe in this report our audio scene recognition system submitted to
the DCASE 2016 challenge. Firstly, given the label set of the scenes, a label
tree is automatically constructed. This category taxonomy is then used in the
feature extraction step in which an audio scene instance is represented by a
label tree embedding image. Different convolutional neural networks, which are
tailored for the task at hand, are finally learned on top of the image features
for scene recognition. Our system reaches an overall recognition accuracy of
81.2% and 83.3% and outperforms the DCASE 2016 baseline with absolute
improvements of 8.7% and 6.1% on the development and test data, respectively.
|
CaR-FOREST: Joint Classification-Regression Decision Forests for
Overlapping Audio Event Detection | This report describes our submissions to Task2 and Task3 of the DCASE 2016
challenge. The systems aim at dealing with the detection of overlapping audio
events in continuous streams, where the detectors are based on random decision
forests. The proposed forests are jointly trained for classification and
regression simultaneously. Initially, the training is classification-oriented
to encourage the trees to select discriminative features from overlapping
mixtures to separate positive audio segments from the negative ones. The
regression phase is then carried out to let the positive audio segments vote
for the event onsets and offsets, and therefore model the temporal structure of
audio events. One random decision forest is specifically trained for each event
category of interest. Experimental results on the development data show that
our systems significantly outperform the baseline on the Task2 evaluation while
they are inferior to the baseline in the Task3 evaluation.
|
Collaborative Training of Tensors for Compositional Distributional
Semantics | Type-based compositional distributional semantic models present an
interesting line of research into functional representations of linguistic
meaning. One of the drawbacks of such models, however, is the lack of training
data required to train each word-type combination. In this paper we address
this by introducing training methods that share parameters between similar
words. We show that these methods enable zero-shot learning for words that have
no training data at all, as well as enabling construction of high-quality
tensors from very few training examples per word.
|
Watch This: Scalable Cost-Function Learning for Path Planning in Urban
Environments | In this work, we present an approach to learn cost maps for driving in
complex urban environments from a very large number of demonstrations of
driving behaviour by human experts. The learned cost maps are constructed
directly from raw sensor measurements, bypassing the effort of manually
designing cost maps as well as features. When deploying the learned cost maps,
the trajectories generated not only replicate human-like driving behaviour but
are also demonstrably robust against systematic errors in putative robot
configuration. To achieve this we deploy a Maximum Entropy based, non-linear
IRL framework which uses Fully Convolutional Neural Networks (FCNs) to
represent the cost model underlying expert driving behaviour. Using a deep,
parametric approach enables us to scale efficiently to large datasets and
complex behaviours by being run-time independent of dataset extent during
deployment. We demonstrate the scalability and the performance of the proposed
approach on an ambitious dataset collected over the course of one year
including more than 25k demonstration trajectories extracted from over 120km of
driving around pedestrianised areas in the city of Milton Keynes, UK. We
evaluate the resulting cost representations by showing the advantages over a
carefully manually designed cost map and, in addition, demonstrate its
robustness to systematic errors by learning precise cost-maps even in the
presence of system calibration perturbations.
|
Lower Bounds on Active Learning for Graphical Model Selection | We consider the problem of estimating the underlying graph associated with a
Markov random field, with the added twist that the decoding algorithm can
iteratively choose which subsets of nodes to sample based on the previous
samples, resulting in an active learning setting. Considering both Ising and
Gaussian models, we provide algorithm-independent lower bounds for
high-probability recovery within the class of degree-bounded graphs. Our main
results are minimax lower bounds for the active setting that match the best
known lower bounds for the passive setting, which in turn are known to be tight
in several cases of interest. Our analysis is based on Fano's inequality, along
with novel mutual information bounds for the active learning setting, and the
application of restricted graph ensembles. While we consider ensembles that are
similar or identical to those used in the passive setting, we require different
analysis techniques, with a key challenge being bounding a mutual information
quantity associated with observed subsets of nodes, as opposed to full
observations.
|
Explaining Deep Convolutional Neural Networks on Music Classification | Deep convolutional neural networks (CNNs) have been actively adopted in the
field of music information retrieval, e.g. genre classification, mood
detection, and chord recognition. However, the process of learning and
prediction is little understood, particularly when it is applied to
spectrograms. We introduce auralisation of a CNN to understand its underlying
mechanism, which is based on a deconvolution procedure introduced in [2].
Auralisation of a CNN is converting the learned convolutional features that are
obtained from deconvolution into audio signals. In the experiments and
discussions, we explain trained features of a 5-layer CNN based on the
deconvolved spectrograms and auralised signals. The pairwise correlations per
layers with varying different musical attributes are also investigated to
understand the evolution of the learnt features. It is shown that in the deep
layers, the features are learnt to capture textures, the patterns of continuous
distributions, rather than shapes of lines.
|
Proceedings of the 2016 ICML Workshop on #Data4Good: Machine Learning in
Social Good Applications | This is the Proceedings of the ICML Workshop on #Data4Good: Machine Learning
in Social Good Applications, which was held on June 24, 2016 in New York.
|
Log-Linear RNNs: Towards Recurrent Neural Networks with Flexible Prior
Knowledge | We introduce LL-RNNs (Log-Linear RNNs), an extension of Recurrent Neural
Networks that replaces the softmax output layer by a log-linear output layer,
of which the softmax is a special case. This conceptually simple move has two
main advantages. First, it allows the learner to combat training data sparsity
by allowing it to model words (or more generally, output symbols) as complex
combinations of attributes without requiring that each combination is directly
observed in the training data (as the softmax does). Second, it permits the
inclusion of flexible prior knowledge in the form of a priori specified modular
features, where the neural network component learns to dynamically control the
weights of a log-linear distribution exploiting these features.
We conduct experiments in the domain of language modelling of French, that
exploit morphological prior knowledge and show an important decrease in
perplexity relative to a baseline RNN.
We provide other motivating iillustrations, and finally argue that the
log-linear and the neural-network components contribute complementary strengths
to the LL-RNN: the LL aspect allows the model to incorporate rich prior
knowledge, while the NN aspect, according to the "representation learning"
paradigm, allows the model to discover novel combination of characteristics.
|
Adjusting for Dropout Variance in Batch Normalization and Weight
Initialization | We show how to adjust for the variance introduced by dropout with corrections
to weight initialization and Batch Normalization, yielding higher accuracy.
Though dropout can preserve the expected input to a neuron between train and
test, the variance of the input differs. We thus propose a new weight
initialization by correcting for the influence of dropout rates and an
arbitrary nonlinearity's influence on variance through simple corrective
scalars. Since Batch Normalization trained with dropout estimates the variance
of a layer's incoming distribution with some inputs dropped, the variance also
differs between train and test. After training a network with Batch
Normalization and dropout, we simply update Batch Normalization's variance
moving averages with dropout off and obtain state of the art on CIFAR-10 and
CIFAR-100 without data augmentation.
|
Proceedings of the 2016 ICML Workshop on Human Interpretability in
Machine Learning (WHI 2016) | This is the Proceedings of the 2016 ICML Workshop on Human Interpretability
in Machine Learning (WHI 2016), which was held in New York, NY, June 23, 2016.
Invited speakers were Susan Athey, Rich Caruana, Jacob Feldman, Percy Liang,
and Hanna Wallach.
|
Adversarial examples in the physical world | Most existing machine learning classifiers are highly vulnerable to
adversarial examples. An adversarial example is a sample of input data which
has been modified very slightly in a way that is intended to cause a machine
learning classifier to misclassify it. In many cases, these modifications can
be so subtle that a human observer does not even notice the modification at
all, yet the classifier still makes a mistake. Adversarial examples pose
security concerns because they could be used to perform an attack on machine
learning systems, even if the adversary has no access to the underlying model.
Up to now, all previous work have assumed a threat model in which the adversary
can feed data directly into the machine learning classifier. This is not always
the case for systems operating in the physical world, for example those which
are using signals from cameras and other sensors as an input. This paper shows
that even in such physical world scenarios, machine learning systems are
vulnerable to adversarial examples. We demonstrate this by feeding adversarial
images obtained from cell-phone camera to an ImageNet Inception classifier and
measuring the classification accuracy of the system. We find that a large
fraction of adversarial examples are classified incorrectly even when perceived
through the camera.
|
Learning from Multiway Data: Simple and Efficient Tensor Regression | Tensor regression has shown to be advantageous in learning tasks with
multi-directional relatedness. Given massive multiway data, traditional methods
are often too slow to operate on or suffer from memory bottleneck. In this
paper, we introduce subsampled tensor projected gradient to solve the problem.
Our algorithm is impressively simple and efficient. It is built upon projected
gradient method with fast tensor power iterations, leveraging randomized
sketching for further acceleration. Theoretical analysis shows that our
algorithm converges to the correct solution in fixed number of iterations. The
memory requirement grows linearly with the size of the problem. We demonstrate
superior empirical performance on both multi-linear multi-task learning and
spatio-temporal applications.
|
Online Learning Schemes for Power Allocation in Energy Harvesting
Communications | We consider the problem of power allocation over a time-varying channel with
unknown distribution in energy harvesting communication systems. In this
problem, the transmitter has to choose the transmit power based on the amount
of stored energy in its battery with the goal of maximizing the average rate
obtained over time. We model this problem as a Markov decision process (MDP)
with the transmitter as the agent, the battery status as the state, the
transmit power as the action and the rate obtained as the reward. The average
reward maximization problem over the MDP can be solved by a linear program (LP)
that uses the transition probabilities for the state-action pairs and their
reward values to choose a power allocation policy. Since the rewards associated
the state-action pairs are unknown, we propose two online learning algorithms:
UCLP and Epoch-UCLP that learn these rewards and adapt their policies along the
way. The UCLP algorithm solves the LP at each step to decide its current policy
using the upper confidence bounds on the rewards, while the Epoch-UCLP
algorithm divides the time into epochs, solves the LP only at the beginning of
the epochs and follows the obtained policy in that epoch. We prove that the
reward losses or regrets incurred by both these algorithms are upper bounded by
constants. Epoch-UCLP incurs a higher regret compared to UCLP, but reduces the
computational requirements substantially. We also show that the presented
algorithms work for online learning in cost minimization problems like the
packet scheduling with power-delay tradeoff with minor changes.
|
Uncovering Locally Discriminative Structure for Feature Analysis | Manifold structure learning is often used to exploit geometric information
among data in semi-supervised feature learning algorithms. In this paper, we
find that local discriminative information is also of importance for
semi-supervised feature learning. We propose a method that utilizes both the
manifold structure of data and local discriminant information. Specifically, we
define a local clique for each data point. The k-Nearest Neighbors (kNN) is
used to determine the structural information within each clique. We then employ
a variant of Fisher criterion model to each clique for local discriminant
evaluation and sum all cliques as global integration into the framework. In
this way, local discriminant information is embedded. Labels are also utilized
to minimize distances between data from the same class. In addition, we use the
kernel method to extend our proposed model and facilitate feature learning in a
high-dimensional space after feature mapping. Experimental results show that
our method is superior to all other compared methods over a number of datasets.
|
Visual Dynamics: Probabilistic Future Frame Synthesis via Cross
Convolutional Networks | We study the problem of synthesizing a number of likely future frames from a
single input image. In contrast to traditional methods, which have tackled this
problem in a deterministic or non-parametric way, we propose a novel approach
that models future frames in a probabilistic manner. Our probabilistic model
makes it possible for us to sample and synthesize many possible future frames
from a single input image. Future frame synthesis is challenging, as it
involves low- and high-level image and motion understanding. We propose a novel
network structure, namely a Cross Convolutional Network to aid in synthesizing
future frames; this network structure encodes image and motion information as
feature maps and convolutional kernels, respectively. In experiments, our model
performs well on synthetic data, such as 2D shapes and animated game sprites,
as well as on real-wold videos. We also show that our model can be applied to
tasks such as visual analogy-making, and present an analysis of the learned
network representations.
|
Classifier Risk Estimation under Limited Labeling Resources | In this paper we propose strategies for estimating performance of a
classifier when labels cannot be obtained for the whole test set. The number of
test instances which can be labeled is very small compared to the whole test
data size. The goal then is to obtain a precise estimate of classifier
performance using as little labeling resource as possible. Specifically, we try
to answer, how to select a subset of the large test set for labeling such that
the performance of a classifier estimated on this subset is as close as
possible to the one on the whole test set. We propose strategies based on
stratified sampling for selecting this subset. We show that these strategies
can reduce the variance in estimation of classifier accuracy by a significant
amount compared to simple random sampling (over 65% in several cases). Hence,
our proposed methods are much more precise compared to random sampling for
accuracy estimation under restricted labeling resources. The reduction in
number of samples required (compared to random sampling) to estimate the
classifier accuracy with only 1% error is high as 60% in some cases.
|
Dealing with Class Imbalance using Thresholding | We propose thresholding as an approach to deal with class imbalance. We
define the concept of thresholding as a process of determining a decision
boundary in the presence of a tunable parameter. The threshold is the maximum
value of this tunable parameter where the conditions of a certain decision are
satisfied. We show that thresholding is applicable not only for linear
classifiers but also for non-linear classifiers. We show that this is the
implicit assumption for many approaches to deal with class imbalance in linear
classifiers. We then extend this paradigm beyond linear classification and show
how non-linear classification can be dealt with under this umbrella framework
of thresholding. The proposed method can be used for outlier detection in many
real-life scenarios like in manufacturing. In advanced manufacturing units,
where the manufacturing process has matured over time, the number of instances
(or parts) of the product that need to be rejected (based on a strict regime of
quality tests) becomes relatively rare and are defined as outliers. How to
detect these rare parts or outliers beforehand? How to detect combination of
conditions leading to these outliers? These are the questions motivating our
research. This paper focuses on prediction of outliers and conditions leading
to outliers using classification. We address the problem of outlier detection
using classification. The classes are good parts (those passing the quality
tests) and bad parts (those failing the quality tests and can be considered as
outliers). The rarity of outliers transforms this problem into a
class-imbalanced classification problem.
|
How to Allocate Resources For Features Acquisition? | We study classification problems where features are corrupted by noise and
where the magnitude of the noise in each feature is influenced by the resources
allocated to its acquisition. This is the case, for example, when multiple
sensors share a common resource (power, bandwidth, attention, etc.). We develop
a method for computing the optimal resource allocation for a variety of
scenarios and derive theoretical bounds concerning the benefit that may arise
by non-uniform allocation. We further demonstrate the effectiveness of the
developed method in simulations.
|
On Faster Convergence of Cyclic Block Coordinate Descent-type Methods
for Strongly Convex Minimization | The cyclic block coordinate descent-type (CBCD-type) methods, which performs
iterative updates for a few coordinates (a block) simultaneously throughout the
procedure, have shown remarkable computational performance for solving strongly
convex minimization problems. Typical applications include many popular
statistical machine learning methods such as elastic-net regression, ridge
penalized logistic regression, and sparse additive regression. Existing
optimization literature has shown that for strongly convex minimization, the
CBCD-type methods attain iteration complexity of
$\mathcal{O}(p\log(1/\epsilon))$, where $\epsilon$ is a pre-specified accuracy
of the objective value, and $p$ is the number of blocks. However, such
iteration complexity explicitly depends on $p$, and therefore is at least $p$
times worse than the complexity $\mathcal{O}(\log(1/\epsilon))$ of gradient
descent (GD) methods. To bridge this theoretical gap, we propose an improved
convergence analysis for the CBCD-type methods. In particular, we first show
that for a family of quadratic minimization problems, the iteration complexity
$\mathcal{O}(\log^2(p)\cdot\log(1/\epsilon))$ of the CBCD-type methods matches
that of the GD methods in term of dependency on $p$, up to a $\log^2 p$ factor.
Thus our complexity bounds are sharper than the existing bounds by at least a
factor of $p/\log^2(p)$. We also provide a lower bound to confirm that our
improved complexity bounds are tight (up to a $\log^2 (p)$ factor), under the
assumption that the largest and smallest eigenvalues of the Hessian matrix do
not scale with $p$. Finally, we generalize our analysis to other strongly
convex minimization problems beyond quadratic ones.
|
Tight Lower Bounds for Multiplicative Weights Algorithmic Families | We study the fundamental problem of prediction with expert advice and develop
regret lower bounds for a large family of algorithms for this problem. We
develop simple adversarial primitives, that lend themselves to various
combinations leading to sharp lower bounds for many algorithmic families. We
use these primitives to show that the classic Multiplicative Weights Algorithm
(MWA) has a regret of $\sqrt{\frac{T \ln k}{2}}$, there by completely closing
the gap between upper and lower bounds. We further show a regret lower bound of
$\frac{2}{3}\sqrt{\frac{T\ln k}{2}}$ for a much more general family of
algorithms than MWA, where the learning rate can be arbitrarily varied over
time, or even picked from arbitrary distributions over time. We also use our
primitives to construct adversaries in the geometric horizon setting for MWA to
precisely characterize the regret at $\frac{0.391}{\sqrt{\delta}}$ for the case
of $2$ experts and a lower bound of $\frac{1}{2}\sqrt{\frac{\ln k}{2\delta}}$
for the case of arbitrary number of experts $k$.
|
Classifying Variable-Length Audio Files with All-Convolutional Networks
and Masked Global Pooling | We trained a deep all-convolutional neural network with masked global pooling
to perform single-label classification for acoustic scene classification and
multi-label classification for domestic audio tagging in the DCASE-2016
contest. Our network achieved an average accuracy of 84.5% on the four-fold
cross-validation for acoustic scene recognition, compared to the provided
baseline of 72.5%, and an average equal error rate of 0.17 for domestic audio
tagging, compared to the baseline of 0.21. The network therefore improves the
baselines by a relative amount of 17% and 19%, respectively. The network only
consists of convolutional layers to extract features from the short-time
Fourier transform and one global pooling layer to combine those features. It
particularly possesses neither fully-connected layers, besides the
fully-connected output layer, nor dropout layers.
|
Incremental Factorization Machines for Persistently Cold-starting Online
Item Recommendation | Real-world item recommenders commonly suffer from a persistent cold-start
problem which is caused by dynamically changing users and items. In order to
overcome the problem, several context-aware recommendation techniques have been
recently proposed. In terms of both feasibility and performance, factorization
machine (FM) is one of the most promising methods as generalization of the
conventional matrix factorization techniques. However, since online algorithms
are suitable for dynamic data, the static FMs are still inadequate. Thus, this
paper proposes incremental FMs (iFMs), a general online factorization
framework, and specially extends iFMs into an online item recommender. The
proposed framework can be a promising baseline for further development of the
production recommender systems. Evaluation is done empirically both on
synthetic and real-world unstable datasets.
|
From Behavior to Sparse Graphical Games: Efficient Recovery of
Equilibria | In this paper we study the problem of exact recovery of the pure-strategy
Nash equilibria (PSNE) set of a graphical game from noisy observations of joint
actions of the players alone. We consider sparse linear influence games --- a
parametric class of graphical games with linear payoffs, and represented by
directed graphs of n nodes (players) and in-degree of at most k. We present an
$\ell_1$-regularized logistic regression based algorithm for recovering the
PSNE set exactly, that is both computationally efficient --- i.e. runs in
polynomial time --- and statistically efficient --- i.e. has logarithmic sample
complexity. Specifically, we show that the sufficient number of samples
required for exact PSNE recovery scales as $\mathcal{O}(\mathrm{poly}(k) \log
n)$. We also validate our theoretical results using synthetic experiments.
|
Learning a metric for class-conditional KNN | Naive Bayes Nearest Neighbour (NBNN) is a simple and effective framework
which addresses many of the pitfalls of K-Nearest Neighbour (KNN)
classification. It has yielded competitive results on several computer vision
benchmarks. Its central tenet is that during NN search, a query is not compared
to every example in a database, ignoring class information. Instead, NN
searches are performed within each class, generating a score per class. A key
problem with NN techniques, including NBNN, is that they fail when the data
representation does not capture perceptual (e.g.~class-based) similarity. NBNN
circumvents this by using independent engineered descriptors (e.g.~SIFT). To
extend its applicability outside of image-based domains, we propose to learn a
metric which captures perceptual similarity. Similar to how Neighbourhood
Components Analysis optimizes a differentiable form of KNN classification, we
propose "Class Conditional" metric learning (CCML), which optimizes a soft form
of the NBNN selection rule. Typical metric learning algorithms learn either a
global or local metric. However, our proposed method can be adjusted to a
particular level of locality by tuning a single parameter. An empirical
evaluation on classification and retrieval tasks demonstrates that our proposed
method clearly outperforms existing learned distance metrics across a variety
of image and non-image datasets.
|
Proximal Quasi-Newton Methods for Regularized Convex Optimization with
Linear and Accelerated Sublinear Convergence Rates | In [19], a general, inexact, efficient proximal quasi-Newton algorithm for
composite optimization problems has been proposed and a sublinear global
convergence rate has been established. In this paper, we analyze the
convergence properties of this method, both in the exact and inexact setting,
in the case when the objective function is strongly convex. We also investigate
a practical variant of this method by establishing a simple stopping criterion
for the subproblem optimization. Furthermore, we consider an accelerated
variant, based on FISTA [1], to the proximal quasi-Newton algorithm. A similar
accelerated method has been considered in [7], where the convergence rate
analysis relies on very strong impractical assumptions. We present a modified
analysis while relaxing these assumptions and perform a practical comparison of
the accelerated proximal quasi- Newton algorithm and the regular one. Our
analysis and computational results show that acceleration may not bring any
benefit in the quasi-Newton setting.
|
Kernel-based methods for bandit convex optimization | We consider the adversarial convex bandit problem and we build the first
$\mathrm{poly}(T)$-time algorithm with $\mathrm{poly}(n) \sqrt{T}$-regret for
this problem. To do so we introduce three new ideas in the derivative-free
optimization literature: (i) kernel methods, (ii) a generalization of Bernoulli
convolutions, and (iii) a new annealing schedule for exponential weights (with
increasing learning rate). The basic version of our algorithm achieves
$\tilde{O}(n^{9.5} \sqrt{T})$-regret, and we show that a simple variant of this
algorithm can be run in $\mathrm{poly}(n \log(T))$-time per step at the cost of
an additional $\mathrm{poly}(n) T^{o(1)}$ factor in the regret. These results
improve upon the $\tilde{O}(n^{11} \sqrt{T})$-regret and
$\exp(\mathrm{poly}(T))$-time result of the first two authors, and the
$\log(T)^{\mathrm{poly}(n)} \sqrt{T}$-regret and
$\log(T)^{\mathrm{poly}(n)}$-time result of Hazan and Li. Furthermore we
conjecture that another variant of the algorithm could achieve
$\tilde{O}(n^{1.5} \sqrt{T})$-regret, and moreover that this regret is
unimprovable (the current best lower bound being $\Omega(n \sqrt{T})$ and it is
achieved with linear functions). For the simpler situation of zeroth order
stochastic convex optimization this corresponds to the conjecture that the
optimal query complexity is of order $n^3 / \epsilon^2$.
|
Recurrent Memory Array Structures | The following report introduces ideas augmenting standard Long Short Term
Memory (LSTM) architecture with multiple memory cells per hidden unit in order
to improve its generalization capabilities. It considers both deterministic and
stochastic variants of memory operation. It is shown that the nondeterministic
Array-LSTM approach improves state-of-the-art performance on character level
text prediction achieving 1.402 BPC on enwik8 dataset. Furthermore, this report
estabilishes baseline neural-based results of 1.12 BPC and 1.19 BPC for enwik9
and enwik10 datasets respectively.
|
Stream-based Online Active Learning in a Contextual Multi-Armed Bandit
Framework | We study the stream-based online active learning in a contextual multi-armed
bandit framework. In this framework, the reward depends on both the arm and the
context. In a stream-based active learning setting, obtaining the ground truth
of the reward is costly, and the conventional contextual multi-armed bandit
algorithm fails to achieve a sublinear regret due to this cost. Hence, the
algorithm needs to determine whether or not to request the ground truth of the
reward at current time slot. In our framework, we consider a stream-based
active learning setting in which a query request for the ground truth is sent
to the annotator, together with some prior information of the ground truth.
Depending on the accuracy of the prior information, the query cost varies. Our
algorithm mainly carries out two operations: the refinement of the context and
arm spaces and the selection of actions. In our algorithm, the partitions of
the context space and the arm space are maintained for a certain time slots,
and then become finer as more information about the rewards accumulates. We use
a strategic way to select the arms and to request the ground truth of the
reward, aiming to maximize the total reward. We analytically show that the
regret is sublinear and in the same order with that of the conventional
contextual multi-armed bandit algorithms, where no query cost
|
How to calculate partition functions using convex programming
hierarchies: provable bounds for variational methods | We consider the problem of approximating partition functions for Ising
models. We make use of recent tools in combinatorial optimization: the
Sherali-Adams and Lasserre convex programming hierarchies, in combination with
variational methods to get algorithms for calculating partition functions in
these families. These techniques give new, non-trivial approximation guarantees
for the partition function beyond the regime of correlation decay. They also
generalize some classical results from statistical physics about the
Curie-Weiss ferromagnetic Ising model, as well as provide a partition function
counterpart of classical results about max-cut on dense graphs
\cite{arora1995polynomial}. With this, we connect techniques from two
apparently disparate research areas -- optimization and counting/partition
function approximations. (i.e. \#-P type of problems).
Furthermore, we design to the best of our knowledge the first provable,
convex variational methods. Though in the literature there are a host of convex
versions of variational methods \cite{wainwright2003tree, wainwright2005new,
heskes2006convexity, meshi2009convexifying}, they come with no guarantees
(apart from some extremely special cases, like e.g. the graph has a single
cycle \cite{weiss2000correctness}). We consider dense and low threshold rank
graphs, and interestingly, the reason our approach works on these types of
graphs is because local correlations propagate to global correlations --
completely the opposite of algorithms based on correlation decay. In the
process we design novel entropy approximations based on the low-order moments
of a distribution.
Our proof techniques are very simple and generic, and likely to be applicable
to many other settings other than Ising models.
|
On Deterministic Conditions for Subspace Clustering under Missing Data | In this paper we present deterministic conditions for success of sparse
subspace clustering (SSC) under missing data, when data is assumed to come from
a Union of Subspaces (UoS) model. We consider two algorithms, which are
variants of SSC with entry-wise zero-filling that differ in terms of the
optimization problems used to find affinity matrix for spectral clustering. For
both the algorithms, we provide deterministic conditions for any pattern of
missing data such that perfect clustering can be achieved. We provide extensive
sets of simulation results for clustering as well as completion of data at
missing entries, under the UoS model. Our experimental results indicate that in
contrast to the full data case, accurate clustering does not imply accurate
subspace identification and completion, indicating the natural order of
relative hardness of these problems.
|
Multi-Step Bayesian Optimization for One-Dimensional Feasibility
Determination | Bayesian optimization methods allocate limited sampling budgets to maximize
expensive-to-evaluate functions. One-step-lookahead policies are often used,
but computing optimal multi-step-lookahead policies remains a challenge. We
consider a specialized Bayesian optimization problem: finding the superlevel
set of an expensive one-dimensional function, with a Markov process prior. We
compute the Bayes-optimal sampling policy efficiently, and characterize the
suboptimality of one-step lookahead. Our numerical experiments demonstrate that
the one-step lookahead policy is close to optimal in this problem, performing
within 98% of optimal in the experimental settings considered.
|
Information Projection and Approximate Inference for Structured Sparse
Variables | Approximate inference via information projection has been recently introduced
as a general-purpose approach for efficient probabilistic inference given
sparse variables. This manuscript goes beyond classical sparsity by proposing
efficient algorithms for approximate inference via information projection that
are applicable to any structure on the set of variables that admits enumeration
using a \emph{matroid}. We show that the resulting information projection can
be reduced to combinatorial submodular optimization subject to matroid
constraints. Further, leveraging recent advances in submodular optimization, we
provide an efficient greedy algorithm with strong optimization-theoretic
guarantees. The class of probabilistic models that can be expressed in this way
is quite broad and, as we show, includes group sparse regression, group sparse
principal components analysis and sparse canonical correlation analysis, among
others. Moreover, empirical results on simulated data and high dimensional
neuroimaging data highlight the superior performance of the information
projection approach as compared to established baselines for a range of
probabilistic models.
|
Network Trimming: A Data-Driven Neuron Pruning Approach towards
Efficient Deep Architectures | State-of-the-art neural networks are getting deeper and wider. While their
performance increases with the increasing number of layers and neurons, it is
crucial to design an efficient deep architecture in order to reduce
computational and memory costs. Designing an efficient neural network, however,
is labor intensive requiring many experiments, and fine-tunings. In this paper,
we introduce network trimming which iteratively optimizes the network by
pruning unimportant neurons based on analysis of their outputs on a large
dataset. Our algorithm is inspired by an observation that the outputs of a
significant portion of neurons in a large network are mostly zero, regardless
of what inputs the network received. These zero activation neurons are
redundant, and can be removed without affecting the overall accuracy of the
network. After pruning the zero activation neurons, we retrain the network
using the weights before pruning as initialization. We alternate the pruning
and retraining to further reduce zero activations in a network. Our experiments
on the LeNet and VGG-16 show that we can achieve high compression ratio of
parameters without losing or even achieving higher accuracy than the original
network.
|
Predicting the evolution of stationary graph signals | An emerging way of tackling the dimensionality issues arising in the modeling
of a multivariate process is to assume that the inherent data structure can be
captured by a graph. Nevertheless, though state-of-the-art graph-based methods
have been successful for many learning tasks, they do not consider
time-evolving signals and thus are not suitable for prediction. Based on the
recently introduced joint stationarity framework for time-vertex processes,
this letter considers multivariate models that exploit the graph topology so as
to facilitate the prediction. The resulting method yields similar accuracy to
the joint (time-graph) mean-squared error estimator but at lower complexity,
and outperforms purely time-based methods.
|
DeepBinaryMask: Learning a Binary Mask for Video Compressive Sensing | In this paper, we propose a novel encoder-decoder neural network model
referred to as DeepBinaryMask for video compressive sensing. In video
compressive sensing one frame is acquired using a set of coded masks (sensing
matrix) from which a number of video frames is reconstructed, equal to the
number of coded masks. The proposed framework is an end-to-end model where the
sensing matrix is trained along with the video reconstruction. The encoder
learns the binary elements of the sensing matrix and the decoder is trained to
recover the unknown video sequence. The reconstruction performance is found to
improve when using the trained sensing mask from the network as compared to
other mask designs such as random, across a wide variety of compressive sensing
reconstruction algorithms. Finally, our analysis and discussion offers insights
into understanding the characteristics of the trained mask designs that lead to
the improved reconstruction quality.
|
Approximate maximum entropy principles via Goemans-Williamson with
applications to provable variational methods | The well known maximum-entropy principle due to Jaynes, which states that
given mean parameters, the maximum entropy distribution matching them is in an
exponential family, has been very popular in machine learning due to its
"Occam's razor" interpretation. Unfortunately, calculating the potentials in
the maximum-entropy distribution is intractable \cite{bresler2014hardness}. We
provide computationally efficient versions of this principle when the mean
parameters are pairwise moments: we design distributions that approximately
match given pairwise moments, while having entropy which is comparable to the
maximum entropy distribution matching those moments.
We additionally provide surprising applications of the approximate maximum
entropy principle to designing provable variational methods for partition
function calculations for Ising models without any assumptions on the
potentials of the model. More precisely, we show that in every temperature, we
can get approximation guarantees for the log-partition function comparable to
those in the low-temperature limit, which is the setting of optimization of
quadratic forms over the hypercube. \cite{alon2006approximating}
|
Parsimonious Mixed-Effects HodgeRank for Crowdsourced Preference
Aggregation | In crowdsourced preference aggregation, it is often assumed that all the
annotators are subject to a common preference or utility function which
generates their comparison behaviors in experiments. However, in reality
annotators are subject to variations due to multi-criteria, abnormal, or a
mixture of such behaviors. In this paper, we propose a parsimonious
mixed-effects model based on HodgeRank, which takes into account both the fixed
effect that the majority of annotators follows a common linear utility model,
and the random effect that a small subset of annotators might deviate from the
common significantly and exhibits strongly personalized preferences. HodgeRank
has been successfully applied to subjective quality evaluation of multimedia
and resolves pairwise crowdsourced ranking data into a global consensus ranking
and cyclic conflicts of interests. As an extension, our proposed methodology
further explores the conflicts of interests through the random effect in
annotator specific variations. The key algorithm in this paper establishes a
dynamic path from the common utility to individual variations, with different
levels of parsimony or sparsity on personalization, based on newly developed
Linearized Bregman Algorithms with Inverse Scale Space method. Finally the
validity of the methodology are supported by experiments with both simulated
examples and three real-world crowdsourcing datasets, which shows that our
proposed method exhibits better performance (i.e. smaller test error) compared
with HodgeRank due to its parsimonious property.
|
Learning in Quantum Control: High-Dimensional Global Optimization for
Noisy Quantum Dynamics | Quantum control is valuable for various quantum technologies such as
high-fidelity gates for universal quantum computing, adaptive quantum-enhanced
metrology, and ultra-cold atom manipulation. Although supervised machine
learning and reinforcement learning are widely used for optimizing control
parameters in classical systems, quantum control for parameter optimization is
mainly pursued via gradient-based greedy algorithms. Although the quantum
fitness landscape is often compatible with greedy algorithms, sometimes greedy
algorithms yield poor results, especially for large-dimensional quantum
systems. We employ differential evolution algorithms to circumvent the
stagnation problem of non-convex optimization. We improve quantum control
fidelity for noisy system by averaging over the objective function. To reduce
computational cost, we introduce heuristics for early termination of runs and
for adaptive selection of search subspaces. Our implementation is massively
parallel and vectorized to reduce run time even further. We demonstrate our
methods with two examples, namely quantum phase estimation and quantum gate
design, for which we achieve superior fidelity and scalability than obtained
using greedy algorithms.
|
Incomplete Pivoted QR-based Dimensionality Reduction | High-dimensional big data appears in many research fields such as image
recognition, biology and collaborative filtering. Often, the exploration of
such data by classic algorithms is encountered with difficulties due to `curse
of dimensionality' phenomenon. Therefore, dimensionality reduction methods are
applied to the data prior to its analysis. Many of these methods are based on
principal components analysis, which is statistically driven, namely they map
the data into a low-dimension subspace that preserves significant statistical
properties of the high-dimensional data. As a consequence, such methods do not
directly address the geometry of the data, reflected by the mutual distances
between multidimensional data point. Thus, operations such as classification,
anomaly detection or other machine learning tasks may be affected.
This work provides a dictionary-based framework for geometrically driven data
analysis that includes dimensionality reduction, out-of-sample extension and
anomaly detection. It embeds high-dimensional data in a low-dimensional
subspace. This embedding preserves the original high-dimensional geometry of
the data up to a user-defined distortion rate. In addition, it identifies a
subset of landmark data points that constitute a dictionary for the analyzed
dataset. The dictionary enables to have a natural extension of the
low-dimensional embedding to out-of-sample data points, which gives rise to a
distortion-based criterion for anomaly detection. The suggested method is
demonstrated on synthetic and real-world datasets and achieves good results for
classification, anomaly detection and out-of-sample tasks.
|
LazySVD: Even Faster SVD Decomposition Yet Without Agonizing Pain | We study $k$-SVD that is to obtain the first $k$ singular vectors of a matrix
$A$. Recently, a few breakthroughs have been discovered on $k$-SVD: Musco and
Musco [1] proved the first gap-free convergence result using the block Krylov
method, Shamir [2] discovered the first variance-reduction stochastic method,
and Bhojanapalli et al. [3] provided the fastest $O(\mathsf{nnz}(A) +
\mathsf{poly}(1/\varepsilon))$-time algorithm using alternating minimization.
In this paper, we put forward a new and simple LazySVD framework to improve
the above breakthroughs. This framework leads to a faster gap-free method
outperforming [1], and the first accelerated and stochastic method
outperforming [2]. In the $O(\mathsf{nnz}(A) + \mathsf{poly}(1/\varepsilon))$
running-time regime, LazySVD outperforms [3] in certain parameter regimes
without even using alternating minimization.
|
Recurrent Highway Networks | Many sequential processing tasks require complex nonlinear transition
functions from one step to the next. However, recurrent neural networks with
'deep' transition functions remain difficult to train, even when using Long
Short-Term Memory (LSTM) networks. We introduce a novel theoretical analysis of
recurrent networks based on Gersgorin's circle theorem that illuminates several
modeling and optimization issues and improves our understanding of the LSTM
cell. Based on this analysis we propose Recurrent Highway Networks, which
extend the LSTM architecture to allow step-to-step transition depths larger
than one. Several language modeling experiments demonstrate that the proposed
architecture results in powerful and efficient models. On the Penn Treebank
corpus, solely increasing the transition depth from 1 to 10 improves word-level
perplexity from 90.6 to 65.4 using the same number of parameters. On the larger
Wikipedia datasets for character prediction (text8 and enwik8), RHNs outperform
all previous results and achieve an entropy of 1.27 bits per character.
|
Nystrom Method for Approximating the GMM Kernel | The GMM (generalized min-max) kernel was recently proposed (Li, 2016) as a
measure of data similarity and was demonstrated effective in machine learning
tasks. In order to use the GMM kernel for large-scale datasets, the prior work
resorted to the (generalized) consistent weighted sampling (GCWS) to convert
the GMM kernel to linear kernel. We call this approach as ``GMM-GCWS''.
In the machine learning literature, there is a popular algorithm which we
call ``RBF-RFF''. That is, one can use the ``random Fourier features'' (RFF) to
convert the ``radial basis function'' (RBF) kernel to linear kernel. It was
empirically shown in (Li, 2016) that RBF-RFF typically requires substantially
more samples than GMM-GCWS in order to achieve comparable accuracies.
The Nystrom method is a general tool for computing nonlinear kernels, which
again converts nonlinear kernels into linear kernels. We apply the Nystrom
method for approximating the GMM kernel, a strategy which we name as
``GMM-NYS''. In this study, our extensive experiments on a set of fairly large
datasets confirm that GMM-NYS is also a strong competitor of RBF-RFF.
|
Deep Reconstruction-Classification Networks for Unsupervised Domain
Adaptation | In this paper, we propose a novel unsupervised domain adaptation algorithm
based on deep learning for visual object recognition. Specifically, we design a
new model called Deep Reconstruction-Classification Network (DRCN), which
jointly learns a shared encoding representation for two tasks: i) supervised
classification of labeled source data, and ii) unsupervised reconstruction of
unlabeled target data.In this way, the learnt representation not only preserves
discriminability, but also encodes useful information from the target domain.
Our new DRCN model can be optimized by using backpropagation similarly as the
standard neural networks.
We evaluate the performance of DRCN on a series of cross-domain object
recognition tasks, where DRCN provides a considerable improvement (up to ~8% in
accuracy) over the prior state-of-the-art algorithms. Interestingly, we also
observe that the reconstruction pipeline of DRCN transforms images from the
source domain into images whose appearance resembles the target dataset. This
suggests that DRCN's performance is due to constructing a single composite
representation that encodes information about both the structure of target
images and the classification of source images. Finally, we provide a formal
analysis to justify the algorithm's objective in domain adaptation context.
|
Improved Multi-Class Cost-Sensitive Boosting via Estimation of the
Minimum-Risk Class | We present a simple unified framework for multi-class cost-sensitive
boosting. The minimum-risk class is estimated directly, rather than via an
approximation of the posterior distribution. Our method jointly optimizes
binary weak learners and their corresponding output vectors, requiring classes
to share features at each iteration. By training in a cost-sensitive manner,
weak learners are invested in separating classes whose discrimination is
important, at the expense of less relevant classification boundaries.
Additional contributions are a family of loss functions along with proof that
our algorithm is Boostable in the theoretical sense, as well as an efficient
procedure for growing decision trees for use as weak learners. We evaluate our
method on a variety of datasets: a collection of synthetic planar data, common
UCI datasets, MNIST digits, SUN scenes, and CUB-200 birds. Results show
state-of-the-art performance across all datasets against several strong
baselines, including non-boosting multi-class approaches.
|
Fast Sampling for Strongly Rayleigh Measures with Application to
Determinantal Point Processes | In this note we consider sampling from (non-homogeneous) strongly Rayleigh
probability measures. As an important corollary, we obtain a fast mixing Markov
Chain sampler for Determinantal Point Processes.
|
Estimating Uncertainty Online Against an Adversary | Assessing uncertainty is an important step towards ensuring the safety and
reliability of machine learning systems. Existing uncertainty estimation
techniques may fail when their modeling assumptions are not met, e.g. when the
data distribution differs from the one seen at training time. Here, we propose
techniques that assess a classification algorithm's uncertainty via calibrated
probabilities (i.e. probabilities that match empirical outcome frequencies in
the long run) and which are guaranteed to be reliable (i.e. accurate and
calibrated) on out-of-distribution input, including input generated by an
adversary. This represents an extension of classical online learning that
handles uncertainty in addition to guaranteeing accuracy under adversarial
assumptions. We establish formal guarantees for our methods, and we validate
them on two real-world problems: question answering and medical diagnosis from
genomic data.
|
Characterizing Driving Styles with Deep Learning | Characterizing driving styles of human drivers using vehicle sensor data,
e.g., GPS, is an interesting research problem and an important real-world
requirement from automotive industries. A good representation of driving
features can be highly valuable for autonomous driving, auto insurance, and
many other application scenarios. However, traditional methods mainly rely on
handcrafted features, which limit machine learning algorithms to achieve a
better performance. In this paper, we propose a novel deep learning solution to
this problem, which could be the first attempt of extending deep learning to
driving behavior analysis based on GPS data. The proposed approach can
effectively extract high level and interpretable features describing complex
driving patterns. It also requires significantly less human experience and
work. The power of the learned driving style representations are validated
through the driver identification problem using a large real dataset.
|
San Francisco Crime Classification | San Francisco Crime Classification is an online competition administered by
Kaggle Inc. The competition aims at predicting the future crimes based on a
given set of geographical and time-based features. In this paper, I achieved a
an accuracy that ranks at top %18, as of May 19th, 2016. I will explore the
data, and explain in details the tools I used to achieve that result.
|
Unsupervised Feature Learning Based on Deep Models for Environmental
Audio Tagging | Environmental audio tagging aims to predict only the presence or absence of
certain acoustic events in the interested acoustic scene. In this paper we make
contributions to audio tagging in two parts, respectively, acoustic modeling
and feature learning. We propose to use a shrinking deep neural network (DNN)
framework incorporating unsupervised feature learning to handle the multi-label
classification task. For the acoustic modeling, a large set of contextual
frames of the chunk are fed into the DNN to perform a multi-label
classification for the expected tags, considering that only chunk (or
utterance) level rather than frame-level labels are available. Dropout and
background noise aware training are also adopted to improve the generalization
capability of the DNNs. For the unsupervised feature learning, we propose to
use a symmetric or asymmetric deep de-noising auto-encoder (sDAE or aDAE) to
generate new data-driven features from the Mel-Filter Banks (MFBs) features.
The new features, which are smoothed against background noise and more compact
with contextual information, can further improve the performance of the DNN
baseline. Compared with the standard Gaussian Mixture Model (GMM) baseline of
the DCASE 2016 audio tagging challenge, our proposed method obtains a
significant equal error rate (EER) reduction from 0.21 to 0.13 on the
development set. The proposed aDAE system can get a relative 6.7% EER reduction
compared with the strong DNN baseline on the development set. Finally, the
results also show that our approach obtains the state-of-the-art performance
with 0.15 EER on the evaluation set of the DCASE 2016 audio tagging task while
EER of the first prize of this challenge is 0.17.
|
Hierarchical learning for DNN-based acoustic scene classification | In this paper, we present a deep neural network (DNN)-based acoustic scene
classification framework. Two hierarchical learning methods are proposed to
improve the DNN baseline performance by incorporating the hierarchical taxonomy
information of environmental sounds. Firstly, the parameters of the DNN are
initialized by the proposed hierarchical pre-training. Multi-level objective
function is then adopted to add more constraint on the cross-entropy based loss
function. A series of experiments were conducted on the Task1 of the Detection
and Classification of Acoustic Scenes and Events (DCASE) 2016 challenge. The
final DNN-based system achieved a 22.9% relative improvement on average scene
classification error as compared with the Gaussian Mixture Model (GMM)-based
benchmark system across four standard folds.
|
Sequential Cost-Sensitive Feature Acquisition | We propose a reinforcement learning based approach to tackle the
cost-sensitive learning problem where each input feature has a specific cost.
The acquisition process is handled through a stochastic policy which allows
features to be acquired in an adaptive way. The general architecture of our
approach relies on representation learning to enable performing prediction on
any partially observed sample, whatever the set of its observed features are.
The resulting model is an original mix of representation learning and of
reinforcement learning ideas. It is learned with policy gradient techniques to
minimize a budgeted inference cost. We demonstrate the effectiveness of our
proposed method with several experiments on a variety of datasets for the
sparse prediction problem where all features have the same cost, but also for
some cost-sensitive settings.
|
Possibilistic Networks: Parameters Learning from Imprecise Data and
Evaluation strategy | There has been an ever-increasing interest in multidisciplinary research on
representing and reasoning with imperfect data. Possibilistic networks present
one of the powerful frameworks of interest for representing uncertain and
imprecise information. This paper covers the problem of their parameters
learning from imprecise datasets, i.e., containing multi-valued data. We
propose in the rst part of this paper a possibilistic networks sampling
process. In the second part, we propose a likelihood function which explores
the link between random sets theory and possibility theory. This function is
then deployed to parametrize possibilistic networks.
|
Re-presenting a Story by Emotional Factors using Sentimental Analysis
Method | Remembering an event is affected by personal emotional status. We examined
the psychological status and personal factors; depression (Center for
Epidemiological Studies - Depression, Radloff, 1977), present affective
(Positive Affective and Negative Affective Schedule, Watson et al., 1988), life
orient (Life Orient Test, Scheier & Carver, 1985), self-awareness (Core Self
Evaluation Scale, Judge et al., 2003), and social factor (Social Support,
Sarason et al., 1983) of undergraduate students (N=64) and got summaries of a
story, Chronicle of a Death Foretold (Gabriel Garcia Marquez, 1981) from them.
We implement a sentimental analysis model based on convolutional neural network
(LeCun & Bengio, 1995) to evaluate each summary. From the same vein used for
transfer learning (Pan & Yang, 2010), we collected 38,265 movie review data to
train the model and then use them to score summaries of each student. The
results of CES-D and PANAS show the relationship between emotion and memory
retrieval as follows: depressed people have shown a tendency of representing a
story more negatively, and they seemed less expressive. People with full of
emotion - high in PANAS - have retrieved their memory more expressively than
others, using more negative words then others. The contributions of this study
can be summarized as follows: First, lightening the relationship between
emotion and its effect during times of storing or retrieving a memory. Second,
suggesting objective methods to evaluate the intensity of emotion in natural
language format, using a sentimental analysis model.
|
Learning Shallow Detection Cascades for Wearable Sensor-Based Mobile
Health Applications | The field of mobile health aims to leverage recent advances in wearable
on-body sensing technology and smart phone computing capabilities to develop
systems that can monitor health states and deliver just-in-time adaptive
interventions. However, existing work has largely focused on analyzing
collected data in the off-line setting. In this paper, we propose a novel
approach to learning shallow detection cascades developed explicitly for use in
a real-time wearable-phone or wearable-phone-cloud systems. We apply our
approach to the problem of cigarette smoking detection from a combination of
wrist-worn actigraphy data and respiration chest band data using two and three
stage cascades.
|
A Vector Space for Distributional Semantics for Entailment | Distributional semantics creates vector-space representations that capture
many forms of semantic similarity, but their relation to semantic entailment
has been less clear. We propose a vector-space model which provides a formal
foundation for a distributional semantics of entailment. Using a mean-field
approximation, we develop approximate inference procedures and entailment
operators over vectors of probabilities of features being known (versus
unknown). We use this framework to reinterpret an existing
distributional-semantic model (Word2Vec) as approximating an entailment-based
model of the distributions of words in contexts, thereby predicting lexical
entailment relations. In both unsupervised and semi-supervised experiments on
hyponymy detection, we get substantial improvements over previous results.
|
Feature Extraction and Automated Classification of Heartbeats by Machine
Learning | We present algorithms for the detection of a class of heart arrhythmias with
the goal of eventual adoption by practicing cardiologists. In clinical
practice, detection is based on a small number of meaningful features extracted
from the heartbeat cycle. However, techniques proposed in the literature use
high dimensional vectors consisting of morphological, and time based features
for detection. Using electrocardiogram (ECG) signals, we found smaller subsets
of features sufficient to detect arrhythmias with high accuracy. The features
were found by an iterative step-wise feature selection method. We depart from
common literature in the following aspects: 1. As opposed to a high dimensional
feature vectors, we use a small set of features with meaningful clinical
interpretation, 2. we eliminate the necessity of short-duration
patient-specific ECG data to append to the global training data for
classification 3. We apply semi-parametric classification procedures (in an
ensemble framework) for arrhythmia detection, and 4. our approach is based on a
reduced sampling rate of ~ 115 Hz as opposed to 360 Hz in standard literature.
|
The KIT Motion-Language Dataset | Linking human motion and natural language is of great interest for the
generation of semantic representations of human activities as well as for the
generation of robot activities based on natural language input. However, while
there have been years of research in this area, no standardized and openly
available dataset exists to support the development and evaluation of such
systems. We therefore propose the KIT Motion-Language Dataset, which is large,
open, and extensible. We aggregate data from multiple motion capture databases
and include them in our dataset using a unified representation that is
independent of the capture system or marker set, making it easy to work with
the data regardless of its origin. To obtain motion annotations in natural
language, we apply a crowd-sourcing approach and a web-based tool that was
specifically build for this purpose, the Motion Annotation Tool. We thoroughly
document the annotation process itself and discuss gamification methods that we
used to keep annotators motivated. We further propose a novel method,
perplexity-based selection, which systematically selects motions for further
annotation that are either under-represented in our dataset or that have
erroneous annotations. We show that our method mitigates the two aforementioned
problems and ensures a systematic annotation process. We provide an in-depth
analysis of the structure and contents of our resulting dataset, which, as of
October 10, 2016, contains 3911 motions with a total duration of 11.23 hours
and 6278 annotations in natural language that contain 52,903 words. We believe
this makes our dataset an excellent choice that enables more transparent and
comparable research in this important area.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.