title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Learning in games with continuous action sets and unknown payoff
functions | This paper examines the convergence of no-regret learning in games with
continuous action sets. For concreteness, we focus on learning via "dual
averaging", a widely used class of no-regret learning schemes where players
take small steps along their individual payoff gradients and then "mirror" the
output back to their action sets. In terms of feedback, we assume that players
can only estimate their payoff gradients up to a zero-mean error with bounded
variance. To study the convergence of the induced sequence of play, we
introduce the notion of variational stability, and we show that stable
equilibria are locally attracting with high probability whereas globally stable
equilibria are globally attracting with probability 1. We also discuss some
applications to mixed-strategy learning in finite games, and we provide
explicit estimates of the method's convergence speed.
|
Fundamental Limits of Budget-Fidelity Trade-off in Label Crowdsourcing | Digital crowdsourcing (CS) is a modern approach to perform certain large
projects using small contributions of a large crowd. In CS, a taskmaster
typically breaks down the project into small batches of tasks and assigns them
to so-called workers with imperfect skill levels. The crowdsourcer then
collects and analyzes the results for inference and serving the purpose of the
project. In this work, the CS problem, as a human-in-the-loop computation
problem, is modeled and analyzed in an information theoretic rate-distortion
framework. The purpose is to identify the ultimate fidelity that one can
achieve by any form of query from the crowd and any decoding (inference)
algorithm with a given budget. The results are established by a joint source
channel (de)coding scheme, which represent the query scheme and inference, over
parallel noisy channels, which model workers with imperfect skill levels. We
also present and analyze a query scheme dubbed $k$-ary incidence coding and
study optimized query pricing in this setting.
|
Collaborative Filtering with Recurrent Neural Networks | We show that collaborative filtering can be viewed as a sequence prediction
problem, and that given this interpretation, recurrent neural networks offer
very competitive approach. In particular we study how the long short-term
memory (LSTM) can be applied to collaborative filtering, and how it compares to
standard nearest neighbors and matrix factorization methods on movie
recommendation. We show that the LSTM is competitive in all aspects, and
largely outperforms other methods in terms of item coverage and short term
predictions.
|
Hard Negative Mining for Metric Learning Based Zero-Shot Classification | Zero-Shot learning has been shown to be an efficient strategy for domain
adaptation. In this context, this paper builds on the recent work of Bucher et
al. [1], which proposed an approach to solve Zero-Shot classification problems
(ZSC) by introducing a novel metric learning based objective function. This
objective function allows to learn an optimal embedding of the attributes
jointly with a measure of similarity between images and attributes. This paper
extends their approach by proposing several schemes to control the generation
of the negative pairs, resulting in a significant improvement of the
performance and giving above state-of-the-art results on three challenging ZSC
datasets.
|
Entity Embedding-based Anomaly Detection for Heterogeneous Categorical
Events | Anomaly detection plays an important role in modern data-driven security
applications, such as detecting suspicious access to a socket from a process.
In many cases, such events can be described as a collection of categorical
values that are considered as entities of different types, which we call
heterogeneous categorical events. Due to the lack of intrinsic distance
measures among entities, and the exponentially large event space, most existing
work relies heavily on heuristics to calculate abnormal scores for events.
Different from previous work, we propose a principled and unified probabilistic
model APE (Anomaly detection via Probabilistic pairwise interaction and Entity
embedding) that directly models the likelihood of events. In this model, we
embed entities into a common latent space using their observed co-occurrence in
different events. More specifically, we first model the compatibility of each
pair of entities according to their embeddings. Then we utilize the weighted
pairwise interactions of different entity types to define the event
probability. Using Noise-Contrastive Estimation with "context-dependent" noise
distribution, our model can be learned efficiently regardless of the large
event space. Experimental results on real enterprise surveillance data show
that our methods can accurately detect abnormal events compared to other
state-of-the-art abnormal detection techniques.
|
Leveraging over intact priors for boosting control and dexterity of
prosthetic hands by amputees | Non-invasive myoelectric prostheses require a long training time to obtain
satisfactory control dexterity. These training times could possibly be reduced
by leveraging over training efforts by previous subjects. So-called domain
adaptation algorithms formalize this strategy and have indeed been shown to
significantly reduce the amount of required training data for intact subjects
for myoelectric movements classification. It is not clear, however, whether
these results extend also to amputees and, if so, whether prior information
from amputees and intact subjects is equally useful. To overcome this problem,
we evaluated several domain adaptation algorithms on data coming from both
amputees and intact subjects. Our findings indicate that: (1) the use of
previous experience from other subjects allows us to reduce the training time
by about an order of magnitude; (2) this improvement holds regardless of
whether an amputee exploits previous information from other amputees or from
intact subjects.
|
Clustering and Community Detection with Imbalanced Clusters | Spectral clustering methods which are frequently used in clustering and
community detection applications are sensitive to the specific graph
constructions particularly when imbalanced clusters are present. We show that
ratio cut (RCut) or normalized cut (NCut) objectives are not tailored to
imbalanced cluster sizes since they tend to emphasize cut sizes over cut
values. We propose a graph partitioning problem that seeks minimum cut
partitions under minimum size constraints on partitions to deal with imbalanced
cluster sizes. Our approach parameterizes a family of graphs by adaptively
modulating node degrees on a fixed node set, yielding a set of parameter
dependent cuts reflecting varying levels of imbalance. The solution to our
problem is then obtained by optimizing over these parameters. We present
rigorous limit cut analysis results to justify our approach and demonstrate the
superiority of our method through experiments on synthetic and real datasets
for data clustering, semi-supervised learning and community detection.
|
Interacting with Massive Behavioral Data | In this short paper, we propose the split-diffuse (SD) algorithm that takes
the output of an existing word embedding algorithm, and distributes the data
points uniformly across the visualization space. The result improves the
perceivability and the interactability by the human.
We apply the SD algorithm to analyze the user behavior through access logs
within the cyber security domain. The result, named the topic grids, is a set
of grids on various topics generated from the logs. On the same set of grids,
different behavioral metrics can be shown on different targets over different
periods of time, to provide visualization and interaction to the human experts.
Analysis, investigation, and other types of interaction can be performed on
the topic grids more efficiently than on the output of existing dimension
reduction methods. In addition to the cyber security domain, the topic grids
can be further applied to other domains like e-commerce, credit card
transaction, customer service to analyze the behavior in a large scale.
|
Large Scale Behavioral Analytics via Topical Interaction | We propose the split-diffuse (SD) algorithm that takes the output of an
existing dimension reduction algorithm, and distributes the data points
uniformly across the visualization space. The result, called the topic grids,
is a set of grids on various topics which are generated from the free-form text
content of any domain of interest. The topic grids efficiently utilizes the
visualization space to provide visual summaries for massive data. Topical
analysis, comparison and interaction can be performed on the topic grids in a
more perceivable way.
|
Global analysis of Expectation Maximization for mixtures of two
Gaussians | Expectation Maximization (EM) is among the most popular algorithms for
estimating parameters of statistical models. However, EM, which is an iterative
algorithm based on the maximum likelihood principle, is generally only
guaranteed to find stationary points of the likelihood objective, and these
points may be far from any maximizer. This article addresses this disconnect
between the statistical principles behind EM and its algorithmic properties.
Specifically, it provides a global analysis of EM for specific models in which
the observations comprise an i.i.d. sample from a mixture of two Gaussians.
This is achieved by (i) studying the sequence of parameters from idealized
execution of EM in the infinite sample limit, and fully characterizing the
limit points of the sequence in terms of the initial parameters; and then (ii)
based on this convergence analysis, establishing statistical consistency (or
lack thereof) for the actual sequence of parameters produced by EM.
|
Learning Temporal Dependence from Time-Series Data with Latent Variables | We consider the setting where a collection of time series, modeled as random
processes, evolve in a causal manner, and one is interested in learning the
graph governing the relationships of these processes. A special case of wide
interest and applicability is the setting where the noise is Gaussian and
relationships are Markov and linear. We study this setting with two additional
features: firstly, each random process has a hidden (latent) state, which we
use to model the internal memory possessed by the variables (similar to hidden
Markov models). Secondly, each variable can depend on its latent memory state
through a random lag (rather than a fixed lag), thus modeling memory recall
with differing lags at distinct times. Under this setting, we develop an
estimator and prove that under a genericity assumption, the parameters of the
model can be learned consistently. We also propose a practical adaption of this
estimator, which demonstrates significant performance gains in both synthetic
and real-world datasets.
|
Learning to generalize to new compositions in image understanding | Recurrent neural networks have recently been used for learning to describe
images using natural language. However, it has been observed that these models
generalize poorly to scenes that were not observed during training, possibly
depending too strongly on the statistics of the text in the training data. Here
we propose to describe images using short structured representations, aiming to
capture the crux of a description. These structured representations allow us to
tease-out and evaluate separately two types of generalization: standard
generalization to new images with similar scenes, and generalization to new
combinations of known entities. We compare two learning approaches on the
MS-COCO dataset: a state-of-the-art recurrent network based on an LSTM (Show,
Attend and Tell), and a simple structured prediction model on top of a deep
network. We find that the structured model generalizes to new compositions
substantially better than the LSTM, ~7 times the accuracy of predicting
structured representations. By providing a concrete method to quantify
generalization for unseen combinations, we argue that structured
representations and compositional splits are a useful benchmark for image
captioning, and advocate compositional models that capture linguistic and
visual structure.
|
KSR: A Semantic Representation of Knowledge Graph within a Novel
Unsupervised Paradigm | Knowledge representation is a long-history topic in AI, which is very
important. A variety of models have been proposed for knowledge graph
embedding, which projects symbolic entities and relations into continuous
vector space. However, most related methods merely focus on the data-fitting of
knowledge graph, and ignore the interpretable semantic expression. Thus,
traditional embedding methods are not friendly for applications that require
semantic analysis, such as question answering and entity retrieval. To this
end, this paper proposes a semantic representation method for knowledge graph
\textbf{(KSR)}, which imposes a two-level hierarchical generative process that
globally extracts many aspects and then locally assigns a specific category in
each aspect for every triple. Since both aspects and categories are
semantics-relevant, the collection of categories in each aspect is treated as
the semantic representation of this triple. Extensive experiments show that our
model outperforms other state-of-the-art baselines substantially.
|
A Boundary Tilting Persepective on the Phenomenon of Adversarial
Examples | Deep neural networks have been shown to suffer from a surprising weakness:
their classification outputs can be changed by small, non-random perturbations
of their inputs. This adversarial example phenomenon has been explained as
originating from deep networks being "too linear" (Goodfellow et al., 2014). We
show here that the linear explanation of adversarial examples presents a number
of limitations: the formal argument is not convincing, linear classifiers do
not always suffer from the phenomenon, and when they do their adversarial
examples are different from the ones affecting deep networks.
We propose a new perspective on the phenomenon. We argue that adversarial
examples exist when the classification boundary lies close to the submanifold
of sampled data, and present a mathematical analysis of this new perspective in
the linear case. We define the notion of adversarial strength and show that it
can be reduced to the deviation angle between the classifier considered and the
nearest centroid classifier. Then, we show that the adversarial strength can be
made arbitrarily high independently of the classification performance due to a
mechanism that we call boundary tilting. This result leads us to defining a new
taxonomy of adversarial examples. Finally, we show that the adversarial
strength observed in practice is directly dependent on the level of
regularisation used and the strongest adversarial examples, symptomatic of
overfitting, can be avoided by using a proper level of regularisation.
|
Random Forest for Label Ranking | Label ranking aims to learn a mapping from instances to rankings over a
finite number of predefined labels. Random forest is a powerful and one of the
most successful general-purpose machine learning algorithms of modern times. In
this paper, we present a powerful random forest label ranking method which uses
random decision trees to retrieve nearest neighbors. We have developed a novel
two-step rank aggregation strategy to effectively aggregate neighboring
rankings discovered by the random forest into a final predicted ranking.
Compared with existing methods, the new random forest method has many
advantages including its intrinsically scalable tree data structure, highly
parallel-able computational architecture and much superior performance. We
present extensive experimental results to demonstrate that our new method
achieves the highly competitive performance compared with state-of-the-art
methods for datasets with complete ranking and datasets with only partial
ranking information.
|
Temperature-Based Deep Boltzmann Machines | Deep learning techniques have been paramount in the last years, mainly due to
their outstanding results in a number of applications, that range from speech
recognition to face-based user identification. Despite other techniques
employed for such purposes, Deep Boltzmann Machines are among the most used
ones, which are composed of layers of Restricted Boltzmann Machines (RBMs)
stacked on top of each other. In this work, we evaluate the concept of
temperature in DBMs, which play a key role in Boltzmann-related distributions,
but it has never been considered in this context up to date. Therefore, the
main contribution of this paper is to take into account this information and to
evaluate its influence in DBMs considering the task of binary image
reconstruction. We expect this work can foster future research considering the
usage of different temperatures during learning in DBMs.
|
Bayesian selection for the l2-Potts model regularization parameter: 1D
piecewise constant signal denoising | Piecewise constant denoising can be solved either by deterministic
optimization approaches, based on the Potts model, or by stochastic Bayesian
procedures. The former lead to low computational time but require the selection
of a regularization parameter, whose value significantly impacts the achieved
solution, and whose automated selection remains an involved and challenging
problem. Conversely, fully Bayesian formalisms encapsulate the regularization
parameter selection into hierarchical models, at the price of high
computational costs. This contribution proposes an operational strategy that
combines hierarchical Bayesian and Potts model formulations, with the double
aim of automatically tuning the regularization parameter and of maintaining
computational effciency. The proposed procedure relies on formally connecting a
Bayesian framework to a l2-Potts functional. Behaviors and performance for the
proposed piecewise constant denoising and regularization parameter tuning
techniques are studied qualitatively and assessed quantitatively, and shown to
compare favorably against those of a fully Bayesian hierarchical procedure,
both in accuracy and in computational load.
|
Online Monotone Optimization | This paper presents a new framework for analyzing and designing no-regret
algorithms for dynamic (possibly adversarial) systems. The proposed framework
generalizes the popular online convex optimization framework and extends it to
its natural limit allowing it to capture a notion of regret that is intuitive
for more general problems such as those encountered in game theory and
variational inequalities. The framework hinges on a special choice of a
system-wide loss function we have developed. Using this framework, we prove
that a simple update scheme provides a no-regret algorithm for monotone
systems. While previous results in game theory prove individual agents can
enjoy unilateral no-regret guarantees, our result proves monotonicity
sufficient for guaranteeing no-regret when considering the adjustments of
multiple agent strategies in parallel. Furthermore, to our knowledge, this is
the first framework to provide a suitable notion of regret for variational
inequalities. Most importantly, our proposed framework ensures monotonicity a
sufficient condition for employing multiple online learners safely in parallel.
|
Optimizing Recurrent Neural Networks Architectures under Time
Constraints | Recurrent neural network (RNN)'s architecture is a key factor influencing its
performance. We propose algorithms to optimize hidden sizes under running time
constraint. We convert the discrete optimization into a subset selection
problem. By novel transformations, the objective function becomes submodular
and constraint becomes supermodular. A greedy algorithm with bounds is
suggested to solve the transformed problem. And we show how transformations
influence the bounds. To speed up optimization, surrogate functions are
proposed which balance exploration and exploitation. Experiments show that our
algorithms can find more accurate models or faster models than manually tuned
state-of-the-art and random search. We also compare popular RNN architectures
using our algorithms.
|
Human-Algorithm Interaction Biases in the Big Data Cycle: A Markov Chain
Iterated Learning Framework | Early supervised machine learning algorithms have relied on reliable expert
labels to build predictive models. However, the gates of data generation have
recently been opened to a wider base of users who started participating
increasingly with casual labeling, rating, annotating, etc. The increased
online presence and participation of humans has led not only to a
democratization of unchecked inputs to algorithms, but also to a wide
democratization of the "consumption" of machine learning algorithms' outputs by
general users. Hence, these algorithms, many of which are becoming essential
building blocks of recommender systems and other information filters, started
interacting with users at unprecedented rates. The result is machine learning
algorithms that consume more and more data that is unchecked, or at the very
least, not fitting conventional assumptions made by various machine learning
algorithms. These include biased samples, biased labels, diverging training and
testing sets, and cyclical interaction between algorithms, humans, information
consumed by humans, and data consumed by algorithms. Yet, the continuous
interaction between humans and algorithms is rarely taken into account in
machine learning algorithm design and analysis. In this paper, we present a
preliminary theoretical model and analysis of the mutual interaction between
humans and algorithms, based on an iterated learning framework that is inspired
from the study of human language evolution. We also define the concepts of
human and algorithm blind spots and outline machine learning approaches to mend
iterated bias through two novel notions: antidotes and reactive learning.
|
Relevant based structure learning for feature selection | Feature selection is an important task in many problems occurring in pattern
recognition, bioinformatics, machine learning and data mining applications. The
feature selection approach enables us to reduce the computation burden and the
falling accuracy effect of dealing with huge number of features in typical
learning problems. There is a variety of techniques for feature selection in
supervised learning problems based on different selection metrics. In this
paper, we propose a novel unified framework for feature selection built on the
graphical models and information theoretic tools. The proposed approach
exploits the structure learning among features to select more relevant and less
redundant features to the predictive modeling problem according to a primary
novel likelihood based criterion. In line with the selection of the optimal
subset of features through the proposed method, it provides us the Bayesian
network classifier without the additional cost of model training on the
selected subset of features. The optimal properties of our method are
established through empirical studies and computational complexity analysis.
Furthermore the proposed approach is evaluated on a bunch of benchmark datasets
based on the well-known classification algorithms. Extensive experiments
confirm the significant improvement of the proposed approach compared to the
earlier works.
|
Learning-Based Resource Allocation Scheme for TDD-Based CRAN System | Explosive growth in the use of smart wireless devices has necessitated the
provision of higher data rates and always-on connectivity, which are the main
motivators for designing the fifth generation (5G) systems. To achieve higher
system efficiency, massive antenna deployment with tight coordination is one
potential strategy for designing 5G systems, but has two types of associated
system overhead. First is the synchronization overhead, which can be reduced by
implementing a cloud radio access network (CRAN)-based architecture design,
that separates the baseband processing and radio access functionality to
achieve better system synchronization. Second is the overhead for acquiring
channel state information (CSI) of the users present in the system, which,
however, increases tremendously when instantaneous CSI is used to serve
high-mobility users. To serve a large number of users, a CRAN system with a
dense deployment of remote radio heads (RRHs) is considered, such that each
user has a line-of-sight (LOS) link with the corresponding RRH. Since, the
trajectory of movement for high-mobility users is predictable; therefore,
fairly accurate position estimates for those users can be obtained, and can be
used for resource allocation to serve the considered users. The resource
allocation is dependent upon various correlated system parameters, and these
correlations can be learned using well-known \emph{machine learning}
algorithms. This paper proposes a novel \emph{learning-based resource
allocation scheme} for time division duplex (TDD) based 5G CRAN systems with
dense RRH deployment, by using only the users' position estimates for resource
allocation, thus avoiding the need for CSI acquisition. This reduces the
overall system overhead significantly, while still achieving near-optimal
system performance; thus, better (effective) system efficiency is achieved.
(See the paper for full abstract)
|
Robust Discriminative Clustering with Sparse Regularizers | Clustering high-dimensional data often requires some form of dimensionality
reduction, where clustered variables are separated from "noise-looking"
variables. We cast this problem as finding a low-dimensional projection of the
data which is well-clustered. This yields a one-dimensional projection in the
simplest situation with two clusters, and extends naturally to a multi-label
scenario for more than two clusters. In this paper, (a) we first show that this
joint clustering and dimension reduction formulation is equivalent to
previously proposed discriminative clustering frameworks, thus leading to
convex relaxations of the problem, (b) we propose a novel sparse extension,
which is still cast as a convex relaxation and allows estimation in higher
dimensions, (c) we propose a natural extension for the multi-label scenario,
(d) we provide a new theoretical analysis of the performance of these
formulations with a simple probabilistic model, leading to scalings over the
form $d=O(\sqrt{n})$ for the affine invariant case and $d=O(n)$ for the sparse
case, where $n$ is the number of examples and $d$ the ambient dimension, and
finally, (e) we propose an efficient iterative algorithm with running-time
complexity proportional to $O(nd^2)$, improving on earlier algorithms which had
quadratic complexity in the number of examples.
|
Wasserstein Discriminant Analysis | Wasserstein Discriminant Analysis (WDA) is a new supervised method that can
improve classification of high-dimensional data by computing a suitable linear
map onto a lower dimensional subspace. Following the blueprint of classical
Linear Discriminant Analysis (LDA), WDA selects the projection matrix that
maximizes the ratio of two quantities: the dispersion of projected points
coming from different classes, divided by the dispersion of projected points
coming from the same class. To quantify dispersion, WDA uses regularized
Wasserstein distances, rather than cross-variance measures which have been
usually considered, notably in LDA. Thanks to the the underlying principles of
optimal transport, WDA is able to capture both global (at distribution scale)
and local (at samples scale) interactions between classes. Regularized
Wasserstein distances can be computed using the Sinkhorn matrix scaling
algorithm; We show that the optimization of WDA can be tackled using automatic
differentiation of Sinkhorn iterations. Numerical experiments show promising
results both in terms of prediction and visualization on toy examples and real
life datasets such as MNIST and on deep features obtained from a subset of the
Caltech dataset.
|
Data Poisoning Attacks on Factorization-Based Collaborative Filtering | Recommendation and collaborative filtering systems are important in modern
information and e-commerce applications. As these systems are becoming
increasingly popular in the industry, their outputs could affect business
decision making, introducing incentives for an adversarial party to compromise
the availability or integrity of such systems. We introduce a data poisoning
attack on collaborative filtering systems. We demonstrate how a powerful
attacker with full knowledge of the learner can generate malicious data so as
to maximize his/her malicious objectives, while at the same time mimicking
normal user behavior to avoid being detected. While the complete knowledge
assumption seems extreme, it enables a robust assessment of the vulnerability
of collaborative filtering schemes to highly motivated attacks. We present
efficient solutions for two popular factorization-based collaborative filtering
algorithms: the \emph{alternative minimization} formulation and the
\emph{nuclear norm minimization} method. Finally, we test the effectiveness of
our proposed algorithms on real-world data and discuss potential defensive
strategies.
|
Why does deep and cheap learning work so well? | We show how the success of deep learning could depend not only on mathematics
but also on physics: although well-known mathematical theorems guarantee that
neural networks can approximate arbitrary functions well, the class of
functions of practical interest can frequently be approximated through "cheap
learning" with exponentially fewer parameters than generic ones. We explore how
properties frequently encountered in physics such as symmetry, locality,
compositionality, and polynomial log-probability translate into exceptionally
simple neural networks. We further argue that when the statistical process
generating the data is of a certain hierarchical form prevalent in physics and
machine-learning, a deep neural network can be more efficient than a shallow
one. We formalize these claims using information theory and discuss the
relation to the renormalization group. We prove various "no-flattening
theorems" showing when efficient linear deep networks cannot be accurately
approximated by shallow ones without efficiency loss, for example, we show that
$n$ variables cannot be multiplied using fewer than 2^n neurons in a single
hidden layer.
|
Visualizing and Understanding Sum-Product Networks | Sum-Product Networks (SPNs) are recently introduced deep tractable
probabilistic models by which several kinds of inference queries can be
answered exactly and in a tractable time. Up to now, they have been largely
used as black box density estimators, assessed only by comparing their
likelihood scores only. In this paper we explore and exploit the inner
representations learned by SPNs. We do this with a threefold aim: first we want
to get a better understanding of the inner workings of SPNs; secondly, we seek
additional ways to evaluate one SPN model and compare it against other
probabilistic models, providing diagnostic tools to practitioners; lastly, we
want to empirically evaluate how good and meaningful the extracted
representations are, as in a classic Representation Learning framework. In
order to do so we revise their interpretation as deep neural networks and we
propose to exploit several visualization techniques on their node activations
and network outputs under different types of inference queries. To investigate
these models as feature extractors, we plug some SPNs, learned in a greedy
unsupervised fashion on image datasets, in supervised classification learning
tasks. We extract several embedding types from node activations by filtering
nodes by their type, by their associated feature abstraction level and by their
scope. In a thorough empirical comparison we prove them to be competitive
against those generated from popular feature extractors as Restricted Boltzmann
Machines. Finally, we investigate embeddings generated from random
probabilistic marginal queries as means to compare other tractable
probabilistic models on a common ground, extending our experiments to Mixtures
of Trees.
|
Data Dependent Convergence for Distributed Stochastic Optimization | In this dissertation we propose alternative analysis of distributed
stochastic gradient descent (SGD) algorithms that rely on spectral properties
of the data covariance. As a consequence we can relate questions pertaining to
speedups and convergence rates for distributed SGD to the data distribution
instead of the regularity properties of the objective functions. More precisely
we show that this rate depends on the spectral norm of the sample covariance
matrix. An estimate of this norm can provide practitioners with guidance
towards a potential gain in algorithm performance. For example many sparse
datasets with low spectral norm prove to be amenable to gains in distributed
settings. Towards establishing this data dependence we first study a
distributed consensus-based SGD algorithm and show that the rate of convergence
involves the spectral norm of the sample covariance matrix when the underlying
data is assumed to be independent and identically distributed (homogenous).
This dependence allows us to identify network regimes that prove to be
beneficial for datasets with low sample covariance spectral norm. Existing
consensus based analyses prove to be sub-optimal in the homogenous setting. Our
analysis method also allows us to find data-dependent convergence rates as we
limit the amount of communication. Spreading a fixed amount of data across more
nodes slows convergence; in the asymptotic regime we show that adding more
machines can help when minimizing twice-differentiable losses. Since the
mini-batch results don't follow from the consensus results we propose a
different data dependent analysis thereby providing theoretical validation for
why certain datasets are more amenable to mini-batching. We also provide
empirical evidence for results in this thesis.
|
Multi-Label Classification Method Based on Extreme Learning Machines | In this paper, an Extreme Learning Machine (ELM) based technique for
Multi-label classification problems is proposed and discussed. In multi-label
classification, each of the input data samples belongs to one or more than one
class labels. The traditional binary and multi-class classification problems
are the subset of the multi-label problem with the number of labels
corresponding to each sample limited to one. The proposed ELM based multi-label
classification technique is evaluated with six different benchmark multi-label
datasets from different domains such as multimedia, text and biology. A
detailed comparison of the results is made by comparing the proposed method
with the results from nine state of the arts techniques for five different
evaluation metrics. The nine methods are chosen from different categories of
multi-label methods. The comparative results shows that the proposed Extreme
Learning Machine based multi-label classification technique is a better
alternative than the existing state of the art methods for multi-label
problems.
|
Applying Naive Bayes Classification to Google Play Apps Categorization | There are over one million apps on Google Play Store and over half a million
publishers. Having such a huge number of apps and developers can pose a
challenge to app users and new publishers on the store. Discovering apps can be
challenging if apps are not correctly published in the right category, and, in
turn, reduce earnings for app developers. Additionally, with over 41 categories
on Google Play Store, deciding on the right category to publish an app can be
challenging for developers due to the number of categories they have to choose
from. Machine Learning has been very useful, especially in classification
problems such sentiment analysis, document classification and spam detection.
These strategies can also be applied to app categorization on Google Play Store
to suggest appropriate categories for app publishers using details from their
application.
In this project, we built two variations of the Naive Bayes classifier using
open metadata from top developer apps on Google Play Store in other to classify
new apps on the store. These classifiers are then evaluated using various
evaluation methods and their results compared against each other. The results
show that the Naive Bayes algorithm performs well for our classification
problem and can potentially automate app categorization for Android app
publishers on Google Play Store
|
What makes ImageNet good for transfer learning? | The tremendous success of ImageNet-trained deep features on a wide range of
transfer tasks begs the question: what are the properties of the ImageNet
dataset that are critical for learning good, general-purpose features? This
work provides an empirical investigation of various facets of this question: Is
more pre-training data always better? How does feature quality depend on the
number of training examples per class? Does adding more object classes improve
performance? For the same data budget, how should the data be split into
classes? Is fine-grained recognition necessary for learning good features?
Given the same number of training classes, is it better to have coarse classes
or fine-grained classes? Which is better: more classes or more examples per
class? To answer these and related questions, we pre-trained CNN features on
various subsets of the ImageNet dataset and evaluated transfer performance on
PASCAL detection, PASCAL action classification, and SUN scene classification
tasks. Our overall findings suggest that most changes in the choice of
pre-training data long thought to be critical do not significantly affect
transfer performance.? Given the same number of training classes, is it better
to have coarse classes or fine-grained classes? Which is better: more classes
or more examples per class?
|
Pruning Filters for Efficient ConvNets | The success of CNNs in various applications is accompanied by a significant
increase in the computation and parameter storage costs. Recent efforts toward
reducing these overheads involve pruning and compressing the weights of various
layers without hurting original accuracy. However, magnitude-based pruning of
weights reduces a significant number of parameters from the fully connected
layers and may not adequately reduce the computation costs in the convolutional
layers due to irregular sparsity in the pruned networks. We present an
acceleration method for CNNs, where we prune filters from CNNs that are
identified as having a small effect on the output accuracy. By removing whole
filters in the network together with their connecting feature maps, the
computation costs are reduced significantly. In contrast to pruning weights,
this approach does not result in sparse connectivity patterns. Hence, it does
not need the support of sparse convolution libraries and can work with existing
efficient BLAS libraries for dense matrix multiplications. We show that even
simple filter pruning techniques can reduce inference costs for VGG-16 by up to
34% and ResNet-110 by up to 38% on CIFAR10 while regaining close to the
original accuracy by retraining the networks.
|
Measuring Machine Intelligence Through Visual Question Answering | As machines have become more intelligent, there has been a renewed interest
in methods for measuring their intelligence. A common approach is to propose
tasks for which a human excels, but one which machines find difficult. However,
an ideal task should also be easy to evaluate and not be easily gameable. We
begin with a case study exploring the recently popular task of image captioning
and its limitations as a task for measuring machine intelligence. An
alternative and more promising task is Visual Question Answering that tests a
machine's ability to reason about language and vision. We describe a dataset
unprecedented in size created for the task that contains over 760,000 human
generated questions about images. Using around 10 million human generated
answers, machines may be easily evaluated.
|
hi-RF: Incremental Learning Random Forest for large-scale multi-class
Data Classification | In recent years, dynamically growing data and incrementally growing number of
classes pose new challenges to large-scale data classification research. Most
traditional methods struggle to balance the precision and computational burden
when data and its number of classes increased. However, some methods are with
weak precision, and the others are time-consuming. In this paper, we propose an
incremental learning method, namely, heterogeneous incremental Nearest Class
Mean Random Forest (hi-RF), to handle this issue. It is a heterogeneous method
that either replaces trees or updates trees leaves in the random forest
adaptively, to reduce the computational time in comparable performance, when
data of new classes arrive. Specifically, to keep the accuracy, one proportion
of trees are replaced by new NCM decision trees; to reduce the computational
load, the rest trees are updated their leaves probabilities only. Most of all,
out-of-bag estimation and out-of-bag boosting are proposed to balance the
accuracy and the computational efficiency. Fair experiments were conducted and
demonstrated its comparable precision with much less computational time.
|
A Mathematical Framework for Feature Selection from Real-World Data with
Non-Linear Observations | In this paper, we study the challenge of feature selection based on a
relatively small collection of sample pairs $\{(x_i, y_i)\}_{1 \leq i \leq m}$.
The observations $y_i \in \mathbb{R}$ are thereby supposed to follow a noisy
single-index model, depending on a certain set of signal variables. A major
difficulty is that these variables usually cannot be observed directly, but
rather arise as hidden factors in the actual data vectors $x_i \in
\mathbb{R}^d$ (feature variables). We will prove that a successful variable
selection is still possible in this setup, even when the applied estimator does
not have any knowledge of the underlying model parameters and only takes the
'raw' samples $\{(x_i, y_i)\}_{1 \leq i \leq m}$ as input. The model
assumptions of our results will be fairly general, allowing for non-linear
observations, arbitrary convex signal structures as well as strictly convex
loss functions. This is particularly appealing for practical purposes, since in
many applications, already standard methods, e.g., the Lasso or logistic
regression, yield surprisingly good outcomes. Apart from a general discussion
of the practical scope of our theoretical findings, we will also derive a
rigorous guarantee for a specific real-world problem, namely sparse feature
extraction from (proteomics-based) mass spectrometry data.
|
A High Speed Multi-label Classifier based on Extreme Learning Machines | In this paper a high speed neural network classifier based on extreme
learning machines for multi-label classification problem is proposed and
dis-cussed. Multi-label classification is a superset of traditional binary and
multi-class classification problems. The proposed work extends the extreme
learning machine technique to adapt to the multi-label problems. As opposed to
the single-label problem, both the number of labels the sample belongs to, and
each of those target labels are to be identified for multi-label classification
resulting in in-creased complexity. The proposed high speed multi-label
classifier is applied to six benchmark datasets comprising of different
application areas such as multi-media, text and biology. The training time and
testing time of the classifier are compared with those of the state-of-the-arts
methods. Experimental studies show that for all the six datasets, our proposed
technique have faster execution speed and better performance, thereby
outperforming all the existing multi-label clas-sification methods.
|
A Novel Online Real-time Classifier for Multi-label Data Streams | In this paper, a novel extreme learning machine based online multi-label
classifier for real-time data streams is proposed. Multi-label classification
is one of the actively researched machine learning paradigm that has gained
much attention in the recent years due to its rapidly increasing real world
applications. In contrast to traditional binary and multi-class classification,
multi-label classification involves association of each of the input samples
with a set of target labels simultaneously. There are no real-time online
neural network based multi-label classifier available in the literature. In
this paper, we exploit the inherent nature of high speed exhibited by the
extreme learning machines to develop a novel online real-time classifier for
multi-label data streams. The developed classifier is experimented with
datasets from different application domains for consistency, performance and
speed. The experimental studies show that the proposed method outperforms the
existing state-of-the-art techniques in terms of speed and accuracy and can
classify multi-label data streams in real-time.
|
Recursive Partitioning for Personalization using Observational Data | We study the problem of learning to choose from m discrete treatment options
(e.g., news item or medical drug) the one with best causal effect for a
particular instance (e.g., user or patient) where the training data consists of
passive observations of covariates, treatment, and the outcome of the
treatment. The standard approach to this problem is regress and compare: split
the training data by treatment, fit a regression model in each split, and, for
a new instance, predict all m outcomes and pick the best. By reformulating the
problem as a single learning task rather than m separate ones, we propose a new
approach based on recursively partitioning the data into regimes where
different treatments are optimal. We extend this approach to an optimal
partitioning approach that finds a globally optimal partition, achieving a
compact, interpretable, and impactful personalization model. We develop new
tools for validating and evaluating personalization models on observational
data and use these to demonstrate the power of our novel approaches in a
personalized medicine and a job training application.
|
Hash2Vec, Feature Hashing for Word Embeddings | In this paper we propose the application of feature hashing to create word
embeddings for natural language processing. Feature hashing has been used
successfully to create document vectors in related tasks like document
classification. In this work we show that feature hashing can be applied to
obtain word embeddings in linear time with the size of the data. The results
show that this algorithm, that does not need training, is able to capture the
semantic meaning of words. We compare the results against GloVe showing that
they are similar. As far as we know this is the first application of feature
hashing to the word embeddings problem and the results indicate this is a
scalable technique with practical results for NLP applications.
|
Robustness of classifiers: from adversarial to random noise | Several recent works have shown that state-of-the-art classifiers are
vulnerable to worst-case (i.e., adversarial) perturbations of the datapoints.
On the other hand, it has been empirically observed that these same classifiers
are relatively robust to random noise. In this paper, we propose to study a
\textit{semi-random} noise regime that generalizes both the random and
worst-case noise regimes. We propose the first quantitative analysis of the
robustness of nonlinear classifiers in this general noise regime. We establish
precise theoretical bounds on the robustness of classifiers in this general
regime, which depend on the curvature of the classifier's decision boundary.
Our bounds confirm and quantify the empirical observations that classifiers
satisfying curvature constraints are robust to random noise. Moreover, we
quantify the robustness of classifiers in terms of the subspace dimension in
the semi-random noise regime, and show that our bounds remarkably interpolate
between the worst-case and random noise regimes. We perform experiments and
show that the derived bounds provide very accurate estimates when applied to
various state-of-the-art deep neural networks and datasets. This result
suggests bounds on the curvature of the classifiers' decision boundaries that
we support experimentally, and more generally offers important insights onto
the geometry of high dimensional classification problems.
|
Towards Transparent AI Systems: Interpreting Visual Question Answering
Models | Deep neural networks have shown striking progress and obtained
state-of-the-art results in many AI research fields in the recent years.
However, it is often unsatisfying to not know why they predict what they do. In
this paper, we address the problem of interpreting Visual Question Answering
(VQA) models. Specifically, we are interested in finding what part of the input
(pixels in images or words in questions) the VQA model focuses on while
answering the question. To tackle this problem, we use two visualization
techniques -- guided backpropagation and occlusion -- to find important words
in the question and important regions in the image. We then present qualitative
and quantitative analyses of these importance maps. We found that even without
explicit attention mechanisms, VQA models may sometimes be implicitly attending
to relevant regions in the image, and often to appropriate words in the
question.
|
Towards Competitive Classifiers for Unbalanced Classification Problems:
A Study on the Performance Scores | Although a great methodological effort has been invested in proposing
competitive solutions to the class-imbalance problem, little effort has been
made in pursuing a theoretical understanding of this matter.
In order to shed some light on this topic, we perform, through a novel
framework, an exhaustive analysis of the adequateness of the most commonly used
performance scores to assess this complex scenario. We conclude that using
unweighted H\"older means with exponent $p \leq 1$ to average the recalls of
all the classes produces adequate scores which are capable of determining
whether a classifier is competitive.
Then, we review the major solutions presented in the class-imbalance
literature. Since any learning task can be defined as an optimisation problem
where a loss function, usually connected to a particular score, is minimised,
our goal, here, is to find whether the learning tasks found in the literature
are also oriented to maximise the previously detected adequate scores. We
conclude that they usually maximise the unweighted H\"older mean with $p = 1$
(a-mean).
Finally, we provide bounds on the values of the studied performance scores
which guarantee a classifier with a higher recall than the random classifier in
each and every class.
|
Learning Syntactic Program Transformations from Examples | IDEs, such as Visual Studio, automate common transformations, such as Rename
and Extract Method refactorings. However, extending these catalogs of
transformations is complex and time-consuming. A similar phenomenon appears in
intelligent tutoring systems where instructors have to write cumbersome code
transformations that describe "common faults" to fix similar student
submissions to programming assignments. We present REFAZER, a technique for
automatically generating program transformations. REFAZER builds on the
observation that code edits performed by developers can be used as examples for
learning transformations. Example edits may share the same structure but
involve different variables and subexpressions, which must be generalized in a
transformation at the right level of abstraction. To learn transformations,
REFAZER leverages state-of-the-art programming-by-example methodology using the
following key components: (a) a novel domain-specific language (DSL) for
describing program transformations, (b) domain-specific deductive algorithms
for synthesizing transformations in the DSL, and (c) functions for ranking the
synthesized transformations. We instantiate and evaluate REFAZER in two
domains. First, given examples of edits used by students to fix incorrect
programming assignment submissions, we learn transformations that can fix other
students' submissions with similar faults. In our evaluation conducted on 4
programming tasks performed by 720 students, our technique helped to fix
incorrect submissions for 87% of the students. In the second domain, we use
repetitive edits applied by developers to the same project to synthesize a
program transformation that applies these edits to other locations in the code.
In our evaluation conducted on 59 scenarios of repetitive edits taken from 3 C#
open-source projects, REFAZER learns the intended program transformation in 83%
of the cases.
|
A Tutorial on Online Supervised Learning with Applications to Node
Classification in Social Networks | We revisit the elegant observation of T. Cover '65 which, perhaps, is not as
well-known to the broader community as it should be. The first goal of the
tutorial is to explain---through the prism of this elementary result---how to
solve certain sequence prediction problems by modeling sets of solutions rather
than the unknown data-generating mechanism. We extend Cover's observation in
several directions and focus on computational aspects of the proposed
algorithms. The applicability of the methods is illustrated on several
examples, including node classification in a network.
The second aim of this tutorial is to demonstrate the following phenomenon:
it is possible to predict as well as a combinatorial "benchmark" for which we
have a certain multiplicative approximation algorithm, even if the exact
computation of the benchmark given all the data is NP-hard. The proposed
prediction methods, therefore, circumvent some of the computational
difficulties associated with finding the best model given the data. These
difficulties arise rather quickly when one attempts to develop a probabilistic
model for graph-based or other problems with a combinatorial structure.
|
Neural Network Architecture Optimization through Submodularity and
Supermodularity | Deep learning models' architectures, including depth and width, are key
factors influencing models' performance, such as test accuracy and computation
time. This paper solves two problems: given computation time budget, choose an
architecture to maximize accuracy, and given accuracy requirement, choose an
architecture to minimize computation time. We convert this architecture
optimization into a subset selection problem. With accuracy's submodularity and
computation time's supermodularity, we propose efficient greedy optimization
algorithms. The experiments demonstrate our algorithm's ability to find more
accurate models or faster models. By analyzing architecture evolution with
growing time budget, we discuss relationships among accuracy, time and
architecture, and give suggestions on neural network architecture design.
|
A Novel Progressive Learning Technique for Multi-class Classification | In this paper, a progressive learning technique for multi-class
classification is proposed. This newly developed learning technique is
independent of the number of class constraints and it can learn new classes
while still retaining the knowledge of previous classes. Whenever a new class
(non-native to the knowledge learnt thus far) is encountered, the neural
network structure gets remodeled automatically by facilitating new neurons and
interconnections, and the parameters are calculated in such a way that it
retains the knowledge learnt thus far. This technique is suitable for
real-world applications where the number of classes is often unknown and online
learning from real-time data is required. The consistency and the complexity of
the progressive learning technique are analyzed. Several standard datasets are
used to evaluate the performance of the developed technique. A comparative
study shows that the developed technique is superior.
|
A novel online multi-label classifier for high-speed streaming data
applications | In this paper, a high-speed online neural network classifier based on extreme
learning machines for multi-label classification is proposed. In multi-label
classification, each of the input data sample belongs to one or more than one
of the target labels. The traditional binary and multi-class classification
where each sample belongs to only one target class forms the subset of
multi-label classification. Multi-label classification problems are far more
complex than binary and multi-class classification problems, as both the number
of target labels and each of the target labels corresponding to each of the
input samples are to be identified. The proposed work exploits the high-speed
nature of the extreme learning machines to achieve real-time multi-label
classification of streaming data. A new threshold-based online sequential
learning algorithm is proposed for high speed and streaming data classification
of multi-label problems. The proposed method is experimented with six different
datasets from different application domains such as multimedia, text, and
biology. The hamming loss, accuracy, training time and testing time of the
proposed technique is compared with nine different state-of-the-art methods.
Experimental studies shows that the proposed technique outperforms the existing
multi-label classifiers in terms of performance and speed.
|
Neural Coarse-Graining: Extracting slowly-varying latent degrees of
freedom with neural networks | We present a loss function for neural networks that encompasses an idea of
trivial versus non-trivial predictions, such that the network jointly
determines its own prediction goals and learns to satisfy them. This permits
the network to choose sub-sets of a problem which are most amenable to its
abilities to focus on solving, while discarding 'distracting' elements that
interfere with its learning. To do this, the network first transforms the raw
data into a higher-level categorical representation, and then trains a
predictor from that new time series to its future. To prevent a trivial
solution of mapping the signal to zero, we introduce a measure of
non-triviality via a contrast between the prediction error of the learned model
with a naive model of the overall signal statistics. The transform can learn to
discard uninformative and unpredictable components of the signal in favor of
the features which are both highly predictive and highly predictable. This
creates a coarse-grained model of the time-series dynamics, focusing on
predicting the slowly varying latent parameters which control the statistics of
the time-series, rather than predicting the fast details directly. The result
is a semi-supervised algorithm which is capable of extracting latent
parameters, segmenting sections of time-series with differing statistics, and
building a higher-level representation of the underlying dynamics from
unlabeled data.
|
Reward Augmented Maximum Likelihood for Neural Structured Prediction | A key problem in structured output prediction is direct optimization of the
task reward function that matters for test evaluation. This paper presents a
simple and computationally efficient approach to incorporate task reward into a
maximum likelihood framework. By establishing a link between the log-likelihood
and expected reward objectives, we show that an optimal regularized expected
reward is achieved when the conditional distribution of the outputs given the
inputs is proportional to their exponentiated scaled rewards. Accordingly, we
present a framework to smooth the predictive probability of the outputs using
their corresponding rewards. We optimize the conditional log-probability of
augmented outputs that are sampled proportionally to their exponentiated scaled
rewards. Experiments on neural sequence to sequence models for speech
recognition and machine translation show notable improvements over a maximum
likelihood baseline by using reward augmented maximum likelihood (RAML), where
the rewards are defined as the negative edit distance between the outputs and
the ground truth labels.
|
Employing traditional machine learning algorithms for big data streams
analysis: the case of object trajectory prediction | In this paper, we model the trajectory of sea vessels and provide a service
that predicts in near-real time the position of any given vessel in 4', 10',
20' and 40' time intervals. We explore the necessary tradeoffs between
accuracy, performance and resource utilization are explored given the large
volume and update rates of input data. We start with building models based on
well-established machine learning algorithms using static datasets and
multi-scan training approaches and identify the best candidate to be used in
implementing a single-pass predictive approach, under real-time constraints.
The results are measured in terms of accuracy and performance and are compared
against the baseline kinematic equations. Results show that it is possible to
efficiently model the trajectory of multiple vessels using a single model,
which is trained and evaluated using an adequately large, static dataset, thus
achieving a significant gain in terms of resource usage while not compromising
accuracy.
|
Ternary Neural Networks for Resource-Efficient AI Applications | The computation and storage requirements for Deep Neural Networks (DNNs) are
usually high. This issue limits their deployability on ubiquitous computing
devices such as smart phones, wearables and autonomous drones. In this paper,
we propose ternary neural networks (TNNs) in order to make deep learning more
resource-efficient. We train these TNNs using a teacher-student approach based
on a novel, layer-wise greedy methodology. Thanks to our two-stage training
procedure, the teacher network is still able to use state-of-the-art methods
such as dropout and batch normalization to increase accuracy and reduce
training time. Using only ternary weights and activations, the student ternary
network learns to mimic the behavior of its teacher network without using any
multiplication. Unlike its -1,1 binary counterparts, a ternary neural network
inherently prunes the smaller weights by setting them to zero during training.
This makes them sparser and thus more energy-efficient. We design a
purpose-built hardware architecture for TNNs and implement it on FPGA and ASIC.
We evaluate TNNs on several benchmark datasets and demonstrate up to 3.1x
better energy efficiency with respect to the state of the art while also
improving accuracy.
|
Testing $k$-Monotonicity | A Boolean $k$-monotone function defined over a finite poset domain ${\cal D}$
alternates between the values $0$ and $1$ at most $k$ times on any ascending
chain in ${\cal D}$. Therefore, $k$-monotone functions are natural
generalizations of the classical monotone functions, which are the $1$-monotone
functions. Motivated by the recent interest in $k$-monotone functions in the
context of circuit complexity and learning theory, and by the central role that
monotonicity testing plays in the context of property testing, we initiate a
systematic study of $k$-monotone functions, in the property testing model. In
this model, the goal is to distinguish functions that are $k$-monotone (or are
close to being $k$-monotone) from functions that are far from being
$k$-monotone. Our results include the following:
- We demonstrate a separation between testing $k$-monotonicity and testing
monotonicity, on the hypercube domain $\{0,1\}^d$, for $k\geq 3$;
- We demonstrate a separation between testing and learning on $\{0,1\}^d$,
for $k=\omega(\log d)$: testing $k$-monotonicity can be performed with
$2^{O(\sqrt d \cdot \log d\cdot \log{1/\varepsilon})}$ queries, while learning
$k$-monotone functions requires $2^{\Omega(k\cdot \sqrt
d\cdot{1/\varepsilon})}$ queries (Blais et al. (RANDOM 2015)).
- We present a tolerant test for functions $f\colon[n]^d\to \{0,1\}$ with
complexity independent of $n$, which makes progress on a problem left open by
Berman et al. (STOC 2014).
Our techniques exploit the testing-by-learning paradigm, use novel
applications of Fourier analysis on the grid $[n]^d$, and draw connections to
distribution testing techniques.
|
A Unified View of Multi-Label Performance Measures | Multi-label classification deals with the problem where each instance is
associated with multiple class labels. Because evaluation in multi-label
classification is more complicated than single-label setting, a number of
performance measures have been proposed. It is noticed that an algorithm
usually performs differently on different measures. Therefore, it is important
to understand which algorithms perform well on which measure(s) and why. In
this paper, we propose a unified margin view to revisit eleven performance
measures in multi-label classification. In particular, we define label-wise
margin and instance-wise margin, and prove that through maximizing these
margins, different corresponding performance measures will be optimized. Based
on the defined margins, a max-margin approach called LIMO is designed and
empirical results verify our theoretical findings.
|
Least Ambiguous Set-Valued Classifiers with Bounded Error Levels | In most classification tasks there are observations that are ambiguous and
therefore difficult to correctly label. Set-valued classifiers output sets of
plausible labels rather than a single label, thereby giving a more appropriate
and informative treatment to the labeling of ambiguous instances. We introduce
a framework for multiclass set-valued classification, where the classifiers
guarantee user-defined levels of coverage or confidence (the probability that
the true label is contained in the set) while minimizing the ambiguity (the
expected size of the output). We first derive oracle classifiers assuming the
true distribution to be known. We show that the oracle classifiers are obtained
from level sets of the functions that define the conditional probability of
each class. Then we develop estimators with good asymptotic and finite sample
properties. The proposed estimators build on existing single-label classifiers.
The optimal classifier can sometimes output the empty set, but we provide two
solutions to fix this issue that are suitable for various practical needs.
|
A deep learning model for estimating story points | Although there has been substantial research in software analytics for effort
estimation in traditional software projects, little work has been done for
estimation in agile projects, especially estimating user stories or issues.
Story points are the most common unit of measure used for estimating the effort
involved in implementing a user story or resolving an issue. In this paper, we
offer for the \emph{first} time a comprehensive dataset for story points-based
estimation that contains 23,313 issues from 16 open source projects. We also
propose a prediction model for estimating story points based on a novel
combination of two powerful deep learning architectures: long short-term memory
and recurrent highway network. Our prediction system is \emph{end-to-end}
trainable from raw input data to prediction outcomes without any manual feature
engineering. An empirical evaluation demonstrates that our approach
consistently outperforms three common effort estimation baselines and two
alternatives in both Mean Absolute Error and the Standardized Accuracy.
|
Doubly stochastic large scale kernel learning with the empirical kernel
map | With the rise of big data sets, the popularity of kernel methods declined and
neural networks took over again. The main problem with kernel methods is that
the kernel matrix grows quadratically with the number of data points. Most
attempts to scale up kernel methods solve this problem by discarding data
points or basis functions of some approximation of the kernel map. Here we
present a simple yet effective alternative for scaling up kernel methods that
takes into account the entire data set via doubly stochastic optimization of
the emprical kernel map. The algorithm is straightforward to implement, in
particular in parallel execution settings; it leverages the full power and
versatility of classical kernel functions without the need to explicitly
formulate a kernel map approximation. We provide empirical evidence that the
algorithm works on large data sets.
|
SEBOOST - Boosting Stochastic Learning Using Subspace Optimization
Techniques | We present SEBOOST, a technique for boosting the performance of existing
stochastic optimization methods. SEBOOST applies a secondary optimization
process in the subspace spanned by the last steps and descent directions. The
method was inspired by the SESOP optimization method for large-scale problems,
and has been adapted for the stochastic learning framework. It can be applied
on top of any existing optimization method with no need to tweak the internal
algorithm. We show that the method is able to boost the performance of
different algorithms, and make them more robust to changes in their
hyper-parameters. As the boosting steps of SEBOOST are applied between large
sets of descent steps, the additional subspace optimization hardly increases
the overall computational burden. We introduce two hyper-parameters that
control the balance between the baseline method and the secondary optimization
process. The method was evaluated on several deep learning tasks, demonstrating
promising results.
|
Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep
Learning Model | Recently exciting progress has been made on protein contact prediction, but
the predicted contacts for proteins without many sequence homologs is still of
low quality and not very useful for de novo structure prediction. This paper
presents a new deep learning method that predicts contacts by integrating both
evolutionary coupling (EC) and sequence conservation information through an
ultra-deep neural network formed by two deep residual networks. This deep
neural network allows us to model very complex sequence-contact relationship as
well as long-range inter-contact correlation. Our method greatly outperforms
existing contact prediction methods and leads to much more accurate
contact-assisted protein folding. Tested on three datasets of 579 proteins, the
average top L long-range prediction accuracy obtained our method, the
representative EC method CCMpred and the CASP11 winner MetaPSICOV is 0.47, 0.21
and 0.30, respectively; the average top L/10 long-range accuracy of our method,
CCMpred and MetaPSICOV is 0.77, 0.47 and 0.59, respectively. Ab initio folding
using our predicted contacts as restraints can yield correct folds (i.e.,
TMscore>0.6) for 203 test proteins, while that using MetaPSICOV- and
CCMpred-predicted contacts can do so for only 79 and 62 proteins, respectively.
Further, our contact-assisted models have much better quality than
template-based models. Using our predicted contacts as restraints, we can (ab
initio) fold 208 of the 398 membrane proteins with TMscore>0.5. By contrast,
when the training proteins of our method are used as templates, homology
modeling can only do so for 10 of them. One interesting finding is that even if
we do not train our prediction models with any membrane proteins, our method
works very well on membrane protein prediction. Finally, in recent blind CAMEO
benchmark our method successfully folded 5 test proteins with a novel fold.
|
Single photon in hierarchical architecture for physical reinforcement
learning: Photon intelligence | Understanding and using natural processes for intelligent functionalities,
referred to as natural intelligence, has recently attracted interest from a
variety of fields, including post-silicon computing for artificial intelligence
and decision making in the behavioural sciences. In a past study, we
successfully used the wave-particle duality of single photons to solve the
two-armed bandit problem, which constitutes the foundation of reinforcement
learning and decision making. In this study, we propose and confirm a
hierarchical architecture for single-photon-based reinforcement learning and
decision making that verifies the scalability of the principle. Specifically,
the four-armed bandit problem is solved given zero prior knowledge in a
two-layer hierarchical architecture, where polarization is autonomously adapted
in order to effect adequate decision making using single-photon measurements.
In the hierarchical structure, the notion of layer-dependent decisions emerges.
The optimal solutions in the coarse layer and in the fine layer, however,
conflict with each other in some contradictive problems. We show that while
what we call a tournament strategy resolves such contradictions, the
probabilistic nature of single photons allows for the direct location of the
optimal solution even for contradictive problems, hence manifesting the
exploration ability of single photons. This study provides insights into photon
intelligence in hierarchical architectures for future artificial intelligence
as well as the potential of natural processes for intelligent functionalities.
|
Convolutional Neural Networks for Text Categorization: Shallow
Word-level vs. Deep Character-level | This paper reports the performances of shallow word-level convolutional
neural networks (CNN), our earlier work (2015), on the eight datasets with
relatively large training data that were used for testing the very deep
character-level CNN in Conneau et al. (2016). Our findings are as follows. The
shallow word-level CNNs achieve better error rates than the error rates
reported in Conneau et al., though the results should be interpreted with some
consideration due to the unique pre-processing of Conneau et al. The shallow
word-level CNN uses more parameters and therefore requires more storage than
the deep character-level CNN; however, the shallow word-level CNN computes much
faster.
|
Towards End-to-End Reinforcement Learning of Dialogue Agents for
Information Access | This paper proposes KB-InfoBot -- a multi-turn dialogue agent which helps
users search Knowledge Bases (KBs) without composing complicated queries. Such
goal-oriented dialogue agents typically need to interact with an external
database to access real-world knowledge. Previous systems achieved this by
issuing a symbolic query to the KB to retrieve entries based on their
attributes. However, such symbolic operations break the differentiability of
the system and prevent end-to-end training of neural dialogue agents. In this
paper, we address this limitation by replacing symbolic queries with an induced
"soft" posterior distribution over the KB that indicates which entities the
user is interested in. Integrating the soft retrieval process with a
reinforcement learner leads to higher task success rate and reward in both
simulations and against real users. We also present a fully neural end-to-end
agent, trained entirely from user feedback, and discuss its application towards
personalized dialogue agents. The source code is available at
https://github.com/MiuLab/KB-InfoBot.
|
Randomized Prediction Games for Adversarial Machine Learning | In spam and malware detection, attackers exploit randomization to obfuscate
malicious data and increase their chances of evading detection at test time;
e.g., malware code is typically obfuscated using random strings or byte
sequences to hide known exploits. Interestingly, randomization has also been
proposed to improve security of learning algorithms against evasion attacks, as
it results in hiding information about the classifier to the attacker. Recent
work has proposed game-theoretical formulations to learn secure classifiers, by
simulating different evasion attacks and modifying the classification function
accordingly. However, both the classification function and the simulated data
manipulations have been modeled in a deterministic manner, without accounting
for any form of randomization. In this work, we overcome this limitation by
proposing a randomized prediction game, namely, a non-cooperative
game-theoretic formulation in which the classifier and the attacker make
randomized strategy selections according to some probability distribution
defined over the respective strategy set. We show that our approach allows one
to improve the trade-off between attack detection and false alarms with respect
to state-of-the-art secure classifiers, even against attacks that are different
from those hypothesized during design, on application examples including
handwritten digit recognition, spam and malware detection.
|
An Online Universal Classifier for Binary, Multi-class and Multi-label
Classification | Classification involves the learning of the mapping function that associates
input samples to corresponding target label. There are two major categories of
classification problems: Single-label classification and Multi-label
classification. Traditional binary and multi-class classifications are
sub-categories of single-label classification. Several classifiers are
developed for binary, multi-class and multi-label classification problems, but
there are no classifiers available in the literature capable of performing all
three types of classification. In this paper, a novel online universal
classifier capable of performing all the three types of classification is
proposed. Being a high speed online classifier, the proposed technique can be
applied to streaming data applications. The performance of the developed
classifier is evaluated using datasets from binary, multi-class and multi-label
problems. The results obtained are compared with state-of-the-art techniques
from each of the classification types.
|
Graph-Based Active Learning: A New Look at Expected Error Minimization | In graph-based active learning, algorithms based on expected error
minimization (EEM) have been popular and yield good empirical performance. The
exact computation of EEM optimally balances exploration and exploitation. In
practice, however, EEM-based algorithms employ various approximations due to
the computational hardness of exact EEM. This can result in a lack of either
exploration or exploitation, which can negatively impact the effectiveness of
active learning. We propose a new algorithm TSA (Two-Step Approximation) that
balances between exploration and exploitation efficiently while enjoying the
same computational complexity as existing approximations. Finally, we
empirically show the value of balancing between exploration and exploitation in
both toy and real-world datasets where our method outperforms several
state-of-the-art methods.
|
A Probabilistic Optimum-Path Forest Classifier for Binary Classification
Problems | Probabilistic-driven classification techniques extend the role of traditional
approaches that output labels (usually integer numbers) only. Such techniques
are more fruitful when dealing with problems where one is not interested in
recognition/identification only, but also into monitoring the behavior of
consumers and/or machines, for instance. Therefore, by means of probability
estimates, one can take decisions to work better in a number of scenarios. In
this paper, we propose a probabilistic-based Optimum Path Forest (OPF)
classifier to handle with binary classification problems, and we show it can be
more accurate than naive OPF in a number of datasets. In addition to being just
more accurate or not, probabilistic OPF turns to be another useful tool to the
scientific community.
|
High Dimensional Human Guided Machine Learning | Have you ever looked at a machine learning classification model and thought,
I could have made that? Well, that is what we test in this project, comparing
XGBoost trained on human engineered features to training directly on data. The
human engineered features do not outperform XGBoost trained di- rectly on the
data, but they are comparable. This project con- tributes a novel method for
utilizing human created classifi- cation models on high dimensional datasets.
|
Decoding visual stimuli in human brain by using Anatomical Pattern
Analysis on fMRI images | A universal unanswered question in neuroscience and machine learning is
whether computers can decode the patterns of the human brain. Multi-Voxels
Pattern Analysis (MVPA) is a critical tool for addressing this question.
However, there are two challenges in the previous MVPA methods, which include
decreasing sparsity and noises in the extracted features and increasing the
performance of prediction. In overcoming mentioned challenges, this paper
proposes Anatomical Pattern Analysis (APA) for decoding visual stimuli in the
human brain. This framework develops a novel anatomical feature extraction
method and a new imbalance AdaBoost algorithm for binary classification.
Further, it utilizes an Error-Correcting Output Codes (ECOC) method for
multi-class prediction. APA can automatically detect active regions for each
category of the visual stimuli. Moreover, it enables us to combine homogeneous
datasets for applying advanced classification. Experimental studies on 4 visual
categories (words, consonants, objects and scrambled photos) demonstrate that
the proposed approach achieves superior performance to state-of-the-art
methods.
|
Spectral learning of dynamic systems from nonequilibrium data | Observable operator models (OOMs) and related models are one of the most
important and powerful tools for modeling and analyzing stochastic systems.
They exactly describe dynamics of finite-rank systems and can be efficiently
and consistently estimated through spectral learning under the assumption of
identically distributed data. In this paper, we investigate the properties of
spectral learning without this assumption due to the requirements of analyzing
large-time scale systems, and show that the equilibrium dynamics of a system
can be extracted from nonequilibrium observation data by imposing an
equilibrium constraint. In addition, we propose a binless extension of spectral
learning for continuous data. In comparison with the other continuous-valued
spectral algorithms, the binless algorithm can achieve consistent estimation of
equilibrium dynamics with only linear complexity.
|
Local Maxima in the Likelihood of Gaussian Mixture Models: Structural
Results and Algorithmic Consequences | We provide two fundamental results on the population (infinite-sample)
likelihood function of Gaussian mixture models with $M \geq 3$ components. Our
first main result shows that the population likelihood function has bad local
maxima even in the special case of equally-weighted mixtures of well-separated
and spherical Gaussians. We prove that the log-likelihood value of these bad
local maxima can be arbitrarily worse than that of any global optimum, thereby
resolving an open question of Srebro (2007). Our second main result shows that
the EM algorithm (or a first-order variant of it) with random initialization
will converge to bad critical points with probability at least
$1-e^{-\Omega(M)}$. We further establish that a first-order variant of EM will
not converge to strict saddle points almost surely, indicating that the poor
performance of the first-order method can be attributed to the existence of bad
local maxima rather than bad saddle points. Overall, our results highlight the
necessity of careful initialization when using the EM algorithm in practice,
even when applied in highly favorable settings.
|
Convexified Convolutional Neural Networks | We describe the class of convexified convolutional neural networks (CCNNs),
which capture the parameter sharing of convolutional neural networks in a
convex manner. By representing the nonlinear convolutional filters as vectors
in a reproducing kernel Hilbert space, the CNN parameters can be represented as
a low-rank matrix, which can be relaxed to obtain a convex optimization
problem. For learning two-layer convolutional neural networks, we prove that
the generalization error obtained by a convexified CNN converges to that of the
best possible CNN. For learning deeper networks, we train CCNNs in a layer-wise
manner. Empirically, CCNNs achieve performance competitive with CNNs trained by
backpropagation, SVMs, fully-connected neural networks, stacked denoising
auto-encoders, and other baseline methods.
|
Distribution-Specific Hardness of Learning Neural Networks | Although neural networks are routinely and successfully trained in practice
using simple gradient-based methods, most existing theoretical results are
negative, showing that learning such networks is difficult, in a worst-case
sense over all data distributions. In this paper, we take a more nuanced view,
and consider whether specific assumptions on the "niceness" of the input
distribution, or "niceness" of the target function (e.g. in terms of
smoothness, non-degeneracy, incoherence, random choice of parameters etc.), are
sufficient to guarantee learnability using gradient-based methods. We provide
evidence that neither class of assumptions alone is sufficient: On the one
hand, for any member of a class of "nice" target functions, there are difficult
input distributions. On the other hand, we identify a family of simple target
functions, which are difficult to learn even if the input distribution is
"nice". To prove our results, we develop some tools which may be of independent
interest, such as extending Fourier-based hardness techniques developed in the
context of statistical queries \cite{blum1994weakly}, from the Boolean cube to
Euclidean space and to more general classes of functions.
|
Classifying and sorting cluttered piles of unknown objects with robots:
a learning approach | We consider the problem of sorting a densely cluttered pile of unknown
objects using a robot. This yet unsolved problem is relevant in the robotic
waste sorting business.
By extending previous active learning approaches to grasping, we show a
system that learns the task autonomously. Instead of predicting just whether a
grasp succeeds, we predict the classes of the objects that end up being picked
and thrown onto the target conveyor. Segmenting and identifying objects from
the uncluttered target conveyor, as opposed to the working area, is easier due
to the added structure since the thrown objects will be the only ones present.
Instead of trying to segment or otherwise understand the cluttered working
area in any way, we simply allow the controller to learn a mapping from an RGBD
image in the neighborhood of the grasp to a predicted result---all segmentation
etc. in the working area is implicit in the learned function. The grasp
selection operates in two stages: The first stage is hardcoded and outputs a
distribution of possible grasps that sometimes succeed. The second stage uses a
purely learned criterion to choose the grasp to make from the proposal
distribution created by the first stage.
In an experiment, the system quickly learned to make good pickups and predict
correctly, in advance, which class of object it was going to pick up and was
able to sort the objects from a densely cluttered pile by color.
|
The Player Kernel: Learning Team Strengths Based on Implicit Player
Contributions | In this work, we draw attention to a connection between skill-based models of
game outcomes and Gaussian process classification models. The Gaussian process
perspective enables a) a principled way of dealing with uncertainty and b) rich
models, specified through kernel functions. Using this connection, we tackle
the problem of predicting outcomes of football matches between national teams.
We develop a player kernel that relates any two football matches through the
players lined up on the field. This makes it possible to share knowledge gained
from observing matches between clubs (available in large quantities) and
matches between national teams (available only in limited quantities). We
evaluate our approach on the Euro 2008, 2012 and 2016 final tournaments.
|
Live Orchestral Piano, a system for real-time orchestral music
generation | This paper introduces the first system for performing automatic orchestration
based on a real-time piano input. We believe that it is possible to learn the
underlying regularities existing between piano scores and their orchestrations
by notorious composers, in order to automatically perform this task on novel
piano inputs. To that end, we investigate a class of statistical inference
models called conditional Restricted Boltzmann Machine (cRBM). We introduce a
specific evaluation framework for orchestral generation based on a prediction
task in order to assess the quality of different models. As prediction and
creation are two widely different endeavours, we discuss the potential biases
in evaluating temporal generative models through prediction tasks and their
impact on a creative system. Finally, we introduce an implementation of the
proposed model called Live Orchestral Piano (LOP), which allows to perform
real-time projective orchestration of a MIDI keyboard input.
|
The Robustness of Estimator Composition | We formalize notions of robustness for composite estimators via the notion of
a breakdown point. A composite estimator successively applies two (or more)
estimators: on data decomposed into disjoint parts, it applies the first
estimator on each part, then the second estimator on the outputs of the first
estimator. And so on, if the composition is of more than two estimators.
Informally, the breakdown point is the minimum fraction of data points which if
significantly modified will also significantly modify the output of the
estimator, so it is typically desirable to have a large breakdown point. Our
main result shows that, under mild conditions on the individual estimators, the
breakdown point of the composite estimator is the product of the breakdown
points of the individual estimators. We also demonstrate several scenarios,
ranging from regression to statistical testing, where this analysis is easy to
apply, useful in understanding worst case robustness, and sheds powerful
insights onto the associated data analysis.
|
Evolutionary Synthesis of Deep Neural Networks via Synaptic
Cluster-driven Genetic Encoding | There has been significant recent interest towards achieving highly efficient
deep neural network architectures. A promising paradigm for achieving this is
the concept of evolutionary deep intelligence, which attempts to mimic
biological evolution processes to synthesize highly-efficient deep neural
networks over successive generations. An important aspect of evolutionary deep
intelligence is the genetic encoding scheme used to mimic heredity, which can
have a significant impact on the quality of offspring deep neural networks.
Motivated by the neurobiological phenomenon of synaptic clustering, we
introduce a new genetic encoding scheme where synaptic probability is driven
towards the formation of a highly sparse set of synaptic clusters. Experimental
results for the task of image classification demonstrated that the synthesized
offspring networks using this synaptic cluster-driven genetic encoding scheme
can achieve state-of-the-art performance while having network architectures
that are not only significantly more efficient (with a ~125-fold decrease in
synapses for MNIST) compared to the original ancestor network, but also
tailored for GPU-accelerated machine learning applications.
|
Learning Model Predictive Control for iterative tasks. A Data-Driven
Control Framework | A Learning Model Predictive Controller (LMPC) for iterative tasks is
presented. The controller is reference-free and is able to improve its
performance by learning from previous iterations. A safe set and a terminal
cost function are used in order to guarantee recursive feasibility and
non-increasing performance at each iteration. The paper presents the control
design approach, and shows how to recursively construct terminal set and
terminal cost from state and input trajectories of previous iterations.
Simulation results show the effectiveness of the proposed control logic.
|
Q-Learning with Basic Emotions | Q-learning is a simple and powerful tool in solving dynamic problems where
environments are unknown. It uses a balance of exploration and exploitation to
find an optimal solution to the problem. In this paper, we propose using four
basic emotions: joy, sadness, fear, and anger to influence a Qlearning agent.
Simulations show that the proposed affective agent requires lesser number of
steps to find the optimal path. We found when affective agent finds the optimal
path, the ratio between exploration to exploitation gradually decreases,
indicating lower total step count in the long run
|
Towards Learning and Verifying Invariants of Cyber-Physical Systems by
Code Mutation | Cyber-physical systems (CPS), which integrate algorithmic control with
physical processes, often consist of physically distributed components
communicating over a network. A malfunctioning or compromised component in such
a CPS can lead to costly consequences, especially in the context of public
infrastructure. In this short paper, we argue for the importance of
constructing invariants (or models) of the physical behaviour exhibited by CPS,
motivated by their applications to the control, monitoring, and attestation of
components. To achieve this despite the inherent complexity of CPS, we propose
a new technique for learning invariants that combines machine learning with
ideas from mutation testing. We present a preliminary study on a water
treatment system that suggests the efficacy of this approach, propose
strategies for establishing confidence in the correctness of invariants, then
summarise some research questions and the steps we are taking to investigate
them.
|
Low-rank Bandits with Latent Mixtures | We study the task of maximizing rewards from recommending items (actions) to
users sequentially interacting with a recommender system. Users are modeled as
latent mixtures of C many representative user classes, where each class
specifies a mean reward profile across actions. Both the user features (mixture
distribution over classes) and the item features (mean reward vector per class)
are unknown a priori. The user identity is the only contextual information
available to the learner while interacting. This induces a low-rank structure
on the matrix of expected rewards r a,b from recommending item a to user b. The
problem reduces to the well-known linear bandit when either user or item-side
features are perfectly known. In the setting where each user, with its
stochastically sampled taste profile, interacts only for a small number of
sessions, we develop a bandit algorithm for the two-sided uncertainty. It
combines the Robust Tensor Power Method of Anandkumar et al. (2014b) with the
OFUL linear bandit algorithm of Abbasi-Yadkori et al. (2011). We provide the
first rigorous regret analysis of this combination, showing that its regret
after T user interactions is $\tilde O(C\sqrt{BT})$, with B the number of
users. An ingredient towards this result is a novel robustness property of
OFUL, of independent interest.
|
A Bootstrap Machine Learning Approach to Identify Rare Disease Patients
from Electronic Health Records | Rare diseases are very difficult to identify among large number of other
possible diagnoses. Better availability of patient data and improvement in
machine learning algorithms empower us to tackle this problem computationally.
In this paper, we target one such rare disease - cardiac amyloidosis. We aim to
automate the process of identifying potential cardiac amyloidosis patients with
the help of machine learning algorithms and also learn most predictive factors.
With the help of experienced cardiologists, we prepared a gold standard with 73
positive (cardiac amyloidosis) and 197 negative instances. We achieved high
average cross-validation F1 score of 0.98 using an ensemble machine learning
classifier. Some of the predictive variables were: Age and Diagnosis of cardiac
arrest, chest pain, congestive heart failure, hypertension, prim open angle
glaucoma, and shoulder arthritis. Further studies are needed to validate the
accuracy of the system across an entire health system and its generalizability
for other diseases.
|
Direct Feedback Alignment Provides Learning in Deep Neural Networks | Artificial neural networks are most commonly trained with the
back-propagation algorithm, where the gradient for learning is provided by
back-propagating the error, layer by layer, from the output layer to the hidden
layers. A recently discovered method called feedback-alignment shows that the
weights used for propagating the error backward don't have to be symmetric with
the weights used for propagation the activation forward. In fact, random
feedback weights work evenly well, because the network learns how to make the
feedback useful. In this work, the feedback alignment principle is used for
training hidden layers more independently from the rest of the network, and
from a zero initial condition. The error is propagated through fixed random
feedback connections directly from the output layer to each hidden layer. This
simple method is able to achieve zero training error even in convolutional
networks and very deep networks, completely without error back-propagation. The
method is a step towards biologically plausible machine learning because the
error signal is almost local, and no symmetric or reciprocal weights are
required. Experiments show that the test performance on MNIST and CIFAR is
almost as good as those obtained with back-propagation for fully connected
networks. If combined with dropout, the method achieves 1.45% error on the
permutation invariant MNIST task.
|
Hierarchical Multiscale Recurrent Neural Networks | Learning both hierarchical and temporal representation has been among the
long-standing challenges of recurrent neural networks. Multiscale recurrent
neural networks have been considered as a promising approach to resolve this
issue, yet there has been a lack of empirical evidence showing that this type
of models can actually capture the temporal dependencies by discovering the
latent hierarchical structure of the sequence. In this paper, we propose a
novel multiscale approach, called the hierarchical multiscale recurrent neural
networks, which can capture the latent hierarchical structure in the sequence
by encoding the temporal dependencies with different timescales using a novel
update mechanism. We show some evidence that our proposed multiscale
architecture can discover underlying hierarchical structure in the sequences
without using explicit boundary information. We evaluate our proposed model on
character-level language modelling and handwriting sequence modelling.
|
Semantic Video Trailers | Query-based video summarization is the task of creating a brief visual
trailer, which captures the parts of the video (or a collection of videos) that
are most relevant to the user-issued query. In this paper, we propose an
unsupervised label propagation approach for this task. Our approach effectively
captures the multimodal semantics of queries and videos using state-of-the-art
deep neural networks and creates a summary that is both semantically coherent
and visually attractive. We describe the theoretical framework of our
graph-based approach and empirically evaluate its effectiveness in creating
relevant and attractive trailers. Finally, we showcase example video trailers
generated by our system.
|
Learning Boltzmann Machine with EM-like Method | We propose an expectation-maximization-like(EMlike) method to train Boltzmann
machine with unconstrained connectivity. It adopts Monte Carlo approximation in
the E-step, and replaces the intractable likelihood objective with efficiently
computed objectives or directly approximates the gradient of likelihood
objective in the M-step. The EM-like method is a modification of alternating
minimization. We prove that EM-like method will be the exactly same with
contrastive divergence in restricted Boltzmann machine if the M-step of this
method adopts special approximation. We also propose a new measure to assess
the performance of Boltzmann machine as generative models of data, and its
computational complexity is O(Rmn). Finally, we demonstrate the performance of
EM-like method using numerical experiments.
|
Chaining Bounds for Empirical Risk Minimization | This paper extends the standard chaining technique to prove excess risk upper
bounds for empirical risk minimization with random design settings even if the
magnitude of the noise and the estimates is unbounded. The bound applies to
many loss functions besides the squared loss, and scales only with the
sub-Gaussian or subexponential parameters without further statistical
assumptions such as the bounded kurtosis condition over the hypothesis class. A
detailed analysis is provided for slope constrained and penalized linear least
squares regression with a sub-Gaussian setting, which often proves tight sample
complexity bounds up to logartihmic factors.
|
Polysemous codes | This paper considers the problem of approximate nearest neighbor search in
the compressed domain. We introduce polysemous codes, which offer both the
distance estimation quality of product quantization and the efficient
comparison of binary codes with Hamming distance. Their design is inspired by
algorithms introduced in the 90's to construct channel-optimized vector
quantizers. At search time, this dual interpretation accelerates the search.
Most of the indexed vectors are filtered out with Hamming distance, letting
only a fraction of the vectors to be ranked with an asymmetric distance
estimator.
The method is complementary with a coarse partitioning of the feature space
such as the inverted multi-index. This is shown by our experiments performed on
several public benchmarks such as the BIGANN dataset comprising one billion
vectors, for which we report state-of-the-art results for query times below
0.3\,millisecond per core. Last but not least, our approach allows the
approximate computation of the k-NN graph associated with the Yahoo Flickr
Creative Commons 100M, described by CNN image descriptors, in less than 8 hours
on a single machine.
|
DAiSEE: Towards User Engagement Recognition in the Wild | We introduce DAiSEE, the first multi-label video classification dataset
comprising of 9068 video snippets captured from 112 users for recognizing the
user affective states of boredom, confusion, engagement, and frustration in the
wild. The dataset has four levels of labels namely - very low, low, high, and
very high for each of the affective states, which are crowd annotated and
correlated with a gold standard annotation created using a team of expert
psychologists. We have also established benchmark results on this dataset using
state-of-the-art video classification methods that are available today. We
believe that DAiSEE will provide the research community with challenges in
feature extraction, context-based inference, and development of suitable
machine learning methods for related tasks, thus providing a springboard for
further research. The dataset is available for download at
https://iith.ac.in/~daisee-dataset
|
Doubly Stochastic Neighbor Embedding on Spheres | Stochastic Neighbor Embedding (SNE) methods minimize the divergence between
the similarity matrix of a high-dimensional data set and its counterpart from a
low-dimensional embedding, leading to widely applied tools for data
visualization. Despite their popularity, the current SNE methods experience a
crowding problem when the data include highly imbalanced similarities. This
implies that the data points with higher total similarity tend to get crowded
around the display center. To solve this problem, we introduce a fast
normalization method and normalize the similarity matrix to be doubly
stochastic such that all the data points have equal total similarities.
Furthermore, we show empirically and theoretically that the doubly
stochasticity constraint often leads to embeddings which are approximately
spherical. This suggests replacing a flat space with spheres as the embedding
space. The spherical embedding eliminates the discrepancy between the center
and the periphery in visualization, which efficiently resolves the crowding
problem. We compared the proposed method (DOSNES) with the state-of-the-art SNE
method on three real-world datasets and the results clearly indicate that our
method is more favorable in terms of visualization quality.
|
Human Body Orientation Estimation using Convolutional Neural Network | Personal robots are expected to interact with the user by recognizing the
user's face. However, in most of the service robot applications, the user needs
to move himself/herself to allow the robot to see him/her face to face. To
overcome such limitations, a method for estimating human body orientation is
required. Previous studies used various components such as feature extractors
and classification models to classify the orientation which resulted in low
performance. For a more robust and accurate approach, we propose the light
weight convolutional neural networks, an end to end system, for estimating
human body orientation. Our body orientation estimation model achieved 81.58%
and 94% accuracy with the benchmark dataset and our own dataset respectively.
The proposed method can be used in a wide range of service robot applications
which depend on the ability to estimate human body orientation. To show its
usefulness in service robot applications, we designed a simple robot
application which allows the robot to move towards the user's frontal plane.
With this, we demonstrated an improved face detection rate.
|
Random matrices meet machine learning: a large dimensional analysis of
LS-SVM | This article proposes a performance analysis of kernel least squares support
vector machines (LS-SVMs) based on a random matrix approach, in the regime
where both the dimension of data $p$ and their number $n$ grow large at the
same rate. Under a two-class Gaussian mixture model for the input data, we
prove that the LS-SVM decision function is asymptotically normal with means and
covariances shown to depend explicitly on the derivatives of the kernel
function. This provides improved understanding along with new insights into the
internal workings of SVM-type methods for large datasets.
|
Deep Markov Random Field for Image Modeling | Markov Random Fields (MRFs), a formulation widely used in generative image
modeling, have long been plagued by the lack of expressive power. This issue is
primarily due to the fact that conventional MRFs formulations tend to use
simplistic factors to capture local patterns. In this paper, we move beyond
such limitations, and propose a novel MRF model that uses fully-connected
neurons to express the complex interactions among pixels. Through theoretical
analysis, we reveal an inherent connection between this model and recurrent
neural networks, and thereon derive an approximated feed-forward network that
couples multiple RNNs along opposite directions. This formulation combines the
expressive power of deep neural networks and the cyclic dependency structure of
MRF in a unified model, bringing the modeling capability to a new level. The
feed-forward approximation also allows it to be efficiently learned from data.
Experimental results on a variety of low-level vision tasks show notable
improvement over state-of-the-arts.
|
An improved uncertainty decoding scheme with weighted samples for
DNN-HMM hybrid systems | In this paper, we advance a recently-proposed uncertainty decoding scheme for
DNN-HMM (deep neural network - hidden Markov model) hybrid systems. This
numerical sampling concept averages DNN outputs produced by a finite set of
feature samples (drawn from a probabilistic distortion model) to approximate
the posterior likelihoods of the context-dependent HMM states. As main
innovation, we propose a weighted DNN-output averaging based on a minimum
classification error criterion and apply it to a probabilistic distortion model
for spatial diffuseness features. The experimental evaluation is performed on
the 8-channel REVERB Challenge task using a DNN-HMM hybrid system with
multichannel front-end signal enhancement. We show that the recognition
accuracy of the DNN-HMM hybrid system improves by incorporating uncertainty
decoding based on random sampling and that the proposed weighted DNN-output
averaging further reduces the word error rate scores.
|
Ask the GRU: Multi-Task Learning for Deep Text Recommendations | In a variety of application domains the content to be recommended to users is
associated with text. This includes research papers, movies with associated
plot summaries, news articles, blog posts, etc. Recommendation approaches based
on latent factor models can be extended naturally to leverage text by employing
an explicit mapping from text to factors. This enables recommendations for new,
unseen content, and may generalize better, since the factors for all items are
produced by a compactly-parametrized model. Previous work has used topic models
or averages of word embeddings for this mapping. In this paper we present a
method leveraging deep recurrent neural networks to encode the text sequence
into a latent vector, specifically gated recurrent units (GRUs) trained
end-to-end on the collaborative filtering task. For the task of scientific
paper recommendation, this yields models with significantly higher accuracy. In
cold-start scenarios, we beat the previous state-of-the-art, all of which
ignore word order. Performance is further improved by multi-task learning,
where the text encoder network is trained for a combination of content
recommendation and item metadata prediction. This regularizes the collaborative
filtering model, ameliorating the problem of sparsity of the observed rating
matrix.
|
UberNet: Training a `Universal' Convolutional Neural Network for Low-,
Mid-, and High-Level Vision using Diverse Datasets and Limited Memory | In this work we introduce a convolutional neural network (CNN) that jointly
handles low-, mid-, and high-level vision tasks in a unified architecture that
is trained end-to-end. Such a universal network can act like a `swiss knife'
for vision tasks; we call this architecture an UberNet to indicate its
overarching nature.
We address two main technical challenges that emerge when broadening up the
range of tasks handled by a single CNN: (i) training a deep architecture while
relying on diverse training sets and (ii) training many (potentially unlimited)
tasks with a limited memory budget. Properly addressing these two problems
allows us to train accurate predictors for a host of tasks, without
compromising accuracy.
Through these advances we train in an end-to-end manner a CNN that
simultaneously addresses (a) boundary detection (b) normal estimation (c)
saliency estimation (d) semantic segmentation (e) human part segmentation (f)
semantic boundary detection, (g) region proposal generation and object
detection. We obtain competitive performance while jointly addressing all of
these tasks in 0.7 seconds per frame on a single GPU. A demonstration of this
system can be found at http://cvn.ecp.fr/ubernet/.
|
Discrete Variational Autoencoders | Probabilistic models with discrete latent variables naturally capture
datasets composed of discrete classes. However, they are difficult to train
efficiently, since backpropagation through discrete variables is generally not
possible. We present a novel method to train a class of probabilistic models
with discrete latent variables using the variational autoencoder framework,
including backpropagation through the discrete latent variables. The associated
class of probabilistic models comprises an undirected discrete component and a
directed hierarchical continuous component. The discrete component captures the
distribution over the disconnected smooth manifolds induced by the continuous
component. As a result, this class of models efficiently learns both the class
of objects in an image, and their specific realization in pixels, from
unsupervised data, and outperforms state-of-the-art methods on the
permutation-invariant MNIST, Omniglot, and Caltech-101 Silhouettes datasets.
|
Breaking the Bandwidth Barrier: Geometrical Adaptive Entropy Estimation | Estimators of information theoretic measures such as entropy and mutual
information are a basic workhorse for many downstream applications in modern
data science. State of the art approaches have been either geometric (nearest
neighbor (NN) based) or kernel based (with a globally chosen bandwidth). In
this paper, we combine both these approaches to design new estimators of
entropy and mutual information that outperform state of the art methods. Our
estimator uses local bandwidth choices of $k$-NN distances with a finite $k$,
independent of the sample size. Such a local and data dependent choice improves
performance in practice, but the bandwidth is vanishing at a fast rate, leading
to a non-vanishing bias. We show that the asymptotic bias of the proposed
estimator is universal; it is independent of the underlying distribution.
Hence, it can be pre-computed and subtracted from the estimate. As a byproduct,
we obtain a unified way of obtaining both kernel and NN estimators. The
corresponding theoretical contribution relating the asymptotic geometry of
nearest neighbors to order statistics is of independent mathematical interest.
|
Fitted Learning: Models with Awareness of their Limits | Though deep learning has pushed the boundaries of classification forward, in
recent years hints of the limits of standard classification have begun to
emerge. Problems such as fooling, adding new classes over time, and the need to
retrain learning models only for small changes to the original problem all
point to a potential shortcoming in the classic classification regime, where a
comprehensive a priori knowledge of the possible classes or concepts is
critical. Without such knowledge, classifiers misjudge the limits of their
knowledge and overgeneralization therefore becomes a serious obstacle to
consistent performance. In response to these challenges, this paper extends the
classic regime by reframing classification instead with the assumption that
concepts present in the training set are only a sample of the hypothetical
final set of concepts. To bring learning models into this new paradigm, a novel
elaboration of standard architectures called the competitive overcomplete
output layer (COOL) neural network is introduced. Experiments demonstrate the
effectiveness of COOL by applying it to fooling, separable concept learning,
one-class neural networks, and standard classification benchmarks. The results
suggest that, unlike conventional classifiers, the amount of generalization in
COOL networks can be tuned to match the problem.
|
Learning to learn with backpropagation of Hebbian plasticity | Hebbian plasticity is a powerful principle that allows biological brains to
learn from their lifetime experience. By contrast, artificial neural networks
trained with backpropagation generally have fixed connection weights that do
not change once training is complete. While recent methods can endow neural
networks with long-term memories, Hebbian plasticity is currently not amenable
to gradient descent. Here we derive analytical expressions for activity
gradients in neural networks with Hebbian plastic connections. Using these
expressions, we can use backpropagation to train not just the baseline weights
of the connections, but also their plasticity. As a result, the networks "learn
how to learn" in order to solve the problem at hand: the trained networks
automatically perform fast learning of unpredictable environmental features
during their lifetime, expanding the range of solvable problems. We test the
algorithm on various on-line learning tasks, including pattern completion,
one-shot learning, and reversal learning. The algorithm successfully learns how
to learn the relevant associations from one-shot instruction, and fine-tunes
the temporal dynamics of plasticity to allow for continual learning in response
to changing environmental parameters. We conclude that backpropagation of
Hebbian plasticity offers a powerful model for lifelong learning.
|
Improved Optimistic Mirror Descent for Sparsity and Curvature | Online Convex Optimization plays a key role in large scale machine learning.
Early approaches to this problem were conservative, in which the main focus was
protection against the worst case scenario. But recently several algorithms
have been developed for tightening the regret bounds in easy data instances
such as sparsity, predictable sequences, and curved losses. In this work we
unify some of these existing techniques to obtain new update rules for the
cases when these easy instances occur together. First we analyse an adaptive
and optimistic update rule which achieves tighter regret bound when the loss
sequence is sparse and predictable. Then we explain an update rule that
dynamically adapts to the curvature of the loss function and utilizes the
predictable nature of the loss sequence as well. Finally we extend these
results to composite losses.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.