title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Outperforming Good-Turing: Preliminary Report
|
Estimating a large alphabet probability distribution from a limited number of
samples is a fundamental problem in machine learning and statistics. A variety
of estimation schemes have been proposed over the years, mostly inspired by the
early work of Laplace and the seminal contribution of Good and Turing. One of
the basic assumptions shared by most commonly-used estimators is the unique
correspondence between the symbol's sample frequency and its estimated
probability. In this work we tackle this paradigmatic assumption; we claim that
symbols with "similar" frequencies shall be assigned the same estimated
probability value. This way we regulate the number of parameters and improve
generalization. In this preliminary report we show that by applying an ensemble
of such regulated estimators, we introduce a dramatic enhancement in the
estimation accuracy (typically up to 50%), compared to currently known methods.
An implementation of our suggested method is publicly available at the first
author's web-page.
|
Differentially Private Online Submodular Optimization
|
In this paper we develop the first algorithms for online submodular
minimization that preserve differential privacy under full information feedback
and bandit feedback. A sequence of $T$ submodular functions over a collection
of $n$ elements arrive online, and at each timestep the algorithm must choose a
subset of $[n]$ before seeing the function. The algorithm incurs a cost equal
to the function evaluated on the chosen set, and seeks to choose a sequence of
sets that achieves low expected regret.
Our first result is in the full information setting, where the algorithm can
observe the entire function after making its decision at each timestep. We give
an algorithm in this setting that is $\epsilon$-differentially private and
achieves expected regret
$\tilde{O}\left(\frac{n^{3/2}\sqrt{T}}{\epsilon}\right)$. This algorithm works
by relaxing submodular function to a convex function using the Lovasz
extension, and then simulating an algorithm for differentially private online
convex optimization.
Our second result is in the bandit setting, where the algorithm can only see
the cost incurred by its chosen set, and does not have access to the entire
function. This setting is significantly more challenging because the algorithm
does not receive enough information to compute the Lovasz extension or its
subgradients. Instead, we construct an unbiased estimate using a single-point
estimation, and then simulate private online convex optimization using this
estimate. Our algorithm using bandit feedback is $\epsilon$-differentially
private and achieves expected regret
$\tilde{O}\left(\frac{n^{3/2}T^{3/4}}{\epsilon}\right)$.
|
Combinatorial Bandits for Incentivizing Agents with Dynamic Preferences
|
The design of personalized incentives or recommendations to improve user
engagement is gaining prominence as digital platform providers continually
emerge. We propose a multi-armed bandit framework for matching incentives to
users, whose preferences are unknown a priori and evolving dynamically in time,
in a resource constrained environment. We design an algorithm that combines
ideas from three distinct domains: (i) a greedy matching paradigm, (ii) the
upper confidence bound algorithm (UCB) for bandits, and (iii) mixing times from
the theory of Markov chains. For this algorithm, we provide theoretical bounds
on the regret and demonstrate its performance via both synthetic and realistic
(matching supply and demand in a bike-sharing platform) examples.
|
A survey on policy search algorithms for learning robot controllers in a
handful of trials
|
Most policy search algorithms require thousands of training episodes to find
an effective policy, which is often infeasible with a physical robot. This
survey article focuses on the extreme other end of the spectrum: how can a
robot adapt with only a handful of trials (a dozen) and a few minutes? By
analogy with the word "big-data", we refer to this challenge as "micro-data
reinforcement learning". We show that a first strategy is to leverage prior
knowledge on the policy structure (e.g., dynamic movement primitives), on the
policy parameters (e.g., demonstrations), or on the dynamics (e.g.,
simulators). A second strategy is to create data-driven surrogate models of the
expected reward (e.g., Bayesian optimization) or the dynamical model (e.g.,
model-based policy search), so that the policy optimizer queries the model
instead of the real system. Overall, all successful micro-data algorithms
combine these two strategies by varying the kind of model and prior knowledge.
The current scientific challenges essentially revolve around scaling up to
complex robots (e.g., humanoids), designing generic priors, and optimizing the
computing time.
|
JUMPER: Learning When to Make Classification Decisions in Reading
|
In early years, text classification is typically accomplished by
feature-based machine learning models; recently, deep neural networks, as a
powerful learning machine, make it possible to work with raw input as the text
stands. However, exiting end-to-end neural networks lack explicit
interpretation of the prediction. In this paper, we propose a novel framework,
JUMPER, inspired by the cognitive process of text reading, that models text
classification as a sequential decision process. Basically, JUMPER is a neural
system that scans a piece of text sequentially and makes classification
decisions at the time it wishes. Both the classification result and when to
make the classification are part of the decision process, which is controlled
by a policy network and trained with reinforcement learning. Experimental
results show that a properly trained JUMPER has the following properties: (1)
It can make decisions whenever the evidence is enough, therefore reducing total
text reading by 30-40% and often finding the key rationale of prediction. (2)
It achieves classification accuracy better than or comparable to
state-of-the-art models in several benchmark and industrial datasets.
|
Memory Augmented Policy Optimization for Program Synthesis and Semantic
Parsing
|
We present Memory Augmented Policy Optimization (MAPO), a simple and novel
way to leverage a memory buffer of promising trajectories to reduce the
variance of policy gradient estimate. MAPO is applicable to deterministic
environments with discrete actions, such as structured prediction and
combinatorial optimization tasks. We express the expected return objective as a
weighted sum of two terms: an expectation over the high-reward trajectories
inside the memory buffer, and a separate expectation over trajectories outside
the buffer. To make an efficient algorithm of MAPO, we propose: (1) memory
weight clipping to accelerate and stabilize training; (2) systematic
exploration to discover high-reward trajectories; (3) distributed sampling from
inside and outside of the memory buffer to scale up training. MAPO improves the
sample efficiency and robustness of policy gradient, especially on tasks with
sparse rewards. We evaluate MAPO on weakly supervised program synthesis from
natural language (semantic parsing). On the WikiTableQuestions benchmark, we
improve the state-of-the-art by 2.6%, achieving an accuracy of 46.3%. On the
WikiSQL benchmark, MAPO achieves an accuracy of 74.9% with only weak
supervision, outperforming several strong baselines with full supervision. Our
source code is available at
https://github.com/crazydonkey200/neural-symbolic-machines
|
Sum-Product Networks for Sequence Labeling
|
We consider higher-order linear-chain conditional random fields (HO-LC-CRFs)
for sequence modelling, and use sum-product networks (SPNs) for representing
higher-order input- and output-dependent factors. SPNs are a recently
introduced class of deep models for which exact and efficient inference can be
performed. By combining HO-LC-CRFs with SPNs, expressive models over both the
output labels and the hidden variables are instantiated while still enabling
efficient exact inference. Furthermore, the use of higher-order factors allows
us to capture relations of multiple input segments and multiple output labels
as often present in real-world data. These relations can not be modelled by the
commonly used first-order models and higher-order models with local factors
including only a single output label. We demonstrate the effectiveness of our
proposed models for sequence labeling. In extensive experiments, we outperform
other state-of-the-art methods in optical character recognition and achieve
competitive results in phone classification.
|
Cause-Effect Deep Information Bottleneck For Systematically Missing
Covariates
|
Estimating the causal effects of an intervention from high-dimensional
observational data is difficult due to the presence of confounding. The task is
often complicated by the fact that we may have a systematic missingness in our
data at test time. Our approach uses the information bottleneck to perform a
low-dimensional compression of covariates by explicitly considering the
relevance of information. Based on the sufficiently reduced covariate, we
transfer the relevant information to cases where data is missing at test time,
allowing us to reliably and accurately estimate the effects of an intervention,
even where data is incomplete. Our results on causal inference benchmarks and a
real application for treating sepsis show that our method achieves state-of-the
art performance, without sacrificing interpretability.
|
A Variational Time Series Feature Extractor for Action Prediction
|
We propose a Variational Time Series Feature Extractor (VTSFE), inspired by
the VAE-DMP model of Chen et al., to be used for action recognition and
prediction. Our method is based on variational autoencoders. It improves
VAE-DMP in that it has a better noise inference model, a simpler transition
model constraining the acceleration in the trajectories of the latent space,
and a tighter lower bound for the variational inference. We apply the method
for classification and prediction of whole-body movements on a dataset with 7
tasks and 10 demonstrations per task, recorded with a wearable motion capture
suit. The comparison with VAE and VAE-DMP suggests the better performance of
our method for feature extraction. An open-source software implementation of
each method with TensorFlow is also provided. In addition, a more detailed
version of this work can be found in the indicated code repository. Although it
was meant to, the VTSFE hasn't been tested for action prediction, due to a lack
of time in the context of Maxime Chaveroche's Master thesis at INRIA.
|
Near Optimal Exploration-Exploitation in Non-Communicating Markov
Decision Processes
|
While designing the state space of an MDP, it is common to include states
that are transient or not reachable by any policy (e.g., in mountain car, the
product space of speed and position contains configurations that are not
physically reachable). This leads to defining weakly-communicating or
multi-chain MDPs. In this paper, we introduce \tucrl, the first algorithm able
to perform efficient exploration-exploitation in any finite Markov Decision
Process (MDP) without requiring any form of prior knowledge. In particular, for
any MDP with $S^{\texttt{C}}$ communicating states, $A$ actions and
$\Gamma^{\texttt{C}} \leq S^{\texttt{C}}$ possible communicating next states,
we derive a $\widetilde{O}(D^{\texttt{C}} \sqrt{\Gamma^{\texttt{C}}
S^{\texttt{C}} AT})$ regret bound, where $D^{\texttt{C}}$ is the diameter
(i.e., the longest shortest path) of the communicating part of the MDP. This is
in contrast with optimistic algorithms (e.g., UCRL, Optimistic PSRL) that
suffer linear regret in weakly-communicating MDPs, as well as posterior
sampling or regularised algorithms (e.g., REGAL), which require prior knowledge
on the bias span of the optimal policy to bias the exploration to achieve
sub-linear regret. We also prove that in weakly-communicating MDPs, no
algorithm can ever achieve a logarithmic growth of the regret without first
suffering a linear regret for a number of steps that is exponential in the
parameters of the MDP. Finally, we report numerical simulations supporting our
theoretical findings and showing how TUCRL overcomes the limitations of the
state-of-the-art.
|
A Structured Prediction Approach for Label Ranking
|
We propose to solve a label ranking problem as a structured output regression
task. We adopt a least square surrogate loss approach that solves a supervised
learning problem in two steps: the regression step in a well-chosen feature
space and the pre-image step. We use specific feature maps/embeddings for
ranking data, which convert any ranking/permutation into a vector
representation. These embeddings are all well-tailored for our approach, either
by resulting in consistent estimators, or by solving trivially the pre-image
problem which is often the bottleneck in structured prediction. We also propose
their natural extension to the case of partial rankings and prove their
efficiency on real-world datasets.
|
Extracting Actionable Knowledge from Domestic Violence Discourses on
Social Media
|
Domestic Violence (DV) is considered as big social issue and there exists a
strong relationship between DV and health impacts of the public. Existing
research studies have focused on social media to track and analyse real world
events like emerging trends, natural disasters, user sentiment analysis,
political opinions, and health care. However there is less attention given on
social welfare issues like DV and its impact on public health. Recently, the
victims of DV turned to social media platforms to express their feelings in the
form of posts and seek the social and emotional support, for sympathetic
encouragement, to show compassion and empathy among public. But, it is
difficult to mine the actionable knowledge from large conversational datasets
from social media due to the characteristics of high dimensions, short, noisy,
huge volume, high velocity, and so on. Hence, this paper will propose a novel
framework to model and discover the various themes related to DV from the
public domain. The proposed framework would possibly provide unprecedentedly
valuable information to the public health researchers, national family health
organizations, government and public with data enrichment and consolidation to
improve the social welfare of the community. Thus provides actionable knowledge
by monitoring and analysing continuous and rich user generated content.
|
Learning a Representation Map for Robot Navigation using Deep
Variational Autoencoder
|
The aim of this work is to use Variational Autoencoder (VAE) to learn a
representation of an indoor environment that can be used for robot navigation.
We use images extracted from a video, in which a camera takes a tour around a
house, for training the VAE model with a 4 dimensional latent space. After the
model is trained, each real frame has a corresponding representation point on
manifold in the latent space, and each representation point has corresponding
reconstructed image. For the navigation problem, we map the starting image and
destination image to the latent space, then optimize a path on the learned
manifold connecting the two points, and finally map the path back through
decoder to a sequence of images. The ideal sequence of images should correspond
to a route that is spatially continuous - i.e. neighbor images in the route
should correspond to neighbor locations in physical space. Such a route could
be used for navigation with computer vision techniques, i.e. a robot could
follow the image sequence from starting location to destination in the
environment step by step. We implement this algorithm, but find in our
experimental results that the resulting route is not satisfactory. The route
consist of several discontinuous image frames along the ideal routes, so that
the route could not be followed by a robot with computer vision techniques in
practice. In our evaluation, we propose two reasons for our failure to
automatically find continuous routes: (1) The VAE tends to capture global
structures, but discard the details; (2) the Euclidean similarity metric used
for measuring continuity between house images is sub-optimal. For further work,
we propose: trying other generative models like VAE-GANs which may be better at
reconstructing the details to learn the representation map, and adjusting the
similarity metric in the path selecting algorithm.
|
Multi-Task Learning with Incomplete Data for Healthcare
|
Multi-task learning is a type of transfer learning that trains multiple tasks
simultaneously and leverages the shared information between related tasks to
improve the generalization performance. However, missing features in the input
matrix is a much more difficult problem which needs to be carefully addressed.
Removing records with missing values can significantly reduce the sample size,
which is impractical for datasets with large percentage of missing values.
Popular imputation methods often distort the covariance structure of the data,
which causes inaccurate inference. In this paper we propose using plug-in
covariance matrix estimators to tackle the challenge of missing features.
Specifically, we analyze the plug-in estimators under the framework of robust
multi-task learning with LASSO and graph regularization, which captures the
relatedness between tasks via graph regularization. We use the Alzheimer's
disease progression dataset as an example to show how the proposed framework is
effective for prediction and model estimation when missing data is present.
|
A Review of Different Word Embeddings for Sentiment Classification using
Deep Learning
|
The web is loaded with textual content, and Natural Language Processing is a
standout amongst the most vital fields in Machine Learning. But when data is
huge simple Machine Learning algorithms are not able to handle it and that is
when Deep Learning comes into play which based on Neural Networks. However
since neural networks cannot process raw text, we have to change over them
through some diverse strategies of word embedding. This paper demonstrates
those distinctive word embedding strategies implemented on an Amazon Review
Dataset, which has two sentiments to be classified: Happy and Unhappy based on
numerous customer reviews. Moreover we demonstrate the distinction in accuracy
with a discourse about which word embedding to apply when.
|
Deep Multiple Instance Feature Learning via Variational Autoencoder
|
We describe a novel weakly supervised deep learning framework that combines
both the discriminative and generative models to learn meaningful
representation in the multiple instance learning (MIL) setting. MIL is a weakly
supervised learning problem where labels are associated with groups of
instances (referred as bags) instead of individual instances. To address the
essential challenge in MIL problems raised from the uncertainty of positive
instances label, we use a discriminative model regularized by variational
autoencoders (VAEs) to maximize the differences between latent representations
of all instances and negative instances. As a result, the hidden layer of the
variational autoencoder learns meaningful representation. This representation
can effectively be used for MIL problems as illustrated by better performance
on the standard benchmark datasets comparing to the state-of-the-art
approaches. More importantly, unlike most related studies, the proposed
framework can be easily scaled to large dataset problems, as illustrated by the
audio event detection and segmentation task. Visualization also confirms the
effectiveness of the latent representation in discriminating positive and
negative classes.
|
Blockchain as a Service: A Decentralized and Secure Computing Paradigm
|
Thanks to the advances in machine learning, data-driven analysis tools have
become valuable solutions for various applications. However, there still remain
essential challenges to develop effective data-driven methods because of the
need to acquire a large amount of data and to have sufficient computing power
to handle the data. In many instances these challenges are addressed by relying
on a dominant cloud computing vendor, but, although commercial cloud vendors
provide valuable platforms for data analytics, they can suffer from a lack of
transparency, security, and privacy-perservation. Furthermore, reliance on
cloud servers prevents applying big data analytics in environments where the
computing power is scattered. To address these challenges, a decentralize,
secure, and privacy-preserving computing paradigm is proposed to enable an
asynchronized cooperative computing process amongst scattered and untrustworthy
computing nodes that may have limited computing power and computing
intelligence. This paradigm is designed by exploring blockchain, decentralized
learning, homomorphic encryption, and software defined networking(SDN)
techniques. The performance of the proposed paradigm is evaluated via different
scenarios in the simulation section.
|
Fully Scalable Gaussian Processes using Subspace Inducing Inputs
|
We introduce fully scalable Gaussian processes, an implementation scheme that
tackles the problem of treating a high number of training instances together
with high dimensional input data. Our key idea is a representation trick over
the inducing variables called subspace inducing inputs. This is combined with
certain matrix-preconditioning based parametrizations of the variational
distributions that lead to simplified and numerically stable variational lower
bounds. Our illustrative applications are based on challenging extreme
multi-label classification problems with the extra burden of the very large
number of class labels. We demonstrate the usefulness of our approach by
presenting predictive performances together with low computational times in
datasets with extremely large number of instances and input dimensions.
|
3D Steerable CNNs: Learning Rotationally Equivariant Features in
Volumetric Data
|
We present a convolutional network that is equivariant to rigid body motions.
The model uses scalar-, vector-, and tensor fields over 3D Euclidean space to
represent data, and equivariant convolutions to map between such
representations. These SE(3)-equivariant convolutions utilize kernels which are
parameterized as a linear combination of a complete steerable kernel basis,
which is derived analytically in this paper. We prove that equivariant
convolutions are the most general equivariant linear maps between fields over
R^3. Our experimental results confirm the effectiveness of 3D Steerable CNNs
for the problem of amino acid propensity prediction and protein structure
classification, both of which have inherent SE(3) symmetry.
|
M-ADDA: Unsupervised Domain Adaptation with Deep Metric Learning
|
Unsupervised domain adaptation techniques have been successful for a wide
range of problems where supervised labels are limited. The task is to classify
an unlabeled `target' dataset by leveraging a labeled `source' dataset that
comes from a slightly similar distribution. We propose metric-based adversarial
discriminative domain adaptation (M-ADDA) which performs two main steps. First,
it uses a metric learning approach to train the source model on the source
dataset by optimizing the triplet loss function. This results in clusters where
embeddings of the same label are close to each other and those with different
labels are far from one another. Next, it uses the adversarial approach (as
that used in ADDA \cite{2017arXiv170205464T}) to make the extracted features
from the source and target datasets indistinguishable. Simultaneously, we
optimize a novel loss function that encourages the target dataset's embeddings
to form clusters. While ADDA and M-ADDA use similar architectures, we show that
M-ADDA performs significantly better on the digits adaptation datasets of MNIST
and USPS. This suggests that using metric-learning for domain adaptation can
lead to large improvements in classification accuracy for the domain adaptation
task. The code is available at \url{https://github.com/IssamLaradji/M-ADDA}.
|
Deep Learning for Launching and Mitigating Wireless Jamming Attacks
|
An adversarial machine learning approach is introduced to launch jamming
attacks on wireless communications and a defense strategy is presented. A
cognitive transmitter uses a pre-trained classifier to predict the current
channel status based on recent sensing results and decides whether to transmit
or not, whereas a jammer collects channel status and ACKs to build a deep
learning classifier that reliably predicts the next successful transmissions
and effectively jams them. This jamming approach is shown to reduce the
transmitter's performance much more severely compared with random or
sensing-based jamming. The deep learning classification scores are used by the
jammer for power control subject to an average power constraint. Next, a
generative adversarial network (GAN) is developed for the jammer to reduce the
time to collect the training dataset by augmenting it with synthetic samples.
As a defense scheme, the transmitter deliberately takes a small number of wrong
actions in spectrum access (in form of a causative attack against the jammer)
and therefore prevents the jammer from building a reliable classifier. The
transmitter systematically selects when to take wrong actions and adapts the
level of defense to mislead the jammer into making prediction errors and
consequently increase its throughput.
|
The Goldilocks zone: Towards better understanding of neural network loss
landscapes
|
We explore the loss landscape of fully-connected and convolutional neural
networks using random, low-dimensional hyperplanes and hyperspheres. Evaluating
the Hessian, $H$, of the loss function on these hypersurfaces, we observe 1) an
unusual excess of the number of positive eigenvalues of $H$, and 2) a large
value of $\mathrm{Tr}(H) / ||H||$ at a well defined range of configuration
space radii, corresponding to a thick, hollow, spherical shell we refer to as
the \textit{Goldilocks zone}. We observe this effect for fully-connected neural
networks over a range of network widths and depths on MNIST and CIFAR-10
datasets with the $\mathrm{ReLU}$ and $\tanh$ non-linearities, and a similar
effect for convolutional networks. Using our observations, we demonstrate a
close connection between the Goldilocks zone, measures of local
convexity/prevalence of positive curvature, and the suitability of a network
initialization. We show that the high and stable accuracy reached when
optimizing on random, low-dimensional hypersurfaces is directly related to the
overlap between the hypersurface and the Goldilocks zone, and as a corollary
demonstrate that the notion of intrinsic dimension is initialization-dependent.
We note that common initialization techniques initialize neural networks in
this particular region of unusually high convexity/prevalence of positive
curvature, and offer a geometric intuition for their success. Furthermore, we
demonstrate that initializing a neural network at a number of points and
selecting for high measures of local convexity such as $\mathrm{Tr}(H) /
||H||$, number of positive eigenvalues of $H$, or low initial loss, leads to
statistically significantly faster training on MNIST. Based on our
observations, we hypothesize that the Goldilocks zone contains an unusually
high density of suitable initialization configurations.
|
Gaussian Processes and Kernel Methods: A Review on Connections and
Equivalences
|
This paper is an attempt to bridge the conceptual gaps between researchers
working on the two widely used approaches based on positive definite kernels:
Bayesian learning or inference using Gaussian processes on the one side, and
frequentist kernel methods based on reproducing kernel Hilbert spaces on the
other. It is widely known in machine learning that these two formalisms are
closely related; for instance, the estimator of kernel ridge regression is
identical to the posterior mean of Gaussian process regression. However, they
have been studied and developed almost independently by two essentially
separate communities, and this makes it difficult to seamlessly transfer
results between them. Our aim is to overcome this potential difficulty. To this
end, we review several old and new results and concepts from either side, and
juxtapose algorithmic quantities from each framework to highlight close
similarities. We also provide discussions on subtle philosophical and
theoretical differences between the two approaches.
|
Generative Probabilistic Novelty Detection with Adversarial Autoencoders
|
Novelty detection is the problem of identifying whether a new data point is
considered to be an inlier or an outlier. We assume that training data is
available to describe only the inlier distribution. Recent approaches primarily
leverage deep encoder-decoder network architectures to compute a reconstruction
error that is used to either compute a novelty score or to train a one-class
classifier. While we too leverage a novel network of that kind, we take a
probabilistic approach and effectively compute how likely is that a sample was
generated by the inlier distribution. We achieve this with two main
contributions. First, we make the computation of the novelty probability
feasible because we linearize the parameterized manifold capturing the
underlying structure of the inlier distribution, and show how the probability
factorizes and can be computed with respect to local coordinates of the
manifold tangent space. Second, we improved the training of the autoencoder
network. An extensive set of results show that the approach achieves
state-of-the-art results on several benchmark datasets.
|
From Text to Topics in Healthcare Records: An Unsupervised Graph
Partitioning Methodology
|
Electronic Healthcare Records contain large volumes of unstructured data,
including extensive free text. Yet this source of detailed information often
remains under-used because of a lack of methodologies to extract interpretable
content in a timely manner. Here we apply network-theoretical tools to analyse
free text in Hospital Patient Incident reports from the National Health
Service, to find clusters of documents with similar content in an unsupervised
manner at different levels of resolution. We combine deep neural network
paragraph vector text-embedding with multiscale Markov Stability community
detection applied to a sparsified similarity graph of document vectors, and
showcase the approach on incident reports from Imperial College Healthcare NHS
Trust, London. The multiscale community structure reveals different levels of
meaning in the topics of the dataset, as shown by descriptive terms extracted
from the clusters of records. We also compare a posteriori against hand-coded
categories assigned by healthcare personnel, and show that our approach
outperforms LDA-based models. Our content clusters exhibit good correspondence
with two levels of hand-coded categories, yet they also provide further medical
detail in certain areas and reveal complementary descriptors of incidents
beyond the external classification taxonomy.
|
Synthetic Sampling for Multi-Class Malignancy Prediction
|
We explore several oversampling techniques for an imbalanced multi-label
classification problem, a setting often encountered when developing models for
Computer-Aided Diagnosis (CADx) systems. While most CADx systems aim to
optimize classifiers for overall accuracy without considering the relative
distribution of each class, we look into using synthetic sampling to increase
per-class performance when predicting the degree of malignancy. Using low-level
image features and a random forest classifier, we show that using synthetic
oversampling techniques increases the sensitivity of the minority classes by an
average of 7.22% points, with as much as a 19.88% point increase in sensitivity
for a particular minority class. Furthermore, the analysis of low-level image
feature distributions for the synthetic nodules reveals that these nodules can
provide insights on how to preprocess image data for better classification
performance or how to supplement the original datasets when more data
acquisition is feasible.
|
Anytime Neural Prediction via Slicing Networks Vertically
|
The pioneer deep neural networks (DNNs) have emerged to be deeper or wider
for improving their accuracy in various applications of artificial
intelligence. However, DNNs are often too heavy to deploy in practice, and it
is often required to control their architectures dynamically given computing
resource budget, i.e., anytime prediction. While most existing approaches have
focused on training multiple shallow sub-networks jointly, we study training
thin sub-networks instead. To this end, we first build many inclusive thin
sub-networks (of the same depth) under a minor modification of existing
multi-branch DNNs, and found that they can significantly outperform the
state-of-art dense architecture for anytime prediction. This is remarkable due
to their simplicity and effectiveness, but training many thin sub-networks
jointly faces a new challenge on training complexity. To address the issue, we
also propose a novel DNN architecture by forcing a certain sparsity pattern on
multi-branch network parameters, making them train efficiently for the purpose
of anytime prediction. In our experiments on the ImageNet dataset, its
sub-networks have up to $43.3\%$ smaller sizes (FLOPs) compared to those of the
state-of-art anytime model with respect to the same accuracy. Finally, we also
propose an alternative task under the proposed architecture using a
hierarchical taxonomy, which brings a new angle for anytime prediction.
|
Gradient Hyperalignment for multi-subject fMRI data alignment
|
Multi-subject fMRI data analysis is an interesting and challenging problem in
human brain decoding studies. The inherent anatomical and functional
variability across subjects make it necessary to do both anatomical and
functional alignment before classification analysis. Besides, when it comes to
big data, time complexity becomes a problem that cannot be ignored. This paper
proposes Gradient Hyperalignment (Gradient-HA) as a gradient-based functional
alignment method that is suitable for multi-subject fMRI datasets with large
amounts of samples and voxels. The advantage of Gradient-HA is that it can
solve independence and high dimension problems by using Independent Component
Analysis (ICA) and Stochastic Gradient Ascent (SGA). Validation using
multi-classification tasks on big data demonstrates that Gradient-HA method has
less time complexity and better or comparable performance compared with other
state-of-the-art functional alignment methods.
|
Predicting Infant Motor Development Status using Day Long Movement Data
from Wearable Sensors
|
Infants with a variety of complications at or before birth are classified as
being at risk for developmental delays (AR). As they grow older, they are
followed by healthcare providers in an effort to discern whether they are on a
typical or impaired developmental trajectory. Often, it is difficult to make an
accurate determination early in infancy as infants with typical development
(TD) display high variability in their developmental trajectories both in
content and timing. Studies have shown that spontaneous movements have the
potential to differentiate typical and atypical trajectories early in life
using sensors and kinematic analysis systems. In this study, machine learning
classification algorithms are used to take inertial movement from wearable
sensors placed on an infant for a day and predict if the infant is AR or TD,
thus further establishing the connection between early spontaneous movement and
developmental trajectory.
|
Optimistic mirror descent in saddle-point problems: Going the extra
(gradient) mile
|
Owing to their connection with generative adversarial networks (GANs),
saddle-point problems have recently attracted considerable interest in machine
learning and beyond. By necessity, most theoretical guarantees revolve around
convex-concave (or even linear) problems; however, making theoretical inroads
towards efficient GAN training depends crucially on moving beyond this classic
framework. To make piecemeal progress along these lines, we analyze the
behavior of mirror descent (MD) in a class of non-monotone problems whose
solutions coincide with those of a naturally associated variational inequality
- a property which we call coherence. We first show that ordinary, "vanilla" MD
converges under a strict version of this condition, but not otherwise; in
particular, it may fail to converge even in bilinear models with a unique
solution. We then show that this deficiency is mitigated by optimism: by taking
an "extra-gradient" step, optimistic mirror descent (OMD) converges in all
coherent problems. Our analysis generalizes and extends the results of
Daskalakis et al. (2018) for optimistic gradient descent (OGD) in bilinear
problems, and makes concrete headway for establishing convergence beyond
convex-concave games. We also provide stochastic analogues of these results,
and we validate our analysis by numerical experiments in a wide array of GAN
models (including Gaussian mixture models, as well as the CelebA and CIFAR-10
datasets).
|
When Work Matters: Transforming Classical Network Structures to Graph
CNN
|
Numerous pattern recognition applications can be formed as learning from
graph-structured data, including social network, protein-interaction network,
the world wide web data, knowledge graph, etc. While convolutional neural
network (CNN) facilitates great advances in gridded image/video understanding
tasks, very limited attention has been devoted to transform these successful
network structures (including Inception net, Residual net, Dense net, etc.) to
establish convolutional networks on graph, due to its irregularity and
complexity geometric topologies (unordered vertices, unfixed number of adjacent
edges/vertices). In this paper, we aim to give a comprehensive analysis of when
work matters by transforming different classical network structures to graph
CNN, particularly in the basic graph recognition problem. Specifically, we
firstly review the general graph CNN methods, especially in its spectral
filtering operation on the irregular graph data. We then introduce the basic
structures of ResNet, Inception and DenseNet into graph CNN and construct these
network structures on graph, named as G_ResNet, G_Inception, G_DenseNet. In
particular, it seeks to help graph CNNs by shedding light on how these
classical network structures work and providing guidelines for choosing
appropriate graph network frameworks. Finally, we comprehensively evaluate the
performance of these different network structures on several public graph
datasets (including social networks and bioinformatic datasets), and
demonstrate how different network structures work on graph CNN in the graph
recognition task.
|
Robust and Scalable Differentiable Neural Computer for Question
Answering
|
Deep learning models are often not easily adaptable to new tasks and require
task-specific adjustments. The differentiable neural computer (DNC), a
memory-augmented neural network, is designed as a general problem solver which
can be used in a wide range of tasks. But in reality, it is hard to apply this
model to new tasks. We analyze the DNC and identify possible improvements
within the application of question answering. This motivates a more robust and
scalable DNC (rsDNC). The objective precondition is to keep the general
character of this model intact while making its application more reliable and
speeding up its required training time. The rsDNC is distinguished by a more
robust training, a slim memory unit and a bidirectional architecture. We not
only achieve new state-of-the-art performance on the bAbI task, but also
minimize the performance variance between different initializations.
Furthermore, we demonstrate the simplified applicability of the rsDNC to new
tasks with passable results on the CNN RC task without adaptions.
|
A Supervised Geometry-Aware Mapping Approach for Classification of
Hyperspectral Images
|
The lack of proper class discrimination among the Hyperspectral (HS) data
points poses a potential challenge in HS classification. To address this issue,
this paper proposes an optimal geometry-aware transformation for enhancing the
classification accuracy. The underlying idea of this method is to obtain a
linear projection matrix by solving a nonlinear objective function based on the
intrinsic geometrical structure of the data. The objective function is
constructed to quantify the discrimination between the points from dissimilar
classes on the projected data space. Then the obtained projection matrix is
used to linearly map the data to more discriminative space. The effectiveness
of the proposed transformation is illustrated with three benchmark real-world
HS data sets. The experiments reveal that the classification and dimensionality
reduction methods on the projected discriminative space outperform their
counterpart in the original space.
|
VFPred: A Fusion of Signal Processing and Machine Learning techniques in
Detecting Ventricular Fibrillation from ECG Signals
|
Ventricular Fibrillation (VF), one of the most dangerous arrhythmias, is
responsible for sudden cardiac arrests. Thus, various algorithms have been
developed to predict VF from Electrocardiogram (ECG), which is a binary
classification problem. In the literature, we find a number of algorithms based
on signal processing, where, after some robust mathematical operations the
decision is given based on a predefined threshold over a single value. On the
other hand, some machine learning based algorithms are also reported in the
literature; however, these algorithms merely combine some parameters and make a
prediction using those as features. Both the approaches have their perks and
pitfalls; thus our motivation was to coalesce them to get the best out of the
both worlds. Hence we have developed, VFPred that, in addition to employing a
signal processing pipeline, namely, Empirical Mode Decomposition and Discrete
Time Fourier Transform for useful feature extraction, uses a Support Vector
Machine for efficient classification. VFPred turns out to be a robust algorithm
as it is able to successfully segregate the two classes with equal confidence
(Sensitivity = 99.99%, Specificity = 98.40%) even from a short signal of 5
seconds long, whereas existing works though requires longer signals, flourishes
in one but fails in the other.
|
Approximate Leave-One-Out for Fast Parameter Tuning in High Dimensions
|
Consider the following class of learning schemes: $$\hat{\boldsymbol{\beta}}
:= \arg\min_{\boldsymbol{\beta}}\;\sum_{j=1}^n
\ell(\boldsymbol{x}_j^\top\boldsymbol{\beta}; y_j) + \lambda
R(\boldsymbol{\beta}),\qquad\qquad (1) $$ where $\boldsymbol{x}_i \in
\mathbb{R}^p$ and $y_i \in \mathbb{R}$ denote the $i^{\text{th}}$ feature and
response variable respectively. Let $\ell$ and $R$ be the loss function and
regularizer, $\boldsymbol{\beta}$ denote the unknown weights, and $\lambda$ be
a regularization parameter. Finding the optimal choice of $\lambda$ is a
challenging problem in high-dimensional regimes where both $n$ and $p$ are
large. We propose two frameworks to obtain a computationally efficient
approximation ALO of the leave-one-out cross validation (LOOCV) risk for
nonsmooth losses and regularizers. Our two frameworks are based on the primal
and dual formulations of (1). We prove the equivalence of the two approaches
under smoothness conditions. This equivalence enables us to justify the
accuracy of both methods under such conditions. We use our approaches to obtain
a risk estimate for several standard problems, including generalized LASSO,
nuclear norm regularization, and support vector machines. We empirically
demonstrate the effectiveness of our results for non-differentiable cases.
|
DeepSource: Point Source Detection using Deep Learning
|
Point source detection at low signal-to-noise is challenging for astronomical
surveys, particularly in radio interferometry images where the noise is
correlated. Machine learning is a promising solution, allowing the development
of algorithms tailored to specific telescope arrays and science cases. We
present DeepSource - a deep learning solution - that uses convolutional neural
networks to achieve these goals. DeepSource enhances the Signal-to-Noise Ratio
(SNR) of the original map and then uses dynamic blob detection to detect
sources. Trained and tested on two sets of 500 simulated 1 deg x 1 deg MeerKAT
images with a total of 300,000 sources, DeepSource is essentially perfect in
both purity and completeness down to SNR = 4 and outperforms PyBDSF in all
metrics. For uniformly-weighted images it achieves a Purity x Completeness (PC)
score at SNR = 3 of 0.73, compared to 0.31 for the best PyBDSF model. For
natural-weighting we find a smaller improvement of ~40% in the PC score at SNR
= 3. If instead we ask where either of the purity or completeness first drop to
90%, we find that DeepSource reaches this value at SNR = 3.6 compared to the
4.3 of PyBDSF (natural-weighting). A key advantage of DeepSource is that it can
learn to optimally trade off purity and completeness for any science case under
consideration. Our results show that deep learning is a promising approach to
point source detection in astronomical images.
|
Improving DNN-based Music Source Separation using Phase Features
|
Music source separation with deep neural networks typically relies only on
amplitude features. In this paper we show that additional phase features can
improve the separation performance. Using the theoretical relationship between
STFT phase and amplitude, we conjecture that derivatives of the phase are a
good feature representation opposed to the raw phase. We verify this conjecture
experimentally and propose a new DNN architecture which combines amplitude and
phase. This joint approach achieves a better signal-to distortion ratio on the
DSD100 dataset for all instruments compared to a network that uses only
amplitude features. Especially, the bass instrument benefits from the phase
information.
|
A Deep-Learning-Based Geological Parameterization for History Matching
Complex Models
|
A new low-dimensional parameterization based on principal component analysis
(PCA) and convolutional neural networks (CNN) is developed to represent complex
geological models. The CNN-PCA method is inspired by recent developments in
computer vision using deep learning. CNN-PCA can be viewed as a generalization
of an existing optimization-based PCA (O-PCA) method. Both CNN-PCA and O-PCA
entail post-processing a PCA model to better honor complex geological features.
In CNN-PCA, rather than use a histogram-based regularization as in O-PCA, a new
regularization involving a set of metrics for multipoint statistics is
introduced. The metrics are based on summary statistics of the nonlinear filter
responses of geological models to a pre-trained deep CNN. In addition, in the
CNN-PCA formulation presented here, a convolutional neural network is trained
as an explicit transform function that can post-process PCA models quickly.
CNN-PCA is shown to provide both unconditional and conditional realizations
that honor the geological features present in reference SGeMS geostatistical
realizations for a binary channelized system. Flow statistics obtained through
simulation of random CNN-PCA models closely match results for random SGeMS
models for a demanding case in which O-PCA models lead to significant
discrepancies. Results for history matching are also presented. In this
assessment CNN-PCA is applied with derivative-free optimization, and a subspace
randomized maximum likelihood method is used to provide multiple posterior
models. Data assimilation and significant uncertainty reduction are achieved
for existing wells, and physically reasonable predictions are also obtained for
new wells. Finally, the CNN-PCA method is extended to a more complex
non-stationary bimodal deltaic fan system, and is shown to provide high-quality
realizations for this challenging example.
|
Data-driven Upsampling of Point Clouds
|
High quality upsampling of sparse 3D point clouds is critically useful for a
wide range of geometric operations such as reconstruction, rendering, meshing,
and analysis. In this paper, we propose a data-driven algorithm that enables an
upsampling of 3D point clouds without the need for hard-coded rules. Our
approach uses a deep network with Chamfer distance as the loss function,
capable of learning the latent features in point clouds belonging to different
object categories. We evaluate our algorithm across different amplification
factors, with upsampling learned and performed on objects belonging to the same
category as well as different categories. We also explore the desirable
characteristics of input point clouds as a function of the distribution of the
point samples. Finally, we demonstrate the performance of our algorithm in
single-category training versus multi-category training scenarios. The final
proposed model is compared against a baseline, optimization-based upsampling
method. Results indicate that our algorithm is capable of generating more
uniform and accurate upsamplings.
|
Financial Trading as a Game: A Deep Reinforcement Learning Approach
|
An automatic program that generates constant profit from the financial market
is lucrative for every market practitioner. Recent advance in deep
reinforcement learning provides a framework toward end-to-end training of such
trading agent. In this paper, we propose an Markov Decision Process (MDP) model
suitable for the financial trading task and solve it with the state-of-the-art
deep recurrent Q-network (DRQN) algorithm. We propose several modifications to
the existing learning algorithm to make it more suitable under the financial
trading setting, namely 1. We employ a substantially small replay memory (only
a few hundreds in size) compared to ones used in modern deep reinforcement
learning algorithms (often millions in size.) 2. We develop an action
augmentation technique to mitigate the need for random exploration by providing
extra feedback signals for all actions to the agent. This enables us to use
greedy policy over the course of learning and shows strong empirical
performance compared to more commonly used epsilon-greedy exploration. However,
this technique is specific to financial trading under a few market assumptions.
3. We sample a longer sequence for recurrent neural network training. A side
product of this mechanism is that we can now train the agent for every T steps.
This greatly reduces training time since the overall computation is down by a
factor of T. We combine all of the above into a complete online learning
algorithm and validate our approach on the spot foreign exchange market.
|
BALSON: Bayesian Least Squares Optimization with Nonnegative L1-Norm
Constraint
|
A Bayesian approach termed BAyesian Least Squares Optimization with
Nonnegative L1-norm constraint (BALSON) is proposed. The error distribution of
data fitting is described by Gaussian likelihood. The parameter distribution is
assumed to be a Dirichlet distribution. With the Bayes rule, searching for the
optimal parameters is equivalent to finding the mode of the posterior
distribution. In order to explicitly characterize the nonnegative L1-norm
constraint of the parameters, we further approximate the true posterior
distribution by a Dirichlet distribution. We estimate the statistics of the
approximating Dirichlet posterior distribution by sampling methods. Four
sampling methods have been introduced. With the estimated posterior
distributions, the original parameters can be effectively reconstructed in
polynomial fitting problems, and the BALSON framework is found to perform
better than conventional methods.
|
Distillation Techniques for Pseudo-rehearsal Based Incremental Learning
|
The ability to learn from incrementally arriving data is essential for any
life-long learning system. However, standard deep neural networks forget the
knowledge about the old tasks, a phenomenon called catastrophic forgetting,
when trained on incrementally arriving data. We discuss the biases in current
Generative Adversarial Networks (GAN) based approaches that learn the
classifier by knowledge distillation from previously trained classifiers. These
biases cause the trained classifier to perform poorly. We propose an approach
to remove these biases by distilling knowledge from the classifier of AC-GAN.
Experiments on MNIST and CIFAR10 show that this method is comparable to current
state of the art rehearsal based approaches. The code for this paper is
available at https://bit.ly/incremental-learning
|
Revisiting Distillation and Incremental Classifier Learning
|
One of the key differences between the learning mechanism of humans and
Artificial Neural Networks (ANNs) is the ability of humans to learn one task at
a time. ANNs, on the other hand, can only learn multiple tasks simultaneously.
Any attempts at learning new tasks incrementally cause them to completely
forget about previous tasks. This lack of ability to learn incrementally,
called Catastrophic Forgetting, is considered a major hurdle in building a true
AI system. In this paper, our goal is to isolate the truly effective existing
ideas for incremental learning from those that only work under certain
conditions. To this end, we first thoroughly analyze the current state of the
art (iCaRL) method for incremental learning and demonstrate that the good
performance of the system is not because of the reasons presented in the
existing literature. We conclude that the success of iCaRL is primarily due to
knowledge distillation and recognize a key limitation of knowledge
distillation, i.e, it often leads to bias in classifiers. Finally, we propose a
dynamic threshold moving algorithm that is able to successfully remove this
bias. We demonstrate the effectiveness of our algorithm on CIFAR100 and MNIST
datasets showing near-optimal results. Our implementation is available at
https://github.com/Khurramjaved96/incremental-learning.
|
A Tutorial on Bayesian Optimization
|
Bayesian optimization is an approach to optimizing objective functions that
take a long time (minutes or hours) to evaluate. It is best-suited for
optimization over continuous domains of less than 20 dimensions, and tolerates
stochastic noise in function evaluations. It builds a surrogate for the
objective and quantifies the uncertainty in that surrogate using a Bayesian
machine learning technique, Gaussian process regression, and then uses an
acquisition function defined from this surrogate to decide where to sample. In
this tutorial, we describe how Bayesian optimization works, including Gaussian
process regression and three common acquisition functions: expected
improvement, entropy search, and knowledge gradient. We then discuss more
advanced techniques, including running multiple function evaluations in
parallel, multi-fidelity and multi-information source optimization,
expensive-to-evaluate constraints, random environmental conditions, multi-task
Bayesian optimization, and the inclusion of derivative information. We conclude
with a discussion of Bayesian optimization software and future research
directions in the field. Within our tutorial material we provide a
generalization of expected improvement to noisy evaluations, beyond the
noise-free setting where it is more commonly applied. This generalization is
justified by a formal decision-theoretic argument, standing in contrast to
previous ad hoc modifications.
|
Improving Deep Learning through Automatic Programming
|
Deep learning and deep architectures are emerging as the best machine
learning methods so far in many practical applications such as reducing the
dimensionality of data, image classification, speech recognition or object
segmentation. In fact, many leading technology companies such as Google,
Microsoft or IBM are researching and using deep architectures in their systems
to replace other traditional models. Therefore, improving the performance of
these models could make a strong impact in the area of machine learning.
However, deep learning is a very fast-growing research domain with many core
methodologies and paradigms just discovered over the last few years. This
thesis will first serve as a short summary of deep learning, which tries to
include all of the most important ideas in this research area. Based on this
knowledge, we suggested, and conducted some experiments to investigate the
possibility of improving the deep learning based on automatic programming
(ADATE). Although our experiments did produce good results, there are still
many more possibilities that we could not try due to limited time as well as
some limitations of the current ADATE version. I hope that this thesis can
promote future work on this topic, especially when the next version of ADATE
comes out. This thesis also includes a short analysis of the power of ADATE
system, which could be useful for other researchers who want to know what it is
capable of.
|
Hierarchical stochastic graphlet embedding for graph-based pattern
recognition
|
Despite being very successful within the pattern recognition and machine
learning community, graph-based methods are often unusable because of the lack
of mathematical operations defined in graph domain. Graph embedding, which maps
graphs to a vectorial space, has been proposed as a way to tackle these
difficulties enabling the use of standard machine learning techniques. However,
it is well known that graph embedding functions usually suffer from the loss of
structural information. In this paper, we consider the hierarchical structure
of a graph as a way to mitigate this loss of information. The hierarchical
structure is constructed by topologically clustering the graph nodes, and
considering each cluster as a node in the upper hierarchical level. Once this
hierarchical structure is constructed, we consider several configurations to
define the mapping into a vector space given a classical graph embedding, in
particular, we propose to make use of the Stochastic Graphlet Embedding (SGE).
Broadly speaking, SGE produces a distribution of uniformly sampled low to high
order graphlets as a way to embed graphs into the vector space. In what
follows, the coarse-to-fine structure of a graph hierarchy and the statistics
fetched by the SGE complements each other and includes important structural
information with varied contexts. Altogether, these two techniques
substantially cope with the usual information loss involved in graph embedding
techniques, obtaining a more robust graph representation. This fact has been
corroborated through a detailed experimental evaluation on various benchmark
graph datasets, where we outperform the state-of-the-art methods.
|
Replicated Siamese LSTM in Ticketing System for Similarity Learning and
Retrieval in Asymmetric Texts
|
The goal of our industrial ticketing system is to retrieve a relevant
solution for an input query, by matching with historical tickets stored in
knowledge base. A query is comprised of subject and description, while a
historical ticket consists of subject, description and solution. To retrieve a
relevant solution, we use textual similarity paradigm to learn similarity in
the query and historical tickets. The task is challenging due to significant
term mismatch in the query and ticket pairs of asymmetric lengths, where
subject is a short text but description and solution are multi-sentence texts.
We present a novel Replicated Siamese LSTM model to learn similarity in
asymmetric text pairs, that gives 22% and 7% gain (Accuracy@10) for retrieval
task, respectively over unsupervised and supervised baselines. We also show
that the topic and distributed semantic features for short and long texts
improved both similarity learning and retrieval.
|
Learning The Sequential Temporal Information with Recurrent Neural
Networks
|
Recurrent Networks are one of the most powerful and promising artificial
neural network algorithms to processing the sequential data such as natural
languages, sound, time series data. Unlike traditional feed-forward network,
Recurrent Network has a inherent feed back loop that allows to store the
temporal context information and pass the state of information to the entire
sequences of the events. This helps to achieve the state of art performance in
many important tasks such as language modeling, stock market prediction, image
captioning, speech recognition, machine translation and object tracking etc.,
However, training the fully connected RNN and managing the gradient flow are
the complicated process. Many studies are carried out to address the mentioned
limitation. This article is intent to provide the brief details about recurrent
neurons, its variances and trips & tricks to train the fully recurrent neural
network. This review work is carried out as a part of our IPO studio software
module 'Multiple Object Tracking'.
|
Large Margin Few-Shot Learning
|
The key issue of few-shot learning is learning to generalize. This paper
proposes a large margin principle to improve the generalization capacity of
metric based methods for few-shot learning. To realize it, we develop a unified
framework to learn a more discriminative metric space by augmenting the
classification loss function with a large margin distance loss function for
training. Extensive experiments on two state-of-the-art few-shot learning
methods, graph neural networks and prototypical networks, show that our method
can improve the performance of existing models substantially with very little
computational overhead, demonstrating the effectiveness of the large margin
principle and the potential of our method.
|
Separability is not the best goal for machine learning
|
Neural networks use their hidden layers to transform input data into linearly
separable data clusters, with a linear or a perceptron type output layer making
the final projection on the line perpendicular to the discriminating
hyperplane. For complex data with multimodal distributions this transformation
is difficult to learn. Projection on $k\geq 2$ line segments is the simplest
extension of linear separability, defining much easier goal for the learning
process. Simple problems are 2-separable, but problems with inherent complex
logic may be solved in a simple way by $k$-separable projections. The
difficulty of learning non-linear data distributions is shifted to separation
of line intervals, simplifying the transformation of data by hidden network
layers. For classification of difficult Boolean problems, such as the parity
problem, linear projection combined with \ksep is sufficient and provides a
powerful new target for learning. More complex targets may also be defined,
changing the goal of learning from linear discrimination to creation of data
distributions that can easily be handled by specialized models selected to
analyze output distributions. This approach can replace many layers of
transformation required by deep learning models.
|
Machine Learning in High Energy Physics Community White Paper
|
Machine learning has been applied to several problems in particle physics
research, beginning with applications to high-level physics analysis in the
1990s and 2000s, followed by an explosion of applications in particle and event
identification and reconstruction in the 2010s. In this document we discuss
promising future research and development areas for machine learning in
particle physics. We detail a roadmap for their implementation, software and
hardware resource requirements, collaborative initiatives with the data science
community, academia and industry, and training the particle physics community
in data science. The main objective of the document is to connect and motivate
these areas of research and development with the physics drivers of the
High-Luminosity Large Hadron Collider and future neutrino experiments and
identify the resource needs for their implementation. Additionally we identify
areas where collaboration with external communities will be of great benefit.
|
Stochastic Block Model for Hypergraphs: Statistical limits and a
semidefinite programming approach
|
We study the problem of community detection in a random hypergraph model
which we call the stochastic block model for $k$-uniform hypergraphs ($k$-SBM).
We investigate the exact recovery problem in $k$-SBM and show that a sharp
phase transition occurs around a threshold: below the threshold it is
impossible to recover the communities with non-vanishing probability, yet above
the threshold there is an estimator which recovers the communities almost
asymptotically surely. We also consider a simple, efficient algorithm for the
exact recovery problem which is based on a semidefinite relaxation technique.
|
Auto Deep Compression by Reinforcement Learning Based Actor-Critic
Structure
|
Model-based compression is an effective, facilitating, and expanded model of
neural network models with limited computing and low power. However,
conventional models of compression techniques utilize crafted features [2,3,12]
and explore specialized areas for exploration and design of large spaces in
terms of size, speed, and accuracy, which usually have returns Less and time is
up. This paper will effectively analyze deep auto compression (ADC) and
reinforcement learning strength in an effective sample and space design, and
improve the compression quality of the model. The results of compression of the
advanced model are obtained without any human effort and in a completely
automated way. With a 4- fold reduction in FLOP, the accuracy of 2.8% is higher
than the manual compression model for VGG-16 in ImageNet.
|
Automated labeling of bugs and tickets using attention-based mechanisms
in recurrent neural networks
|
We explore solutions for automated labeling of content in bug trackers and
customer support systems. In order to do that, we classify content in terms of
several criteria, such as priority or product area. In the first part of the
paper, we provide an overview of existing methods used for text classification.
These methods fall into two categories - the ones that rely on neural networks
and the ones that don't. We evaluate results of several solutions of both
kinds. In the second part of the paper we present our own recurrent neural
network solution based on hierarchical attention paradigm. It consists of
several Hierarchical Attention network blocks with varying Gated Recurrent Unit
cell sizes and a complementary shallow network that goes alongside. Lastly, we
evaluate above-mentioned methods when predicting fields from two datasets -
Arch Linux bug tracker and Chromium bug tracker. Our contributions include a
comprehensive benchmark between a variety of methods on relevant datasets; a
novel solution that outperforms previous generation methods; and two new
datasets that are made public for further research.
|
Partial Policy-based Reinforcement Learning for Anatomical Landmark
Localization in 3D Medical Images
|
Deploying the idea of long-term cumulative return, reinforcement learning has
shown remarkable performance in various fields. We propose a formulation of the
landmark localization in 3D medical images as a reinforcement learning problem.
Whereas value-based methods have been widely used to solve similar problems, we
adopt an actor-critic based direct policy search method framed in a temporal
difference learning approach. Successful behavior learning is challenging in
large state and/or action spaces, requiring many trials. We introduce a partial
policy-based reinforcement learning to enable solving the large problem of
localization by learning the optimal policy on smaller partial domains.
Independent actors efficiently learn the corresponding partial policies, each
utilizing their own independent critic. The proposed policy reconstruction from
the partial policies ensures a robust and efficient localization utilizing the
sub-agents solving simple binary decision problems in their corresponding
partial action spaces. The proposed reinforcement learning requires a small
number of trials to learn the optimal behavior compared with the original
behavior learning scheme.
|
Model Agnostic Supervised Local Explanations
|
Model interpretability is an increasingly important component of practical
machine learning. Some of the most common forms of interpretability systems are
example-based, local, and global explanations. One of the main challenges in
interpretability is designing explanation systems that can capture aspects of
each of these explanation types, in order to develop a more thorough
understanding of the model. We address this challenge in a novel model called
MAPLE that uses local linear modeling techniques along with a dual
interpretation of random forests (both as a supervised neighborhood approach
and as a feature selection method). MAPLE has two fundamental advantages over
existing interpretability systems. First, while it is effective as a black-box
explanation system, MAPLE itself is a highly accurate predictive model that
provides faithful self explanations, and thus sidesteps the typical
accuracy-interpretability trade-off. Specifically, we demonstrate, on several
UCI datasets, that MAPLE is at least as accurate as random forests and that it
produces more faithful local explanations than LIME, a popular interpretability
system. Second, MAPLE provides both example-based and local explanations and
can detect global patterns, which allows it to diagnose limitations in its
local explanations.
|
Domain2Vec: Deep Domain Generalization
|
We address the problem of domain generalization where a decision function is
learned from the data of several related domains, and the goal is to apply it
on an unseen domain successfully. It is assumed that there is plenty of labeled
data available in source domains (also called as training domain), but no
labeled data is available for the unseen domain (also called a target domain or
test domain). We propose a novel neural network architecture, Domain2Vec (D2V)
that learns domain-specific embedding and then uses this embedding to
generalize the learning across related domains. The proposed algorithm, D2V
extends the idea of distribution regression and kernelized domain
generalization to the neural networks setting. We propose a neural network
architecture to learn domain-specific embedding and then use this embedding
along with the data point specific features to label it. We show the
effectiveness of the architecture by accurately estimating domain to domain
similarity. We evaluate our algorithm against standard domain generalization
datasets for image classification and outperform other state of the art
algorithms.
|
Zero-shot Domain Adaptation without Domain Semantic Descriptors
|
We propose a method to infer domain-specific models such as classifiers for
unseen domains, from which no data are given in the training phase, without
domain semantic descriptors. When training and test distributions are
different, standard supervised learning methods perform poorly. Zero-shot
domain adaptation attempts to alleviate this problem by inferring models that
generalize well to unseen domains by using training data in multiple source
domains. Existing methods use observed semantic descriptors characterizing
domains such as time information to infer the domain-specific models for the
unseen domains. However, it cannot always be assumed that such metadata can be
used in real-world applications. The proposed method can infer appropriate
domain-specific models without any semantic descriptors by introducing the
concept of latent domain vectors, which are latent representations for the
domains and are used for inferring the models. The latent domain vector for the
unseen domain is inferred from the set of the feature vectors in the
corresponding domain, which is given in the testing phase. The domain-specific
models consist of two components: the first is for extracting a representation
of a feature vector to be predicted, and the second is for inferring model
parameters given the latent domain vector. The posterior distributions of the
latent domain vectors and the domain-specific models are parametrized by neural
networks, and are optimized by maximizing the variational lower bound using
stochastic gradient descent. The effectiveness of the proposed method was
demonstrated through experiments using one regression and two classification
tasks.
|
Jointly learning relevant subgraph patterns and nonlinear models of
their indicators
|
Classification and regression in which the inputs are graphs of arbitrary
size and shape have been paid attention in various fields such as computational
chemistry and bioinformatics. Subgraph indicators are often used as the most
fundamental features, but the number of possible subgraph patterns are
intractably large due to the combinatorial explosion. We propose a novel
efficient algorithm to jointly learn relevant subgraph patterns and nonlinear
models of their indicators. Previous methods for such joint learning of
subgraph features and models are based on search for single best subgraph
features with specific pruning and boosting procedures of adding their
indicators one by one, which result in linear models of subgraph indicators. In
contrast, the proposed approach is based on directly learning regression trees
for graph inputs using a newly derived bound of the total sum of squares for
data partitions by a given subgraph feature, and thus can learn nonlinear
models through standard gradient boosting. An illustrative example we call the
Graph-XOR problem to consider nonlinearity, numerical experiments with real
datasets, and scalability comparisons to naive approaches using explicit
pattern enumeration are also presented.
|
Decreasing the size of the Restricted Boltzmann machine
|
We propose a method to decrease the number of hidden units of the restricted
Boltzmann machine while avoiding decrease of the performance measured by the
Kullback-Leibler divergence. Then, we demonstrate our algorithm by using
numerical simulations.
|
XNOR Neural Engine: a Hardware Accelerator IP for 21.6 fJ/op Binary
Neural Network Inference
|
Binary Neural Networks (BNNs) are promising to deliver accuracy comparable to
conventional deep neural networks at a fraction of the cost in terms of memory
and energy. In this paper, we introduce the XNOR Neural Engine (XNE), a fully
digital configurable hardware accelerator IP for BNNs, integrated within a
microcontroller unit (MCU) equipped with an autonomous I/O subsystem and hybrid
SRAM / standard cell memory. The XNE is able to fully compute convolutional and
dense layers in autonomy or in cooperation with the core in the MCU to realize
more complex behaviors. We show post-synthesis results in 65nm and 22nm
technology for the XNE IP and post-layout results in 22nm for the full MCU
indicating that this system can drop the energy cost per binary operation to
21.6fJ per operation at 0.4V, and at the same time is flexible and performant
enough to execute state-of-the-art BNN topologies such as ResNet-34 in less
than 2.2mJ per frame at 8.9 fps.
|
Constraint-based Causal Discovery for Non-Linear Structural Causal
Models with Cycles and Latent Confounders
|
We address the problem of causal discovery from data, making use of the
recently proposed causal modeling framework of modular structural causal models
(mSCM) to handle cycles, latent confounders and non-linearities. We introduce
{\sigma}-connection graphs ({\sigma}-CG), a new class of mixed graphs
(containing undirected, bidirected and directed edges) with additional
structure, and extend the concept of {\sigma}-separation, the appropriate
generalization of the well-known notion of d-separation in this setting, to
apply to {\sigma}-CGs. We prove the closedness of {\sigma}-separation under
marginalisation and conditioning and exploit this to implement a test of
{\sigma}-separation on a {\sigma}-CG. This then leads us to the first causal
discovery algorithm that can handle non-linear functional relations, latent
confounders, cyclic causal relationships, and data from different (stochastic)
perfect interventions. As a proof of concept, we show on synthetic data how
well the algorithm recovers features of the causal graph of modular structural
causal models.
|
Pioneer Networks: Progressively Growing Generative Autoencoder
|
We introduce a novel generative autoencoder network model that learns to
encode and reconstruct images with high quality and resolution, and supports
smooth random sampling from the latent space of the encoder. Generative
adversarial networks (GANs) are known for their ability to simulate random
high-quality images, but they cannot reconstruct existing images. Previous
works have attempted to extend GANs to support such inference but, so far, have
not delivered satisfactory high-quality results. Instead, we propose the
Progressively Growing Generative Autoencoder (PIONEER) network which achieves
high-quality reconstruction with $128{\times}128$ images without requiring a
GAN discriminator. We merge recent techniques for progressively building up the
parts of the network with the recently introduced adversarial encoder-generator
network. The ability to reconstruct input images is crucial in many real-world
applications, and allows for precise intelligent manipulation of existing
images. We show promising results in image synthesis and inference, with
state-of-the-art results in CelebA inference tasks.
|
Glow: Generative Flow with Invertible 1x1 Convolutions
|
Flow-based generative models (Dinh et al., 2014) are conceptually attractive
due to tractability of the exact log-likelihood, tractability of exact
latent-variable inference, and parallelizability of both training and
synthesis. In this paper we propose Glow, a simple type of generative flow
using an invertible 1x1 convolution. Using our method we demonstrate a
significant improvement in log-likelihood on standard benchmarks. Perhaps most
strikingly, we demonstrate that a generative model optimized towards the plain
log-likelihood objective is capable of efficient realistic-looking synthesis
and manipulation of large images. The code for our model is available at
https://github.com/openai/glow
|
Deep Learning for Singing Processing: Achievements, Challenges and
Impact on Singers and Listeners
|
This paper summarizes some recent advances on a set of tasks related to the
processing of singing using state-of-the-art deep learning techniques. We
discuss their achievements in terms of accuracy and sound quality, and the
current challenges, such as availability of data and computing resources. We
also discuss the impact that these advances do and will have on listeners and
singers when they are integrated in commercial applications.
|
Position-aware Self-attention with Relative Positional Encodings for
Slot Filling
|
This paper describes how to apply self-attention with relative positional
encodings to the task of relation extraction. We propose to use the
self-attention encoder layer together with an additional position-aware
attention layer that takes into account positions of the query and the object
in the sentence. The self-attention encoder also uses a custom implementation
of relative positional encodings which allow each word in the sentence to take
into account its left and right context. The evaluation of the model is done on
the TACRED dataset. The proposed model relies only on attention (no recurrent
or convolutional layers are used), while improving performance w.r.t. the
previous state of the art.
|
Temporal Difference Learning with Neural Networks - Study of the Leakage
Propagation Problem
|
Temporal-Difference learning (TD) [Sutton, 1988] with function approximation
can converge to solutions that are worse than those obtained by Monte-Carlo
regression, even in the simple case of on-policy evaluation. To increase our
understanding of the problem, we investigate the issue of approximation errors
in areas of sharp discontinuities of the value function being further
propagated by bootstrap updates. We show empirical evidence of this leakage
propagation, and show analytically that it must occur, in a simple Markov
chain, when function approximation errors are present. For reversible policies,
the result can be interpreted as the tension between two terms of the loss
function that TD minimises, as recently described by [Ollivier, 2018]. We show
that the upper bounds from [Tsitsiklis and Van Roy, 1997] hold, but they do not
imply that leakage propagation occurs and under what conditions. Finally, we
test whether the problem could be mitigated with a better state representation,
and whether it can be learned in an unsupervised manner, without rewards or
privileged information.
|
Mammography Assessment using Multi-Scale Deep Classifiers
|
Applying deep learning methods to mammography assessment has remained a
challenging topic. Dense noise with sparse expressions, mega-pixel raw data
resolution, lack of diverse examples have all been factors affecting
performance. The lack of pixel-level ground truths have especially limited
segmentation methods in pushing beyond approximately bounding regions. We
propose a classification approach grounded in high performance tissue
assessment as an alternative to all-in-one localization and assessment models
that is also capable of pinpointing the causal pixels. First, the objective of
the mammography assessment task is formalized in the context of local tissue
classifiers. Then, the accuracy of a convolutional neural net is evaluated on
classifying patches of tissue with suspicious findings at varying scales, where
highest obtained AUC is above $0.9$. The local evaluations of one such expert
tissue classifier is used to augment the results of a heatmap regression model
and additionally recover the exact causal regions at high resolution as a
saliency image suitable for clinical settings.
|
Robust Text-to-SQL Generation with Execution-Guided Decoding
|
We consider the problem of neural semantic parsing, which translates natural
language questions into executable SQL queries. We introduce a new mechanism,
execution guidance, to leverage the semantics of SQL. It detects and excludes
faulty programs during the decoding procedure by conditioning on the execution
of partially generated program. The mechanism can be used with any
autoregressive generative model, which we demonstrate on four state-of-the-art
recurrent or template-based semantic parsing models. We demonstrate that
execution guidance universally improves model performance on various
text-to-SQL datasets with different scales and query complexity: WikiSQL, ATIS,
and GeoQuery. As a result, we achieve new state-of-the-art execution accuracy
of 83.8% on WikiSQL.
|
Sampling and Inference for Beta Neutral-to-the-Left Models of Sparse
Networks
|
Empirical evidence suggests that heavy-tailed degree distributions occurring
in many real networks are well-approximated by power laws with exponents $\eta$
that may take values either less than and greater than two. Models based on
various forms of exchangeability are able to capture power laws with $\eta <
2$, and admit tractable inference algorithms; we draw on previous results to
show that $\eta > 2$ cannot be generated by the forms of exchangeability used
in existing random graph models. Preferential attachment models generate power
law exponents greater than two, but have been of limited use as statistical
models due to the inherent difficulty of performing inference in
non-exchangeable models. Motivated by this gap, we design and implement
inference algorithms for a recently proposed class of models that generates
$\eta$ of all possible values. We show that although they are not exchangeable,
these models have probabilistic structure amenable to inference. Our methods
make a large class of previously intractable models useful for statistical
inference.
|
Deep Global-Connected Net With The Generalized Multi-Piecewise ReLU
Activation in Deep Learning
|
Recent Progress has shown that exploitation of hidden layer neurons in
convolution neural networks incorporating with a carefully designed activation
function can yield better classification results in the field of computer
vision. The paper firstly introduces a novel deep learning architecture aiming
to mitigate the gradient-vanishing problem, in which the earlier hidden layer
neurons could be directly connected with the last hidden layer and feed into
the last layer for classification. We then design a generalized linear
rectifier function as the activation function that can approximate arbitrary
complex functions via training of the parameters. We will show that our design
can achieve similar performance in a number of object recognition and video
action benchmark tasks, under significantly less number of parameters and
shallower network infrastructure, which is not only promising in training in
terms of computation burden and memory usage, but is also applicable to
low-computation, low-memory mobile scenarios.
|
Deep Semantic Segmentation in an AUV for Online Posidonia Oceanica
Meadows identification
|
Recent studies have shown evidence of a significant decline of the Posidonia
oceanica (P.O.) meadows on a global scale. The monitoring and mapping of these
meadows are fundamental tools for measuring their status. We present an
approach based on a deep neural network to automatically perform a
high-precision semantic segmentation of P.O. meadows in sea-floor images,
offering several improvements over the state of the art techniques. Our network
demonstrates outstanding performance over diverse test sets, reaching a
precision of 96.57% and an accuracy of 96.81%, surpassing the reliability of
labelling the images manually. Also, the network is implemented in an
Autonomous Underwater Vehicle (AUV), performing an online P.O. segmentation,
which will be used to generate real-time semantic coverage maps.
|
Towards Radiologist-Level Accurate Deep Learning System for Pulmonary
Screening
|
In this work, we propose advanced pneumonia and Tuberculosis grading system
for X-ray images. The proposed system is a very deep fully convolutional
classification network with online augmentation that outputs confidence values
for diseases prevalence. Its a fully automated system capable of disease
feature understanding without any offline preprocessing step or manual feature
extraction. We have achieved state- of-the- art performance on the public
databases such as ChestXray-14, Mendeley, Shenzhen Hospital X-ray and Belarus
X-ray set.
|
Estimating Bicycle Route Attractivity from Image Data
|
This master thesis focuses on practical application of Convolutional Neural
Network models on the task of road labeling with bike attractivity score. We
start with an abstraction of real world locations into nodes and scored edges
in partially annotated dataset. We enhance information available about each
edge with photographic data from Google Street View service and with additional
neighborhood information from Open Street Map database. We teach a model on
this enhanced dataset and experiment with ImageNet Large Scale Visual
Recognition Competition. We try different dataset enhancing techniques as well
as various model architectures to improve road scoring. We also make use of
transfer learning to use features from a task with rich dataset of ImageNet
into our task with smaller number of images, to prevent model overfitting.
|
Fast Dynamic Convolutional Neural Networks for Visual Tracking
|
Most of the existing tracking methods based on CNN(convolutional neural
networks) are too slow for real-time application despite the excellent tracking
precision compared with the traditional ones. In this paper, a fast dynamic
visual tracking algorithm combining CNN based MDNet(Multi-Domain Network) and
RoIAlign was developed. The major problem of MDNet also lies in the time
efficiency. Considering the computational complexity of MDNet is mainly caused
by the large amount of convolution operations and fine-tuning of the network
during tracking, a RoIPool layer which could conduct the convolution over the
whole image instead of each RoI is added to accelerate the convolution and a
new strategy of fine-tuning the fully-connected layers is used to accelerate
the update. With RoIPool employed, the computation speed has been increased but
the tracking precision has dropped simultaneously. RoIPool could lose some
positioning precision because it can not handle locations represented by
floating numbers. So RoIAlign, instead of RoIPool, which can process floating
numbers of locations by bilinear interpolation has been added to the network.
The results show the target localization precision has been improved and it
hardly increases the computational cost. These strategies can accelerate the
processing and make it 7x faster than MDNet with very low impact on precision
and it can run at around 7 fps. The proposed algorithm has been evaluated on
two benchmarks: OTB100 and VOT2016, on which high precision and speed have been
obtained. The influence of the network structure and training data are also
discussed with experiments.
|
Outfit Generation and Style Extraction via Bidirectional LSTM and
Autoencoder
|
When creating an outfit, style is a criterion in selecting each fashion item.
This means that style can be regarded as a feature of the overall outfit.
However, in various previous studies on outfit generation, there have been few
methods focusing on global information obtained from an outfit. To address this
deficiency, we have incorporated an unsupervised style extraction module into a
model to learn outfits. Using the style information of an outfit as a whole,
the proposed model succeeded in generating outfits more flexibly without
requiring additional information. Moreover, the style information extracted by
the proposed model is easy to interpret. The proposed model was evaluated on
two human-generated outfit datasets. In a fashion item prediction task (missing
prediction task), the proposed model outperformed a baseline method. In a style
extraction task, the proposed model extracted some easily distinguishable
styles. In an outfit generation task, the proposed model generated an outfit
while controlling its styles. This capability allows us to generate fashionable
outfits according to various preferences.
|
Deep Networks with Shape Priors for Nucleus Detection
|
Detection of cell nuclei in microscopic images is a challenging research
topic, because of limitations in cellular image quality and diversity of
nuclear morphology, i.e. varying nuclei shapes, sizes, and overlaps between
multiple cell nuclei. This has been a topic of enduring interest with promising
recent success shown by deep learning methods. These methods train for example
convolutional neural networks (CNNs) with a training set of input images and
known, labeled nuclei locations. Many of these methods are supplemented by
spatial or morphological processing. We develop a new approach that we call
Shape Priors with Convolutional Neural Networks (SP-CNN) to perform
significantly enhanced nuclei detection. A set of canonical shapes is prepared
with the help of a domain expert. Subsequently, we present a new network
structure that can incorporate `expected behavior' of nucleus shapes via two
components: {\em learnable} layers that perform the nucleus detection and a
{\em fixed} processing part that guides the learning with prior information.
Analytically, we formulate a new regularization term that is targeted at
penalizing false positives while simultaneously encouraging detection inside
cell nucleus boundary. Experimental results on a challenging dataset reveal
that SP-CNN is competitive with or outperforms several state-of-the-art
methods.
|
Faster Bounding Box Annotation for Object Detection in Indoor Scenes
|
This paper proposes an approach for rapid bounding box annotation for object
detection datasets. The procedure consists of two stages: The first step is to
annotate a part of the dataset manually, and the second step proposes
annotations for the remaining samples using a model trained with the first
stage annotations. We experimentally study which first/second stage split
minimizes to total workload. In addition, we introduce a new fully labeled
object detection dataset collected from indoor scenes. Compared to other indoor
datasets, our collection has more class categories, different backgrounds,
lighting conditions, occlusion and high intra-class differences. We train deep
learning based object detectors with a number of state-of-the-art models and
compare them in terms of speed and accuracy. The fully annotated dataset is
released freely available for the research community.
|
Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning
|
This paper presents KeypointNet, an end-to-end geometric reasoning framework
to learn an optimal set of category-specific 3D keypoints, along with their
detectors. Given a single image, KeypointNet extracts 3D keypoints that are
optimized for a downstream task. We demonstrate this framework on 3D pose
estimation by proposing a differentiable objective that seeks the optimal set
of keypoints for recovering the relative pose between two views of an object.
Our model discovers geometrically and semantically consistent keypoints across
viewing angles and instances of an object category. Importantly, we find that
our end-to-end framework using no ground-truth keypoint annotations outperforms
a fully supervised baseline using the same neural network architecture on the
task of pose estimation. The discovered 3D keypoints on the car, chair, and
plane categories of ShapeNet are visualized at http://keypointnet.github.io/.
|
Deep Spatio-Temporal Random Fields for Efficient Video Segmentation
|
In this work we introduce a time- and memory-efficient method for structured
prediction that couples neuron decisions across both space at time. We show
that we are able to perform exact and efficient inference on a densely
connected spatio-temporal graph by capitalizing on recent advances on deep
Gaussian Conditional Random Fields (GCRFs). Our method, called VideoGCRF is (a)
efficient, (b) has a unique global minimum, and (c) can be trained end-to-end
alongside contemporary deep networks for video understanding. We experiment
with multiple connectivity patterns in the temporal domain, and present
empirical improvements over strong baselines on the tasks of both semantic and
instance segmentation of videos.
|
Learning models for visual 3D localization with implicit mapping
|
We consider learning based methods for visual localization that do not
require the construction of explicit maps in the form of point clouds or
voxels. The goal is to learn an implicit representation of the environment at a
higher, more abstract level. We propose to use a generative approach based on
Generative Query Networks (GQNs, Eslami et al. 2018), asking the following
questions: 1) Can GQN capture more complex scenes than those it was originally
demonstrated on? 2) Can GQN be used for localization in those scenes? To study
this approach we consider procedurally generated Minecraft worlds, for which we
can generate images of complex 3D scenes along with camera pose coordinates. We
first show that GQNs, enhanced with a novel attention mechanism can capture the
structure of 3D scenes in Minecraft, as evidenced by their samples. We then
apply the models to the localization problem, comparing the results to a
discriminative baseline, and comparing the ways each approach captures the task
uncertainty.
|
Jigsaw Puzzle Solving Using Local Feature Co-Occurrences in Deep Neural
Networks
|
Archaeologists are in dire need of automated object reconstruction methods.
Fragments reassembly is close to puzzle problems, which may be solved by
computer vision algorithms. As they are often beaten on most image related
tasks by deep learning algorithms, we study a classification method that can
solve jigsaw puzzles. In this paper, we focus on classifying the relative
position: given a couple of fragments, we compute their local relation (e.g. on
top). We propose several enhancements over the state of the art in this domain,
which is outperformed by our method by 25\%. We propose an original dataset
composed of pictures from the Metropolitan Museum of Art. We propose a greedy
reconstruction method based on the predicted relative positions.
|
Forecasting Disease Trajectories in Alzheimer's Disease Using Deep
Learning
|
Joint models for longitudinal and time-to-event data are commonly used in
longitudinal studies to forecast disease trajectories over time. Despite the
many advantages of joint modeling, the standard forms suffer from limitations
that arise from a fixed model specification and computational difficulties when
applied to large datasets. We adopt a deep learning approach to address these
limitations, enhancing existing methods with the flexibility and scalability of
deep neural networks while retaining the benefits of joint modeling. Using data
from the Alzheimer's Disease Neuroimaging Institute, we show improvements in
performance and scalability compared to traditional methods.
|
Deep Learning Based Sphere Decoding
|
In this paper, a deep learning (DL)-based sphere decoding algorithm is
proposed, where the radius of the decoding hypersphere is learnt by a deep
neural network (DNN). The performance achieved by the proposed algorithm is
very close to the optimal maximum likelihood decoding (MLD) over a wide range
of signal-to-noise ratios (SNRs), while the computational complexity, compared
to existing sphere decoding variants, is significantly reduced. This
improvement is attributed to DNN's ability of intelligently learning the radius
of the hypersphere used in decoding. The expected complexity of the proposed
DL-based algorithm is analytically derived and compared with existing ones. It
is shown that the number of lattice points inside the decoding hypersphere
drastically reduces in the DL- based algorithm in both the average and
worst-case senses. The effectiveness of the proposed algorithm is shown through
simulation for high-dimensional multiple-input multiple-output (MIMO) systems,
using high-order modulations.
|
Sparse Deep Neural Network Exact Solutions
|
Deep neural networks (DNNs) have emerged as key enablers of machine learning.
Applying larger DNNs to more diverse applications is an important challenge.
The computations performed during DNN training and inference are dominated by
operations on the weight matrices describing the DNN. As DNNs incorporate more
layers and more neurons per layers, these weight matrices may be required to be
sparse because of memory limitations. Sparse DNNs are one possible approach,
but the underlying theory is in the early stages of development and presents a
number of challenges, including determining the accuracy of inference and
selecting nonzero weights for training. Associative array algebra has been
developed by the big data community to combine and extend database, matrix, and
graph/network concepts for use in large, sparse data problems. Applying this
mathematics to DNNs simplifies the formulation of DNN mathematics and reveals
that DNNs are linear over oscillating semirings. This work uses associative
array DNNs to construct exact solutions and corresponding perturbation models
to the rectified linear unit (ReLU) DNN equations that can be used to construct
test vectors for sparse DNN implementations over various precisions. These
solutions can be used for DNN verification, theoretical explorations of DNN
properties, and a starting point for the challenge of sparse training.
|
Data Augmentation for Detection of Architectural Distortion in Digital
Mammography using Deep Learning Approach
|
Early detection of breast cancer can increase treatment efficiency.
Architectural Distortion (AD) is a very subtle contraction of the breast tissue
and may represent the earliest sign of cancer. Since it is very likely to be
unnoticed by radiologists, several approaches have been proposed over the years
but none using deep learning techniques. To train a Convolutional Neural
Network (CNN), which is a deep neural architecture, is necessary a huge amount
of data. To overcome this problem, this paper proposes a data augmentation
approach applied to clinical image dataset to properly train a CNN. Results
using receiver operating characteristic analysis showed that with a very
limited dataset we could train a CNN to detect AD in digital mammography with
area under the curve (AUC = 0.74).
|
NAPS: Natural Program Synthesis Dataset
|
We present a program synthesis-oriented dataset consisting of human written
problem statements and solutions for these problems. The problem statements
were collected via crowdsourcing and the program solutions were extracted from
human-written solutions in programming competitions, accompanied by
input/output examples. We propose using this dataset for the program synthesis
tasks aimed for working with real user-generated data. As a baseline we present
few models, with the best model achieving 8.8% accuracy, showcasing both the
complexity of the dataset and large room for future research.
|
Graph of brain structures grading for early detection of Alzheimer's
disease
|
Alzheimer's disease is the most common dementia leading to an irreversible
neurodegenerative process. To date, subject revealed advanced brain structural
alterations when the diagnosis is established. Therefore, an earlier diagnosis
of this dementia is crucial although it is a challenging task. Recently, many
studies have proposed biomarkers to perform early detection of Alzheimer's
disease. Some of them have proposed methods based on inter-subject similarity
while other approaches have investigated framework using intra-subject
variability. In this work, we propose a novel framework combining both
approaches within an efficient graph of brain structures grading. Subsequently,
we demonstrate the competitive performance of the proposed method compared to
state-of-the-art methods.
|
YouTube for Patient Education: A Deep Learning Approach for
Understanding Medical Knowledge from User-Generated Videos
|
YouTube presents an unprecedented opportunity to explore how machine learning
methods can improve healthcare information dissemination. We propose an
interdisciplinary lens that synthesizes machine learning methods with
healthcare informatics themes to address the critical issue of developing a
scalable algorithmic solution to evaluate videos from a health literacy and
patient education perspective. We develop a deep learning method to understand
the level of medical knowledge encoded in YouTube videos. Preliminary results
suggest that we can extract medical knowledge from YouTube videos and classify
videos according to the embedded knowledge with satisfying performance. Deep
learning methods show great promise in knowledge extraction, natural language
understanding, and image classification, especially in an era of
patient-centric care and precision medicine.
|
Approximate k-space models and Deep Learning for fast photoacoustic
reconstruction
|
We present a framework for accelerated iterative reconstructions using a fast
and approximate forward model that is based on k-space methods for
photoacoustic tomography. The approximate model introduces aliasing artefacts
in the gradient information for the iterative reconstruction, but these
artefacts are highly structured and we can train a CNN that can use the
approximate information to perform an iterative reconstruction. We show
feasibility of the method for human in-vivo measurements in a limited-view
geometry. The proposed method is able to produce superior results to total
variation reconstructions with a speed-up of 32 times.
|
The CodRep Machine Learning on Source Code Competition
|
CodRep is a machine learning competition on source code data. It is carefully
designed so that anybody can enter the competition, whether professional
researchers, students or independent scholars, without specific knowledge in
machine learning or program analysis. In particular, it aims at being a common
playground on which the machine learning and the software engineering research
communities can interact. The competition has started on April 14th 2018 and
has ended on October 14th 2018. The CodRep data is hosted at
https://github.com/KTH/CodRep-competition/.
|
Bandit Online Learning with Unknown Delays
|
This paper deals with bandit online learning problems involving feedback of
unknown delay that can emerge in multi-armed bandit (MAB) and bandit convex
optimization (BCO) settings. MAB and BCO require only values of the objective
function involved that become available through feedback, and are used to
estimate the gradient appearing in the corresponding iterative algorithms.
Since the challenging case of feedback with \emph{unknown} delays prevents one
from constructing the sought gradient estimates, existing MAB and BCO
algorithms become intractable. For such challenging setups, delayed
exploration, exploitation, and exponential (DEXP3) iterations, along with
delayed bandit gradient descent (DBGD) iterations are developed for MAB and
BCO, respectively. Leveraging a unified analysis framework, it is established
that the regret of DEXP3 and DBGD are ${\cal O}\big( \sqrt{K\bar{d}(T+D)}
\big)$ and ${\cal O}\big( \sqrt{K(T+D)} \big)$, respectively, where $\bar{d}$
is the maximum delay and $D$ denotes the delay accumulated over $T$ slots.
Numerical tests using both synthetic and real data validate the performance of
DEXP3 and DBGD.
|
Efficient Decentralized Deep Learning by Dynamic Model Averaging
|
We propose an efficient protocol for decentralized training of deep neural
networks from distributed data sources. The proposed protocol allows to handle
different phases of model training equally well and to quickly adapt to concept
drifts. This leads to a reduction of communication by an order of magnitude
compared to periodically communicating state-of-the-art approaches. Moreover,
we derive a communication bound that scales well with the hardness of the
serialized learning problem. The reduction in communication comes at almost no
cost, as the predictive performance remains virtually unchanged. Indeed, the
proposed protocol retains loss bounds of periodically averaging schemes. An
extensive empirical evaluation validates major improvement of the trade-off
between model performance and communication which could be beneficial for
numerous decentralized learning applications, such as autonomous driving, or
voice recognition and image classification on mobile phones.
|
A New ECOC Algorithm for Multiclass Microarray Data Classification
|
The classification of multi-class microarray datasets is a hard task because
of the small samples size in each class and the heavy overlaps among classes.
To effectively solve these problems, we propose novel Error Correcting Output
Code (ECOC) algorithm by Enhance Class Separability related Data Complexity
measures during encoding process, named as ECOCECS. In this algorithm, two
nearest neighbor related DC measures are deployed to extract the intrinsic
overlapping information from microarray data. Our ECOC algorithm aims to search
an optimal class split scheme by minimizing these measures. The class splitting
process ends when each class is separated from others, and then the class
assignment scheme is mapped as a coding matrix. Experiments are carried out on
five microarray datasets, and results demonstrate the effectiveness and
robustness of our method in comparison with six state-of-art ECOC methods. In
short, our work confirm the probability of applying DC to ECOC framework.
|
An Intriguing Failing of Convolutional Neural Networks and the CoordConv
Solution
|
Few ideas have enjoyed as large an impact on deep learning as convolution.
For any problem involving pixels or spatial representations, common intuition
holds that convolutional neural networks may be appropriate. In this paper we
show a striking counterexample to this intuition via the seemingly trivial
coordinate transform problem, which simply requires learning a mapping between
coordinates in (x,y) Cartesian space and one-hot pixel space. Although
convolutional networks would seem appropriate for this task, we show that they
fail spectacularly. We demonstrate and carefully analyze the failure first on a
toy problem, at which point a simple fix becomes obvious. We call this solution
CoordConv, which works by giving convolution access to its own input
coordinates through the use of extra coordinate channels. Without sacrificing
the computational and parametric efficiency of ordinary convolution, CoordConv
allows networks to learn either complete translation invariance or varying
degrees of translation dependence, as required by the end task. CoordConv
solves the coordinate transform problem with perfect generalization and 150
times faster with 10--100 times fewer parameters than convolution. This stark
contrast raises the question: to what extent has this inability of convolution
persisted insidiously inside other tasks, subtly hampering performance from
within? A complete answer to this question will require further investigation,
but we show preliminary evidence that swapping convolution for CoordConv can
improve models on a diverse set of tasks. Using CoordConv in a GAN produced
less mode collapse as the transform between high-level spatial latents and
pixels becomes easier to learn. A Faster R-CNN detection model trained on MNIST
showed 24% better IOU when using CoordConv, and in the RL domain agents playing
Atari games benefit significantly from the use of CoordConv layers.
|
Data Efficient Lithography Modeling with Transfer Learning and Active
Data Selection
|
Lithography simulation is one of the key steps in physical verification,
enabled by the substantial optical and resist models. A resist model bridges
the aerial image simulation to printed patterns. While the effectiveness of
learning-based solutions for resist modeling has been demonstrated, they are
considerably data-demanding. Meanwhile, a set of manufactured data for a
specific lithography configuration is only valid for the training of one single
model, indicating low data efficiency. Due to the complexity of the
manufacturing process, obtaining enough data for acceptable accuracy becomes
very expensive in terms of both time and cost, especially during the evolution
of technology generations when the design space is intensively explored. In
this work, we propose a new resist modeling framework for contact layers,
utilizing existing data from old technology nodes and active selection of data
in a target technology node, to reduce the amount of data required from the
target lithography configuration. Our framework based on transfer learning and
active learning techniques is effective within a competitive range of accuracy,
i.e., 3-10X reduction on the amount of training data with comparable accuracy
to the state-of-the-art learning approach.
|
A Comparative Study on using Principle Component Analysis with Different
Text Classifiers
|
Text categorization (TC) is the task of automatically organizing a set of
documents into a set of pre-defined categories. Over the last few years,
increased attention has been paid to the use of documents in digital form and
this makes text categorization becomes a challenging issue. The most
significant problem of text categorization is its huge number of features. Most
of these features are redundant, noisy and irrelevant that cause over fitting
with most of the classifiers. Hence, feature extraction is an important step to
improve the overall accuracy and the performance of the text classifiers. In
this paper, we will provide an overview of using principle component analysis
(PCA) as a feature extraction with various classifiers. It was observed that
the performance rate of the classifiers after using PCA to reduce the dimension
of data improved. Experiments are conducted on three UCI data sets, Classic03,
CNAE-9 and DBWorld e-mails. We compare the classification performance results
of using PCA with popular and well-known text classifiers. Results show that
using PCA encouragingly enhances classification performance on most of the
classifiers.
|
Dynamic Pricing with Finitely Many Unknown Valuations
|
Motivated by posted price auctions where buyers are grouped in an unknown
number of latent types characterized by their private values for the good on
sale, we investigate revenue maximization in stochastic dynamic pricing when
the distribution of buyers' private values is supported on an unknown set of
points in [0,1] of unknown cardinality $K$. This setting can be viewed as an
instance of a stochastic $K$-armed bandit problem where the location of the
arms (the $K$ unknown valuations) must be learned as well. In the
distribution-free case, we prove that our setting is just as hard as $K$-armed
stochastic bandits: no algorithm can achieve a regret significantly better than
$\sqrt{KT}$, (where T is the time horizon); we present an efficient algorithm
matching this lower bound up to logarithmic factors. In the
distribution-dependent case, we show that for all $K>2$ our setting is strictly
harder than $K$-armed stochastic bandits by proving that it is impossible to
obtain regret bounds that grow logarithmically in time or slower. On the other
hand, when a lower bound $\gamma>0$ on the smallest drop in the demand curve is
known, we prove an upper bound on the regret of order $(1/\Delta+(\log \log
T)/\gamma^2)(K\log T)$. This is a significant improvement on previously known
regret bounds for discontinuous demand curves, that are at best of order
$(K^{12}/\gamma^8)\sqrt{T}$. When $K=2$ in the distribution-dependent case, the
hardness of our setting reduces to that of a stochastic $2$-armed bandit: we
prove that an upper bound of order $(\log T)/\Delta$ (up to $\log\log$ factors)
on the regret can be achieved with no information on the demand curve. Finally,
we show a $O(\sqrt{T})$ upper bound on the regret for the setting in which the
buyers' decisions are nonstochastic, and the regret is measured with respect to
the best between two fixed valuations one of which is known to the seller.
|
Optimization of a SSP's Header Bidding Strategy using Thompson Sampling
|
Over the last decade, digital media (web or app publishers) generalized the
use of real time ad auctions to sell their ad spaces. Multiple auction
platforms, also called Supply-Side Platforms (SSP), were created. Because of
this multiplicity, publishers started to create competition between SSPs. In
this setting, there are two successive auctions: a second price auction in each
SSP and a secondary, first price auction, called header bidding auction,
between SSPs.In this paper, we consider an SSP competing with other SSPs for ad
spaces. The SSP acts as an intermediary between an advertiser wanting to buy ad
spaces and a web publisher wanting to sell its ad spaces, and needs to define a
bidding strategy to be able to deliver to the advertisers as many ads as
possible while spending as little as possible. The revenue optimization of this
SSP can be written as a contextual bandit problem, where the context consists
of the information available about the ad opportunity, such as properties of
the internet user or of the ad placement.Using classical multi-armed bandit
strategies (such as the original versions of UCB and EXP3) is inefficient in
this setting and yields a low convergence speed, as the arms are very
correlated. In this paper we design and experiment a version of the Thompson
Sampling algorithm that easily takes this correlation into account. We combine
this bayesian algorithm with a particle filter, which permits to handle
non-stationarity by sequentially estimating the distribution of the highest bid
to beat in order to win an auction. We apply this methodology on two real
auction datasets, and show that it significantly outperforms more classical
approaches.The strategy defined in this paper is being developed to be deployed
on thousands of publishers worldwide.
|
Adaptive Adversarial Attack on Scene Text Recognition
|
Recent studies have shown that state-of-the-art deep learning models are
vulnerable to the inputs with small perturbations (adversarial examples). We
observe two critical obstacles in adversarial examples: (i) Strong adversarial
attacks (e.g., C&W attack) require manually tuning hyper-parameters and take a
long time to construct an adversarial example, making it impractical to attack
real-time systems; (ii) Most of the studies focus on non-sequential tasks, such
as image classification, yet only a few consider sequential tasks. In this
work, we speed up adversarial attacks, especially on sequential learning tasks.
By leveraging the uncertainty of each task, we directly learn the adaptive
multi-task weightings, without manually searching hyper-parameters. A unified
architecture is developed and evaluated for both non-sequential tasks and
sequential ones. To validate the effectiveness, we take the scene text
recognition task as a case study. To our best knowledge, our proposed method is
the first attempt to adversarial attack for scene text recognition. Adaptive
Attack achieves over 99.9\% success rate with 3-6X speedup compared to
state-of-the-art adversarial attacks.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.