title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Understanding V2V Driving Scenarios through Traffic Primitives
|
Semantically understanding complex drivers' encountering behavior, wherein
two or multiple vehicles are spatially close to each other, does potentially
benefit autonomous car's decision-making design. This paper presents a
framework of analyzing various encountering behaviors through decomposing
driving encounter data into small building blocks, called driving primitives,
using nonparametric Bayesian learning (NPBL) approaches, which offers a
flexible way to gain an insight into the complex driving encounters without any
prerequisite knowledge. The effectiveness of our proposed primitive-based
framework is validated based on 976 naturalistic driving encounters, from which
more than 4000 driving primitives are learned using NPBL - a sticky HDP-HMM,
combined a hidden Markov model (HMM) with a hierarchical Dirichlet process
(HDP). After that, a dynamic time warping method integrated with k-means
clustering is then developed to cluster all these extracted driving primitives
into groups. Experimental results find that there exist 20 kinds of driving
primitives capable of representing the basic components of driving encounters
in our database. This primitive-based analysis methodology potentially reveals
underlying information of vehicle-vehicle encounters for self-driving
applications.
|
Task Recommendation in Crowdsourcing Based on Learning Preferences and
Reliabilities
|
Workers participating in a crowdsourcing platform can have a wide range of
abilities and interests. An important problem in crowdsourcing is the task
recommendation problem, in which tasks that best match a particular worker's
preferences and reliabilities are recommended to that worker. A task
recommendation scheme that assigns tasks more likely to be accepted by a worker
who is more likely to complete it reliably results in better performance for
the task requester. Without prior information about a worker, his preferences
and reliabilities need to be learned over time. In this paper, we propose a
multi-armed bandit (MAB) framework to learn a worker's preferences and his
reliabilities for different categories of tasks. However, unlike the classical
MAB problem, the reward from the worker's completion of a task is unobservable.
We therefore include the use of gold tasks (i.e., tasks whose solutions are
known \emph{a priori} and which do not produce any rewards) in our task
recommendation procedure. Our model could be viewed as a new variant of MAB, in
which the random rewards can only be observed at those time steps where gold
tasks are used, and the accuracy of estimating the expected reward of
recommending a task to a worker depends on the number of gold tasks used. We
show that the optimal regret is $O(\sqrt{n})$, where $n$ is the number of tasks
recommended to the worker. We develop three task recommendation strategies to
determine the number of gold tasks for different task categories, and show that
they are order optimal. Simulations verify the efficiency of our approaches.
|
Rob-GAN: Generator, Discriminator, and Adversarial Attacker
|
We study two important concepts in adversarial deep learning---adversarial
training and generative adversarial network (GAN). Adversarial training is the
technique used to improve the robustness of discriminator by combining
adversarial attacker and discriminator in the training phase. GAN is commonly
used for image generation by jointly optimizing discriminator and generator. We
show these two concepts are indeed closely related and can be used to
strengthen each other---adding a generator to the adversarial training
procedure can improve the robustness of discriminators, and adding an
adversarial attack to GAN training can improve the convergence speed and lead
to better generators. Combining these two insights, we develop a framework
called Rob-GAN to jointly optimize generator and discriminator in the presence
of adversarial attacks---the generator generates fake images to fool
discriminator; the adversarial attacker perturbs real images to fool the
discriminator, and the discriminator wants to minimize loss under fake and
adversarial images. Through this end-to-end training procedure, we are able to
simultaneously improve the convergence speed of GAN training, the quality of
synthetic images, and the robustness of discriminator under strong adversarial
attacks. Experimental results demonstrate that the obtained classifier is more
robust than the state-of-the-art adversarial training approach, and the
generator outperforms SN-GAN on ImageNet-143.
|
Acceleration through Optimistic No-Regret Dynamics
|
We consider the problem of minimizing a smooth convex function by reducing
the optimization to computing the Nash equilibrium of a particular zero-sum
convex-concave game. Zero-sum games can be solved using online learning
dynamics, where a classical technique involves simulating two no-regret
algorithms that play against each other and, after $T$ rounds, the average
iterate is guaranteed to solve the original optimization problem with error
decaying as $O(\log T/T)$. In this paper we show that the technique can be
enhanced to a rate of $O(1/T^2)$ by extending recent work \cite{RS13,SALS15}
that leverages \textit{optimistic learning} to speed up equilibrium
computation. The resulting optimization algorithm derived from this analysis
coincides \textit{exactly} with the well-known \NA \cite{N83a} method, and
indeed the same story allows us to recover several variants of the Nesterov's
algorithm via small tweaks. We are also able to establish the accelerated
linear rate for a function which is both strongly-convex and smooth. This
methodology unifies a number of different iterative optimization methods: we
show that the \HB algorithm is precisely the non-optimistic variant of \NA, and
recent prior work already established a similar perspective on \FW
\cite{AW17,ALLW18}.
|
AXNet: ApproXimate computing using an end-to-end trainable neural
network
|
Neural network based approximate computing is a universal architecture
promising to gain tremendous energy-efficiency for many error resilient
applications. To guarantee the approximation quality, existing works deploy two
neural networks (NNs), e.g., an approximator and a predictor. The approximator
provides the approximate results, while the predictor predicts whether the
input data is safe to approximate with the given quality requirement. However,
it is non-trivial and time-consuming to make these two neural network
coordinate---they have different optimization objectives---by training them
separately. This paper proposes a novel neural network structure---AXNet---to
fuse two NNs to a holistic end-to-end trainable NN. Leveraging the philosophy
of multi-task learning, AXNet can tremendously improve the invocation
(proportion of safe-to-approximate samples) and reduce the approximation error.
The training effort also decrease significantly. Experiment results show 50.7%
more invocation and substantial cuts of training time when compared to existing
neural network based approximate computing framework.
|
Interpreting recurrent neural networks behaviour via excitable network
attractors
|
Introduction: Machine learning provides fundamental tools both for scientific
research and for the development of technologies with significant impact on
society. It provides methods that facilitate the discovery of regularities in
data and that give predictions without explicit knowledge of the rules
governing a system. However, a price is paid for exploiting such flexibility:
machine learning methods are typically black-boxes where it is difficult to
fully understand what the machine is doing or how it is operating. This poses
constraints on the applicability and explainability of such methods. Methods:
Our research aims to open the black-box of recurrent neural networks, an
important family of neural networks used for processing sequential data. We
propose a novel methodology that provides a mechanistic interpretation of
behaviour when solving a computational task. Our methodology uses mathematical
constructs called excitable network attractors, which are invariant sets in
phase space composed of stable attractors and excitable connections between
them. Results and Discussion: As the behaviour of recurrent neural networks
depends both on training and on inputs to the system, we introduce an algorithm
to extract network attractors directly from the trajectory of a neural network
while solving tasks. Simulations conducted on a controlled benchmark task
confirm the relevance of these attractors for interpreting the behaviour of
recurrent neural networks, at least for tasks that involve learning a finite
number of stable states and transitions between them.
|
Pull Message Passing for Nonparametric Belief Propagation
|
We present a "pull" approach to approximate products of Gaussian mixtures
within message updates for Nonparametric Belief Propagation (NBP) inference.
Existing NBP methods often represent messages between continuous-valued latent
variables as Gaussian mixture models. To avoid computational intractability in
loopy graphs, NBP necessitates an approximation of the product of such
mixtures. Sampling-based product approximations have shown effectiveness for
NBP inference. However, such approximations used within the traditional "push"
message update procedures quickly become computationally prohibitive for
multi-modal distributions over high-dimensional variables. In contrast, we
propose a "pull" method, as the Pull Message Passing for Nonparametric Belief
propagation (PMPNBP) algorithm, and demonstrate its viability for efficient
inference. We report results using an experiment from an existing NBP method,
PAMPAS, for inferring the pose of an articulated structure in clutter. Results
from this illustrative problem found PMPNBP has a greater ability to
efficiently scale the number of components in its mixtures and, consequently,
improve inference accuracy.
|
DeepLink: A Novel Link Prediction Framework based on Deep Learning
|
Recently, link prediction has attracted more attentions from various
disciplines such as computer science, bioinformatics and economics. In this
problem, unknown links between nodes are discovered based on numerous
information such as network topology, profile information and user generated
contents. Most of the previous researchers have focused on the structural
features of the networks. While the recent researches indicate that contextual
information can change the network topology. Although, there are number of
valuable researches which combine structural and content information, but they
face with the scalability issue due to feature engineering. Because, majority
of the extracted features are obtained by a supervised or semi supervised
algorithm. Moreover, the existing features are not general enough to indicate
good performance on different networks with heterogeneous structures. Besides,
most of the previous researches are presented for undirected and unweighted
networks. In this paper, a novel link prediction framework called "DeepLink" is
presented based on deep learning techniques. In contrast to the previous
researches which fail to automatically extract best features for the link
prediction, deep learning reduces the manual feature engineering. In this
framework, both the structural and content information of the nodes are
employed. The framework can use different structural feature vectors, which are
prepared by various link prediction methods. It considers all proximity orders
that are presented in a network during the structural feature learning. We have
evaluated the performance of DeepLink on two real social network datasets
including Telegram and irBlogs. On both datasets, the proposed framework
outperforms several structural and hybrid approaches for link prediction
problem.
|
Enhanced Machine Learning Techniques for Early HARQ Feedback Prediction
in 5G
|
We investigate Early Hybrid Automatic Repeat reQuest (E-HARQ) feedback
schemes enhanced by machine learning techniques as a path towards
ultra-reliable and low-latency communication (URLLC). To this end, we propose
machine learning methods to predict the outcome of the decoding process ahead
of the end of the transmission. We discuss different input features and
classification algorithms ranging from traditional methods to newly developed
supervised autoencoders. These methods are evaluated based on their prospects
of complying with the URLLC requirements of effective block error rates below
$10^{-5}$ at small latency overheads. We provide realistic performance
estimates in a system model incorporating scheduling effects to demonstrate the
feasibility of E-HARQ across different signal-to-noise ratios, subcode lengths,
channel conditions and system loads, and show the benefit over regular HARQ and
existing E-HARQ schemes without machine learning.
|
Global and local evaluation of link prediction tasks with neural
embeddings
|
We focus our attention on the link prediction problem for knowledge graphs,
which is treated herein as a binary classification task on neural embeddings of
the entities. By comparing, combining and extending different methodologies for
link prediction on graph-based data coming from different domains, we formalize
a unified methodology for the quality evaluation benchmark of neural embeddings
for knowledge graphs. This benchmark is then used to empirically investigate
the potential of training neural embeddings globally for the entire graph, as
opposed to the usual way of training embeddings locally for a specific
relation. This new way of testing the quality of the embeddings evaluates the
performance of binary classifiers for scalable link prediction with limited
data. Our evaluation pipeline is made open source, and with this we aim to draw
more attention of the community towards an important issue of transparency and
reproducibility of the neural embeddings evaluations.
|
DeepPhase: Surgical Phase Recognition in CATARACTS Videos
|
Automated surgical workflow analysis and understanding can assist surgeons to
standardize procedures and enhance post-surgical assessment and indexing, as
well as, interventional monitoring. Computer-assisted interventional (CAI)
systems based on video can perform workflow estimation through surgical
instruments' recognition while linking them to an ontology of procedural
phases. In this work, we adopt a deep learning paradigm to detect surgical
instruments in cataract surgery videos which in turn feed a surgical phase
inference recurrent network that encodes temporal aspects of phase steps within
the phase classification. Our models present comparable to state-of-the-art
results for surgical tool detection and phase recognition with accuracies of 99
and 78% respectively.
|
The Helmholtz Method: Using Perceptual Compression to Reduce Machine
Learning Complexity
|
This paper proposes a fundamental answer to a frequently asked question in
multimedia computing and machine learning: Do artifacts from perceptual
compression contribute to error in the machine learning process and if so, how
much? Our approach to the problem is a reinterpretation of the Helmholtz Free
Energy formula from physics to explain the relationship between content and
noise when using sensors (such as cameras or microphones) to capture multimedia
data. The reinterpretation allows a bit-measurement of the noise contained in
images, audio, and video by combining a classifier with perceptual compression,
such as JPEG or MP3. Our experiments on CIFAR-10 as well as Fraunhofer's
IDMT-SMT-Audio-Effects dataset indicate that, at the right quality level,
perceptual compression is actually not harmful but contributes to a significant
reduction of complexity of the machine learning process. That is, our noise
quantification method can be used to speed up the training of deep learning
classifiers significantly while maintaining, or sometimes even improving,
overall classification accuracy. Moreover, our results provide insights into
the reasons for the success of deep learning.
|
Embedded Implementation of a Deep Learning Smile Detector
|
In this paper we study the real time deployment of deep learning algorithms
in low resource computational environments. As the use case, we compare the
accuracy and speed of neural networks for smile detection using different
neural network architectures and their system level implementation on NVidia
Jetson embedded platform. We also propose an asynchronous multithreading scheme
for parallelizing the pipeline. Within this framework, we experimentally
compare thirteen widely used network topologies. The experiments show that low
complexity architectures can achieve almost equal performance as larger ones,
with a fraction of computation required.
|
Sparse Range-constrained Learning and Its Application for Medical Image
Grading
|
Sparse learning has been shown to be effective in solving many real-world
problems. Finding sparse representations is a fundamentally important topic in
many fields of science including signal processing, computer vision, genome
study and medical imaging. One important issue in applying sparse
representation is to find the basis to represent the data,especially in
computer vision and medical imaging where the data is not necessary incoherent.
In medical imaging, clinicians often grade the severity or measure the risk
score of a disease based on images. This process is referred to as medical
image grading. Manual grading of the disease severity or risk score is often
used. However, it is tedious, subjective and expensive. Sparse learning has
been used for automatic grading of medical images for different diseases. In
the grading, we usually begin with one step to find a sparse representation of
the testing image using a set of reference images or atoms from the dictionary.
Then in the second step, the selected atoms are used as references to compute
the grades of the testing images. Since the two steps are conducted
sequentially, the objective function in the first step is not necessarily
optimized for the second step. In this paper, we propose a novel sparse
range-constrained learning(SRCL)algorithm for medical image grading.Different
from most of existing sparse learning algorithms, SRCL integrates the objective
of finding a sparse representation and that of grading the image into one
function. It aims to find a sparse representation of the testing image based on
atoms that are most similar in both the data or feature representation and the
medical grading scores. We apply the new proposed SRCL to CDR computation and
cataract grading. Experimental results show that the proposed method is able to
improve the accuracy in cup-to-disc ratio computation and cataract grading.
|
Two-Layer Mixture Network Ensemble for Apparel Attributes Classification
|
Recognizing apparel attributes has recently drawn great interest in the
computer vision community. Methods based on various deep neural networks have
been proposed for image classification, which could be applied to apparel
attributes recognition. An interesting problem raised is how to ensemble these
methods to further improve the accuracy. In this paper, we propose a two-layer
mixture framework for ensemble different networks. In the first layer of this
framework, two types of ensemble learning methods, bagging and boosting, are
separately applied. Different from traditional methods, our bagging process
makes use of the whole training set, not random subsets, to train each model in
the ensemble, where several differentiated deep networks are used to promote
model variance. To avoid the bias of small-scale samples, the second layer only
adopts bagging to mix the results obtained with bagging and boosting in the
first layer. Experimental results demonstrate that the proposed mixture
framework outperforms any individual network model or either independent
ensemble method in apparel attributes classification.
|
LiDAR and Camera Detection Fusion in a Real Time Industrial Multi-Sensor
Collision Avoidance System
|
Collision avoidance is a critical task in many applications, such as ADAS
(advanced driver-assistance systems), industrial automation and robotics. In an
industrial automation setting, certain areas should be off limits to an
automated vehicle for protection of people and high-valued assets. These areas
can be quarantined by mapping (e.g., GPS) or via beacons that delineate a
no-entry area. We propose a delineation method where the industrial vehicle
utilizes a LiDAR {(Light Detection and Ranging)} and a single color camera to
detect passive beacons and model-predictive control to stop the vehicle from
entering a restricted space. The beacons are standard orange traffic cones with
a highly reflective vertical pole attached. The LiDAR can readily detect these
beacons, but suffers from false positives due to other reflective surfaces such
as worker safety vests. Herein, we put forth a method for reducing false
positive detection from the LiDAR by projecting the beacons in the camera
imagery via a deep learning method and validating the detection using a neural
network-learned projection from the camera to the LiDAR space. Experimental
data collected at Mississippi State University's Center for Advanced Vehicular
Systems (CAVS) shows the effectiveness of the proposed system in keeping the
true detection while mitigating false positives.
|
Deep Learning Hyperspectral Image Classification Using Multiple
Class-based Denoising Autoencoders, Mixed Pixel Training Augmentation, and
Morphological Operations
|
Herein, we present a system for hyperspectral image segmentation that
utilizes multiple class--based denoising autoencoders which are efficiently
trained. Moreover, we present a novel hyperspectral data augmentation method
for labelled HSI data using linear mixtures of pixels from each class, which
helps the system with edge pixels which are almost always mixed pixels.
Finally, we utilize a deep neural network and morphological hole-filling to
provide robust image classification. Results run on the Salinas dataset verify
the high performance of the proposed algorithm.
|
Multi-Scale Gradual Integration CNN for False Positive Reduction in
Pulmonary Nodule Detection
|
Lung cancer is a global and dangerous disease, and its early detection is
crucial to reducing the risks of mortality. In this regard, it has been of
great interest in developing a computer-aided system for pulmonary nodules
detection as early as possible on thoracic CT scans. In general, a nodule
detection system involves two steps: (i) candidate nodule detection at a high
sensitivity, which captures many false positives and (ii) false positive
reduction from candidates. However, due to the high variation of nodule
morphological characteristics and the possibility of mistaking them for
neighboring organs, candidate nodule detection remains a challenge. In this
study, we propose a novel Multi-scale Gradual Integration Convolutional Neural
Network (MGI-CNN), designed with three main strategies: (1) to use multi-scale
inputs with different levels of contextual information, (2) to use abstract
information inherent in different input scales with gradual integration, and
(3) to learn multi-stream feature integration in an end-to-end manner. To
verify the efficacy of the proposed network, we conducted exhaustive
experiments on the LUNA16 challenge datasets by comparing the performance of
the proposed method with state-of-the-art methods in the literature. On two
candidate subsets of the LUNA16 dataset, i.e., V1 and V2, our method achieved
an average CPM of 0.908 (V1) and 0.942 (V2), outperforming comparable methods
by a large margin. Our MGI-CNN is implemented in Python using TensorFlow and
the source code is available from 'https://github.com/ku-milab/MGICNN.'
|
EchoFusion: Tracking and Reconstruction of Objects in 4D Freehand
Ultrasound Imaging without External Trackers
|
Ultrasound (US) is the most widely used fetal imaging technique. However, US
images have limited capture range, and suffer from view dependent artefacts
such as acoustic shadows. Compounding of overlapping 3D US acquisitions into a
high-resolution volume can extend the field of view and remove image artefacts,
which is useful for retrospective analysis including population based studies.
However, such volume reconstructions require information about relative
transformations between probe positions from which the individual volumes were
acquired. In prenatal US scans, the fetus can move independently from the
mother, making external trackers such as electromagnetic or optical tracking
unable to track the motion between probe position and the moving fetus. We
provide a novel methodology for image-based tracking and volume reconstruction
by combining recent advances in deep learning and simultaneous localisation and
mapping (SLAM). Tracking semantics are established through the use of a
Residual 3D U-Net and the output is fed to the SLAM algorithm. As a proof of
concept, experiments are conducted on US volumes taken from a whole body fetal
phantom, and from the heads of real fetuses. For the fetal head segmentation,
we also introduce a novel weak annotation approach to minimise the required
manual effort for ground truth annotation. We evaluate our method
qualitatively, and quantitatively with respect to tissue discrimination
accuracy and tracking robustness.
|
Uncertainty and Interpretability in Convolutional Neural Networks for
Semantic Segmentation of Colorectal Polyps
|
Convolutional Neural Networks (CNNs) are propelling advances in a range of
different computer vision tasks such as object detection and object
segmentation. Their success has motivated research in applications of such
models for medical image analysis. If CNN-based models are to be helpful in a
medical context, they need to be precise, interpretable, and uncertainty in
predictions must be well understood. In this paper, we develop and evaluate
recent advances in uncertainty estimation and model interpretability in the
context of semantic segmentation of polyps from colonoscopy images. We evaluate
and enhance several architectures of Fully Convolutional Networks (FCNs) for
semantic segmentation of colorectal polyps and provide a comparison between
these models. Our highest performing model achieves a 76.06\% mean IOU accuracy
on the EndoScene dataset, a considerable improvement over the previous
state-of-the-art.
|
A Modality-Adaptive Method for Segmenting Brain Tumors and
Organs-at-Risk in Radiation Therapy Planning
|
In this paper we present a method for simultaneously segmenting brain tumors
and an extensive set of organs-at-risk for radiation therapy planning of
glioblastomas. The method combines a contrast-adaptive generative model for
whole-brain segmentation with a new spatial regularization model of tumor shape
using convolutional restricted Boltzmann machines. We demonstrate
experimentally that the method is able to adapt to image acquisitions that
differ substantially from any available training data, ensuring its
applicability across treatment sites; that its tumor segmentation accuracy is
comparable to that of the current state of the art; and that it captures most
organs-at-risk sufficiently well for radiation therapy planning purposes. The
proposed method may be a valuable step towards automating the delineation of
brain tumors and organs-at-risk in glioblastoma patients undergoing radiation
therapy.
|
Metric Embedding Autoencoders for Unsupervised Cross-Dataset Transfer
Learning
|
Cross-dataset transfer learning is an important problem in person
re-identification (Re-ID). Unfortunately, not too many deep transfer Re-ID
models exist for realistic settings of practical Re-ID systems. We propose a
purely deep transfer Re-ID model consisting of a deep convolutional neural
network and an autoencoder. The latent code is divided into metric embedding
and nuisance variables. We then utilize an unsupervised training method that
does not rely on co-training with non-deep models. Our experiments show
improvements over both the baseline and competitors' transfer learning models.
|
A post-processing method to improve the white matter hyperintensity
segmentation accuracy for randomly-initialized U-net
|
White matter hyperintensity (WMH) is commonly found in elder individuals and
appears to be associated with brain diseases. U-net is a convolutional network
that has been widely used for biomedical image segmentation. Recently, U-net
has been successfully applied to WMH segmentation. Random initialization is
usally used to initialize the model weights in the U-net. However, the model
may coverage to different local optima with different randomly initialized
weights. We find a combination of thresholding and averaging the outputs of
U-nets with different random initializations can largely improve the WMH
segmentation accuracy. Based on this observation, we propose a post-processing
technique concerning the way how averaging and thresholding are conducted.
Specifically, we first transfer the score maps from three U-nets to binary
masks via thresholding and then average those binary masks to obtain the final
WMH segmentation. Both quantitative analysis (via the Dice similarity
coefficient) and qualitative analysis (via visual examinations) reveal the
superior performance of the proposed method. This post-processing technique is
independent of the model used. As such, it can also be applied to situations
where other deep learning models are employed, especially when random
initialization is adopted and pre-training is unavailable.
|
A Trace Lasso Regularized L1-norm Graph Cut for Highly Correlated Noisy
Hyperspectral Image
|
This work proposes an adaptive trace lasso regularized L1-norm based graph
cut method for dimensionality reduction of Hyperspectral images, called as
`Trace Lasso-L1 Graph Cut' (TL-L1GC). The underlying idea of this method is to
generate the optimal projection matrix by considering both the sparsity as well
as the correlation of the data samples. The conventional L2-norm used in the
objective function is sensitive to noise and outliers. Therefore, in this work
L1-norm is utilized as a robust alternative to L2-norm. Besides, for further
improvement of the results, we use a penalty function of trace lasso with the
L1GC method. It adaptively balances the L2-norm and L1-norm simultaneously by
considering the data correlation along with the sparsity. We obtain the optimal
projection matrix by maximizing the ratio of between-class dispersion to
within-class dispersion using L1-norm with trace lasso as the penalty.
Furthermore, an iterative procedure for this TL-L1GC method is proposed to
solve the optimization function. The effectiveness of this proposed method is
evaluated on two benchmark HSI datasets.
|
Learning low dimensional word based linear classifiers using Data Shared
Adaptive Bootstrap Aggregated Lasso with application to IMDb data
|
In this article we propose a new supervised ensemble learning method called
Data Shared Adaptive Bootstrap Aggregated (AdaBag) Lasso for capturing low
dimensional useful features for word based sentiment analysis and mining
problems. The literature on ensemble methods is very rich in both statistics
and machine learning. The algorithm is a substantial upgrade of the Data Shared
Lasso uplift algorithm. The most significant conceptual addition to the
existing literature lies in the final selection of bag of predictors through a
special bootstrap aggregation scheme. We apply the algorithm to one simulated
data and perform dimension reduction in grouped IMDb data (drama, comedy and
horror) to extract reduced set of word features for predicting sentiment
ratings of movie reviews demonstrating different aspects. We also compare the
performance of the present method with the classical Principal Components with
associated Linear Discrimination (PCA-LD) as baseline. There are few
limitations in the algorithm. Firstly, the algorithm workflow does not
incorporate online sequential data acquisition and it does not use sentence
based models which are common in ANN algorithms . Our results produce slightly
higher error rate compare to the reported state-of-the-art as a consequence.
|
Dynamical Component Analysis (DyCA): Dimensionality Reduction For
High-Dimensional Deterministic Time-Series
|
Multivariate signal processing is often based on dimensionality reduction
techniques. We propose a new method, Dynamical Component Analysis (DyCA),
leading to a classification of the underlying dynamics and - for a certain type
of dynamics - to a signal subspace representing the dynamics of the data. In
this paper the algorithm is derived leading to a generalized eigenvalue problem
of correlation matrices. The application of the DyCA on high-dimensional
chaotic signals is presented both for simulated data as well as real EEG data
of epileptic seizures.
|
Experimental Implementation of a Quantum Autoencoder via Quantum Adders
|
Quantum autoencoders allow for reducing the amount of resources in a quantum
computation by mapping the original Hilbert space onto a reduced space with the
relevant information. Recently, it was proposed to employ approximate quantum
adders to implement quantum autoencoders in quantum technologies. Here, we
carry out the experimental implementation of this proposal in the Rigetti cloud
quantum computer employing up to three qubits. The experimental fidelities are
in good agreement with the theoretical prediction, thus proving the feasibility
to realize quantum autoencoders via quantum adders in state-of-the-art
superconducting quantum technologies.
|
On the overfly algorithm in deep learning of neural networks
|
In this paper we investigate the supervised backpropagation training of
multilayer neural networks from a dynamical systems point of view. We discuss
some links with the qualitative theory of differential equations and introduce
the overfly algorithm to tackle the local minima problem. Our approach is based
on the existence of first integrals of the generalised gradient system with
build-in dissipation.
|
Resource-Size matters: Improving Neural Named Entity Recognition with
Optimized Large Corpora
|
This study improves the performance of neural named entity recognition by a
margin of up to 11% in F-score on the example of a low-resource language like
German, thereby outperforming existing baselines and establishing a new
state-of-the-art on each single open-source dataset. Rather than designing
deeper and wider hybrid neural architectures, we gather all available resources
and perform a detailed optimization and grammar-dependent morphological
processing consisting of lemmatization and part-of-speech tagging prior to
exposing the raw data to any training process. We test our approach in a
threefold monolingual experimental setup of a) single, b) joint, and c)
optimized training and shed light on the dependency of downstream-tasks on the
size of corpora used to compute word embeddings.
|
Combining Restricted Boltzmann Machines with Neural Networks for Latent
Truth Discovery
|
Latent truth discovery, LTD for short, refers to the problem of aggregating
ltiple claims from various sources in order to estimate the plausibility of
atements about entities. In the absence of a ground truth, this problem is
highly challenging, when some sources provide conflicting claims and others no
claims at all. In this work we provide an unsupervised stochastic inference
procedure on top of a model that combines restricted Boltzmann machines with
feed-forward neural networks to accurately infer the reliability of sources as
well as the plausibility of statements about entities. In comparison to prior
work our approach stands out (1) by allowing the incorporation of arbitrary
features about sources and claims, (2) by generalizing from reliability per
source towards a reliability function, and thus (3) enabling the estimation of
source reliability even for sources that have provided no or very few claims,
(4) by building on efficient and scalable stochastic inference algorithms, and
(5) by outperforming the state-of-the-art by a considerable margin.
|
Learnable: Theory vs Applications
|
Two different views on machine learning problem: Applied learning (machine
learning with business applications) and Agnostic PAC learning are formalized
and compared here. I show that, under some conditions, the theory of PAC
Learnable provides a way to solve the Applied learning problem. However, the
theory requires to have the training sets so large, that it would make the
learning practically useless. I suggest shedding some theoretical
misconceptions about learning to make the theory more aligned with the needs
and experience of practitioners.
|
Infinite Mixture of Inverted Dirichlet Distributions
|
In this work, we develop a novel Bayesian estimation method for the Dirichlet
process (DP) mixture of the inverted Dirichlet distributions, which has been
shown to be very flexible for modeling vectors with positive elements. The
recently proposed extended variational inference (EVI) framework is adopted to
derive an analytically tractable solution. The convergency of the proposed
algorithm is theoretically guaranteed by introducing single lower bound
approximation to the original objective function in the VI framework. In
principle, the proposed model can be viewed as an infinite inverted Dirichelt
mixture model (InIDMM) that allows the automatic determination of the number of
mixture components from data. Therefore, the problem of pre-determining the
optimal number of mixing components has been overcome. Moreover, the problems
of over-fitting and under-fitting are avoided by the Bayesian estimation
approach. Comparing with several recently proposed DP-related methods, the good
performance and effectiveness of the proposed method have been demonstrated
with both synthesized data and real data evaluations.
|
FPGA-Based CNN Inference Accelerator Synthesized from Multi-Threaded C
Software
|
A deep-learning inference accelerator is synthesized from a C-language
software program parallelized with Pthreads. The software implementation uses
the well-known producer/consumer model with parallel threads interconnected by
FIFO queues. The LegUp high-level synthesis (HLS) tool synthesizes threads into
parallel FPGA hardware, translating software parallelism into spatial
parallelism. A complete system is generated where convolution, pooling and
padding are realized in the synthesized accelerator, with remaining tasks
executing on an embedded ARM processor. The accelerator incorporates reduced
precision, and a novel approach for zero-weight-skipping in convolution. On a
mid-sized Intel Arria 10 SoC FPGA, peak performance on VGG-16 is 138 effective
GOPS.
|
End-to-end Deep Learning from Raw Sensor Data: Atrial Fibrillation
Detection using Wearables
|
We present a convolutional-recurrent neural network architecture with long
short-term memory for real-time processing and classification of digital sensor
data. The network implicitly performs typical signal processing tasks such as
filtering and peak detection, and learns time-resolved embeddings of the input
signal. We use a prototype multi-sensor wearable device to collect over 180h of
photoplethysmography (PPG) data sampled at 20Hz, of which 36h are during atrial
fibrillation (AFib). We use end-to-end learning to achieve state-of-the-art
results in detecting AFib from raw PPG data. For classification labels output
every 0.8s, we demonstrate an area under ROC curve of 0.9999, with false
positive and false negative rates both below $2\times 10^{-3}$. This
constitutes a significant improvement on previous results utilising
domain-specific feature engineering, such as heart rate extraction, and brings
large-scale atrial fibrillation screenings within imminent reach.
|
Deep PDF: Probabilistic Surface Optimization and Density Estimation
|
A probability density function (pdf) encodes the entire stochastic knowledge
about data distribution, where data may represent stochastic observations in
robotics, transition state pairs in reinforcement learning or any other
empirically acquired modality. Inferring data pdf is of prime importance,
allowing to analyze various model hypotheses and perform smart decision making.
However, most density estimation techniques are limited in their representation
expressiveness to specific kernel type or predetermined distribution family,
and have other restrictions. For example, kernel density estimation (KDE)
methods require meticulous parameter search and are extremely slow at querying
new points. In this paper we present a novel non-parametric density estimation
approach, DeepPDF, that uses a neural network to approximate a target pdf given
samples from thereof. Such a representation provides high inference accuracy
for a wide range of target pdfs using a relatively simple network structure,
making our method highly statistically robust. This is done via a new
stochastic optimization algorithm, \emph{Probabilistic Surface Optimization}
(PSO), that turns to advantage the stochastic nature of sample points in order
to force network output to be identical to the output of a target pdf. Once
trained, query point evaluation can be efficiently done in DeepPDF by a simple
network forward pass, with linear complexity in the number of query points.
Moreover, the PSO algorithm is capable of inferring the frequency of data
samples and may also be used in other statistical tasks such as conditional
estimation and distribution transformation. We compare the derived approach
with KDE methods showing its superior performance and accuracy.
|
Dictionary Learning in Fourier Transform Scanning Tunneling Spectroscopy
|
Modern high-resolution microscopes, such as the scanning tunneling
microscope, are commonly used to study specimens that have dense and aperiodic
spatial structure. Extracting meaningful information from images obtained from
such microscopes remains a formidable challenge. Fourier analysis is commonly
used to analyze the underlying structure of fundamental motifs present in an
image. However, the Fourier transform fundamentally suffers from severe phase
noise when applied to aperiodic images. Here, we report the development of a
new algorithm based on nonconvex optimization, applicable to any microscopy
modality, that directly uncovers the fundamental motifs present in a real-space
image. Apart from being quantitatively superior to traditional Fourier
analysis, we show that this novel algorithm also uncovers phase sensitive
information about the underlying motif structure. We demonstrate its usefulness
by studying scanning tunneling microscopy images of a Co-doped iron arsenide
superconductor and prove that the application of the algorithm allows for the
complete recovery of quasiparticle interference in this material. Our phase
sensitive quasiparticle interference imaging results indicate that the pairing
symmetry in optimally doped NaFeAs is consistent with a sign-changing s+- order
parameter.
|
A writer-independent approach for offline signature verification using
deep convolutional neural networks features
|
The use of features extracted using a deep convolutional neural network (CNN)
combined with a writer-dependent (WD) SVM classifier resulted in significant
improvement in performance of handwritten signature verification (HSV) when
compared to the previous state-of-the-art methods. In this work it is
investigated whether the use of these CNN features provide good results in a
writer-independent (WI) HSV context, based on the dichotomy transformation
combined with the use of an SVM writer-independent classifier. The experiments
performed in the Brazilian and GPDS datasets show that (i) the proposed
approach outperformed other WI-HSV methods from the literature, (ii) in the
global threshold scenario, the proposed approach was able to outperform the
writer-dependent method with CNN features in the Brazilian dataset, (iii) in an
user threshold scenario, the results are similar to those obtained by the
writer-dependent method with CNN features.
|
False Positive Reduction by Actively Mining Negative Samples for
Pulmonary Nodule Detection in Chest Radiographs
|
Generating large quantities of quality labeled data in medical imaging is
very time consuming and expensive. The performance of supervised algorithms for
various tasks on imaging has improved drastically over the years, however the
availability of data to train these algorithms have become one of the main
bottlenecks for implementation. To address this, we propose a semi-supervised
learning method where pseudo-negative labels from unlabeled data are used to
further refine the performance of a pulmonary nodule detection network in chest
radiographs. After training with the proposed network, the false positive rate
was reduced to 0.1266 from 0.4864 while maintaining sensitivity at 0.89.
|
One-Shot Generation of Near-Optimal Topology through Theory-Driven
Machine Learning
|
We introduce a theory-driven mechanism for learning a neural network model
that performs generative topology design in one shot given a problem setting,
circumventing the conventional iterative process that computational design
tasks usually entail. The proposed mechanism can lead to machines that quickly
response to new design requirements based on its knowledge accumulated through
past experiences of design generation. Achieving such a mechanism through
supervised learning would require an impractically large amount of
problem-solution pairs for training, due to the known limitation of deep neural
networks in knowledge generalization. To this end, we introduce an interaction
between a student (the neural network) and a teacher (the optimality conditions
underlying topology optimization): The student learns from existing data and is
tested on unseen problems. Deviation of the student's solutions from the
optimality conditions is quantified, and used for choosing new data points to
learn from. We call this learning mechanism "theory-driven", as it explicitly
uses domain-specific theories to guide the learning, thus distinguishing itself
from purely data-driven supervised learning. We show through a compliance
minimization problem that the proposed learning mechanism leads to topology
generation with near-optimal structural compliance, much improved from standard
supervised learning under the same computational budget.
|
Improving Neural Sequence Labelling using Additional Linguistic
Information
|
Sequence labelling is the task of assigning categorical labels to a data
sequence. In Natural Language Processing, sequence labelling can be applied to
various fundamental problems, such as Part of Speech (POS) tagging, Named
Entity Recognition (NER), and Chunking. In this study, we propose a method to
add various linguistic features to the neural sequence framework to improve
sequence labelling. Besides word level knowledge, sense embeddings are added to
provide semantic information. Additionally, selective readings of character
embeddings are added to capture contextual as well as morphological features
for each word in a sentence. Compared to previous methods, these added
linguistic features allow us to design a more concise model and perform more
efficient training. Our proposed architecture achieves state of the art results
on the benchmark datasets of POS, NER, and chunking. Moreover, the convergence
rate of our model is significantly better than the previous state of the art
models.
|
MoCoNet: Motion Correction in 3D MPRAGE images using a Convolutional
Neural Network approach
|
Purpose: The suppression of motion artefacts from MR images is a challenging
task. The purpose of this paper is to develop a standalone novel technique to
suppress motion artefacts from MR images using a data-driven deep learning
approach. Methods: A deep learning convolutional neural network (CNN) was
developed to remove motion artefacts in brain MR images. A CNN was trained on
simulated motion corrupted images to identify and suppress artefacts due to the
motion. The network was an encoder-decoder CNN architecture where the encoder
decomposed the motion corrupted images into a set of feature maps. The feature
maps were then combined by the decoder network to generate a motion-corrected
image. The network was tested on an unseen simulated dataset and an
experimental, motion corrupted in vivo brain dataset. Results: The trained
network was able to suppress the motion artefacts in the simulated motion
corrupted images, and the mean percentage error in the motion corrected images
was 2.69 % with a standard deviation of 0.95 %. The network was able to
effectively suppress the motion artefacts from the experimental dataset,
demonstrating the generalisation capability of the trained network. Conclusion:
A novel and generic motion correction technique has been developed that can
suppress motion artefacts from motion corrupted MR images. The proposed
technique is a standalone post-processing method that does not interfere with
data acquisition or reconstruction parameters, thus making it suitable for a
multitude of MR sequences.
|
TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzing
|
Machine learning models are notoriously difficult to interpret and debug.
This is particularly true of neural networks. In this work, we introduce
automated software testing techniques for neural networks that are well-suited
to discovering errors which occur only for rare inputs. Specifically, we
develop coverage-guided fuzzing (CGF) methods for neural networks. In CGF,
random mutations of inputs to a neural network are guided by a coverage metric
toward the goal of satisfying user-specified constraints. We describe how fast
approximate nearest neighbor algorithms can provide this coverage metric. We
then discuss the application of CGF to the following goals: finding numerical
errors in trained neural networks, generating disagreements between neural
networks and quantized versions of those networks, and surfacing undesirable
behavior in character level language models. Finally, we release an open source
library called TensorFuzz that implements the described techniques.
|
Transportation Modes Classification Using Feature Engineering
|
Predicting transportation modes from GPS (Global Positioning System) records
is a hot topic in the trajectory mining domain. Each GPS record is called a
trajectory point and a trajectory is a sequence of these points. Trajectory
mining has applications including but not limited to transportation mode
detection, tourism, traffic congestion, smart cities management, animal
behaviour analysis, environmental preservation, and traffic dynamics are some
of the trajectory mining applications. Transportation modes prediction as one
of the tasks in human mobility and vehicle mobility applications plays an
important role in resource allocation, traffic management systems, tourism
planning and accident detection. In this work, the proposed framework in Etemad
et al. is extended to consider other aspects in the task of transportation
modes prediction. Wrapper search and information retrieval methods were
investigated to find the best subset of trajectory features. Finding the best
classifier and the best feature subset, the framework is compared against two
related papers that applied deep learning methods. The results show that our
framework achieved better performance. Moreover, the ground truth noise removal
improved accuracy of transportation modes prediction task; however, the
assumption of having access to test set labels in pre-processing task is
invalid. Furthermore, the cross validation approaches were investigated and the
performance results show that the random cross validation method provides
optimistic results.
|
Bike Flow Prediction with Multi-Graph Convolutional Networks
|
One fundamental issue in managing bike sharing systems is the bike flow
prediction. Due to the hardness of predicting the flow for a single station,
recent research works often predict the bike flow at cluster-level. While such
studies gain satisfactory prediction accuracy, they cannot directly guide some
fine-grained bike sharing system management issues at station-level. In this
paper, we revisit the problem of the station-level bike flow prediction, aiming
to boost the prediction accuracy leveraging the breakthroughs of deep learning
techniques. We propose a new multi-graph convolutional neural network model to
predict the bike flow at station-level, where the key novelty is viewing the
bike sharing system from the graph perspective. More specifically, we construct
multiple inter-station graphs for a bike sharing system. In each graph, nodes
are stations, and edges are a certain type of relations between stations. Then,
multiple graphs are constructed to reflect heterogeneous relationships (e.g.,
distance, ride record correlation). Afterward, we fuse the multiple graphs and
then apply the convolutional layers on the fused graph to predict station-level
future bike flow. In addition to the estimated bike flow value, our model also
gives the prediction confidence interval so as to help the bike sharing system
managers make decisions. Using New York City and Chicago bike sharing data for
experiments, our model can outperform state-of-the-art station-level prediction
models by reducing 25.1% and 17.0% of prediction error in New York City and
Chicago, respectively.
|
Group-sparse SVD Models and Their Applications in Biological Data
|
Sparse Singular Value Decomposition (SVD) models have been proposed for
biclustering high dimensional gene expression data to identify block patterns
with similar expressions. However, these models do not take into account prior
group effects upon variable selection. To this end, we first propose
group-sparse SVD models with group Lasso (GL1-SVD) and group L0-norm penalty
(GL0-SVD) for non-overlapping group structure of variables. However, such
group-sparse SVD models limit their applicability in some problems with
overlapping structure. Thus, we also propose two group-sparse SVD models with
overlapping group Lasso (OGL1-SVD) and overlapping group L0-norm penalty
(OGL0-SVD). We first adopt an alternating iterative strategy to solve GL1-SVD
based on a block coordinate descent method, and GL0-SVD based on a projection
method. The key of solving OGL1-SVD is a proximal operator with overlapping
group Lasso penalty. We employ an alternating direction method of multipliers
(ADMM) to solve the proximal operator. Similarly, we develop an approximate
method to solve OGL0-SVD. Applications of these methods and comparison with
competing ones using simulated data demonstrate their effectiveness. Extensive
applications of them onto several real gene expression data with gene prior
group knowledge identify some biologically interpretable gene modules.
|
Improving Sequential Determinantal Point Processes for Supervised Video
Summarization
|
It is now much easier than ever before to produce videos. While the
ubiquitous video data is a great source for information discovery and
extraction, the computational challenges are unparalleled. Automatically
summarizing the videos has become a substantial need for browsing, searching,
and indexing visual content. This paper is in the vein of supervised video
summarization using sequential determinantal point process (SeqDPP), which
models diversity by a probabilistic distribution. We improve this model in two
folds. In terms of learning, we propose a large-margin algorithm to address the
exposure bias problem in SeqDPP. In terms of modeling, we design a new
probabilistic distribution such that, when it is integrated into SeqDPP, the
resulting model accepts user input about the expected length of the summary.
Moreover, we also significantly extend a popular video summarization dataset by
1) more egocentric videos, 2) dense user annotations, and 3) a refined
evaluation scheme. We conduct extensive experiments on this dataset (about 60
hours of videos in total) and compare our approach to several competitive
baselines.
|
Optimal Tap Setting of Voltage Regulation Transformers Using Batch
Reinforcement Learning
|
In this paper, we address the problem of setting the tap positions of load
tap changers (LTCs) for voltage regulation in radial power distribution systems
under uncertain load dynamics. The objective is to find a policy to determine
the tap positions that only uses measurements of voltage magnitudes and
topology information so as to minimize the voltage deviation across the system.
We formulate this problem as a Markov decision process (MDP), and propose a
batch reinforcement learning (RL) algorithm to solve it. By taking advantage of
a linearized power flow model, we propose an effective algorithm to estimate
the voltage magnitudes under different tap settings, which allows the RL
algorithm to explore the state and action spaces freely offline without
impacting the system operation. To circumvent the "curse of dimensionality"
resulted from the large state and action spaces, we propose a sequential
learning algorithm to learn an action-value function for each LTC, based on
which the optimal tap positions can be directly determined. The effectiveness
of the proposed algorithm is validated via numerical simulations on the IEEE
13-bus and 123-bus distribution test feeders.
|
A Margin-based MLE for Crowdsourced Partial Ranking
|
A preference order or ranking aggregated from pairwise comparison data is
commonly understood as a strict total order. However, in real-world scenarios,
some items are intrinsically ambiguous in comparisons, which may very well be
an inherent uncertainty of the data. In this case, the conventional total order
ranking can not capture such uncertainty with mere global ranking or utility
scores. In this paper, we are specifically interested in the recent surge in
crowdsourcing applications to predict partial but more accurate (i.e., making
less incorrect statements) orders rather than complete ones. To do so, we
propose a novel framework to learn some probabilistic models of partial orders
as a \emph{margin-based Maximum Likelihood Estimate} (MLE) method. We prove
that the induced MLE is a joint convex optimization problem with respect to all
the parameters, including the global ranking scores and margin parameter.
Moreover, three kinds of generalized linear models are studied, including the
basic uniform model, Bradley-Terry model, and Thurstone-Mosteller model,
equipped with some theoretical analysis on FDR and Power control for the
proposed methods. The validity of these models are supported by experiments
with both simulated and real-world datasets, which shows that the proposed
models exhibit improvements compared with traditional state-of-the-art
algorithms.
|
A Survey of Machine and Deep Learning Methods for Internet of Things
(IoT) Security
|
The Internet of Things (IoT) integrates billions of smart devices that can
communicate with one another with minimal human intervention. It is one of the
fastest developing fields in the history of computing, with an estimated 50
billion devices by the end of 2020. On the one hand, IoT play a crucial role in
enhancing several real-life smart applications that can improve life quality.
On the other hand, the crosscutting nature of IoT systems and the
multidisciplinary components involved in the deployment of such systems
introduced new security challenges. Implementing security measures, such as
encryption, authentication, access control, network security and application
security, for IoT devices and their inherent vulnerabilities is ineffective.
Therefore, existing security methods should be enhanced to secure the IoT
system effectively. Machine learning and deep learning (ML/DL) have advanced
considerably over the last few years, and machine intelligence has transitioned
from laboratory curiosity to practical machinery in several important
applications. Consequently, ML/DL methods are important in transforming the
security of IoT systems from merely facilitating secure communication between
devices to security-based intelligence systems. The goal of this work is to
provide a comprehensive survey of ML /DL methods that can be used to develop
enhanced security methods for IoT systems. IoT security threats that are
related to inherent or newly introduced threats are presented, and various
potential IoT system attack surfaces and the possible threats related to each
surface are discussed. We then thoroughly review ML/DL methods for IoT security
and present the opportunities, advantages and shortcomings of each method. We
discuss the opportunities and challenges involved in applying ML/DL to IoT
security. These opportunities and challenges can serve as potential future
research directions.
|
Consistent polynomial-time unseeded graph matching for Lipschitz
graphons
|
We propose a consistent polynomial-time method for the unseeded node matching
problem for networks with smooth underlying structures. Despite widely
conjectured by the research community that the structured graph matching
problem to be significantly easier than its worst case counterpart, well-known
to be NP-hard, the statistical version of the problem has stood a challenge
that resisted any solution both provable and polynomial-time. The closest
existing work requires quasi-polynomial time. Our method is based on the latest
advances in graphon estimation techniques and analysis on the concentration of
empirical Wasserstein distances. Its core is a simple yet unconventional
sampling-and-matching scheme that reduces the problem from unseeded to seeded.
Our method allows flexible efficiencies, is convenient to analyze and
potentially can be extended to more general settings. Our work enables a rich
variety of subsequent estimations and inferences.
|
Visual Analogies between Atari Games for Studying Transfer Learning in
RL
|
In this work, we ask the following question: Can visual analogies, learned in
an unsupervised way, be used in order to transfer knowledge between pairs of
games and even play one game using an agent trained for another game? We
attempt to answer this research question by creating visual analogies between a
pair of games: a source game and a target game. For example, given a video
frame in the target game, we map it to an analogous state in the source game
and then attempt to play using a trained policy learned for the source game. We
demonstrate convincing visual mapping between four pairs of games (eight
mappings), which are used to evaluate three transfer learning approaches.
|
Towards Automatic Speech Identification from Vocal Tract Shape Dynamics
in Real-time MRI
|
Vocal tract configurations play a vital role in generating distinguishable
speech sounds, by modulating the airflow and creating different resonant
cavities in speech production. They contain abundant information that can be
utilized to better understand the underlying speech production mechanism. As a
step towards automatic mapping of vocal tract shape geometry to acoustics, this
paper employs effective video action recognition techniques, like Long-term
Recurrent Convolutional Networks (LRCN) models, to identify different
vowel-consonant-vowel (VCV) sequences from dynamic shaping of the vocal tract.
Such a model typically combines a CNN based deep hierarchical visual feature
extractor with Recurrent Networks, that ideally makes the network
spatio-temporally deep enough to learn the sequential dynamics of a short video
clip for video classification tasks. We use a database consisting of 2D
real-time MRI of vocal tract shaping during VCV utterances by 17 speakers. The
comparative performances of this class of algorithms under various parameter
settings and for various classification tasks are discussed. Interestingly, the
results show a marked difference in the model performance in the context of
speech classification with respect to generic sequence or video classification
tasks.
|
StructADMM: A Systematic, High-Efficiency Framework of Structured Weight
Pruning for DNNs
|
Weight pruning methods of DNNs have been demonstrated to achieve a good model
pruning rate without loss of accuracy, thereby alleviating the significant
computation/storage requirements of large-scale DNNs. Structured weight pruning
methods have been proposed to overcome the limitation of irregular network
structure and demonstrated actual GPU acceleration. However, in prior work the
pruning rate (degree of sparsity) and GPU acceleration are limited (to less
than 50%) when accuracy needs to be maintained. In this work,we overcome these
limitations by proposing a unified, systematic framework of structured weight
pruning for DNNs. It is a framework that can be used to induce different types
of structured sparsity, such as filter-wise, channel-wise, and shape-wise
sparsity, as well non-structured sparsity. The proposed framework incorporates
stochastic gradient descent with ADMM, and can be understood as a dynamic
regularization method in which the regularization target is analytically
updated in each iteration. Without loss of accuracy on the AlexNet model, we
achieve 2.58X and 3.65X average measured speedup on two GPUs, clearly
outperforming the prior work. The average speedups reach 3.15X and 8.52X when
allowing a moderate ac-curacy loss of 2%. In this case the model compression
for convolutional layers is 15.0X, corresponding to 11.93X measured CPU
speedup. Our experiments on ResNet model and on other data sets like UCF101 and
CIFAR-10 demonstrate the consistently higher performance of our framework.
|
Neural Mesh: Introducing a Notion of Space and Conservation of Energy to
Neural Networks
|
Neural networks are based on a simplified model of the brain. In this
project, we wanted to relax the simplifying assumptions of a traditional neural
network by making a model that more closely emulates the low level interactions
of neurons. Like in an RNN, our model has a state that persists between time
steps, so that the energies of neurons persist. However, unlike an RNN, our
state consists of a 2 dimensional matrix, rather than a 1 dimensional vector,
thereby introducing a concept of distance to other neurons within the state. In
our model, neurons can only fire to adjacent neurons, as in the brain. Like in
the brain, we only allow neurons to fire in a time step if they contain enough
energy, or excitement. We also enforce a notion of conservation of energy, so
that a neuron cannot excite its neighbors more than the excitement it already
contained at that time step. Taken together, these two features allow signals
in the form of activations to flow around in our network over time, making our
neural mesh more closely model signals traveling through the brain the brain.
Although our main goal is to design an architecture to more closely emulate the
brain in the hope of having a correct internal representation of information by
the time we know how to properly train a general intelligence, we did benchmark
our neural mash on a specific task. We found that by increasing the runtime of
the mesh, we were able to increase its accuracy without increasing the number
of parameters.
|
Microsoft Dialogue Challenge: Building End-to-End Task-Completion
Dialogue Systems
|
This proposal introduces a Dialogue Challenge for building end-to-end
task-completion dialogue systems, with the goal of encouraging the dialogue
research community to collaborate and benchmark on standard datasets and
unified experimental environment. In this special session, we will release
human-annotated conversational data in three domains (movie-ticket booking,
restaurant reservation, and taxi booking), as well as an experiment platform
with built-in simulators in each domain, for training and evaluation purposes.
The final submitted systems will be evaluated both in simulated setting and by
human judges.
|
ARM: Augment-REINFORCE-Merge Gradient for Stochastic Binary Networks
|
To backpropagate the gradients through stochastic binary layers, we propose
the augment-REINFORCE-merge (ARM) estimator that is unbiased, exhibits low
variance, and has low computational complexity. Exploiting variable
augmentation, REINFORCE, and reparameterization, the ARM estimator achieves
adaptive variance reduction for Monte Carlo integration by merging two
expectations via common random numbers. The variance-reduction mechanism of the
ARM estimator can also be attributed to either antithetic sampling in an
augmented space, or the use of an optimal anti-symmetric "self-control"
baseline function together with the REINFORCE estimator in that augmented
space. Experimental results show the ARM estimator provides state-of-the-art
performance in auto-encoding variational inference and maximum likelihood
estimation, for discrete latent variable models with one or multiple stochastic
binary layers. Python code for reproducible research is publicly available.
|
Transformationally Identical and Invariant Convolutional Neural Networks
by Combining Symmetric Operations or Input Vectors
|
Transformationally invariant processors constructed by transformed input
vectors or operators have been suggested and applied to many applications. In
this study, transformationally identical processing based on combining results
of all sub-processes with corresponding transformations at one of the
processing steps or at the beginning step were found to be equivalent for a
given condition. This property can be applied to most convolutional neural
network (CNN) systems. Specifically, a transformationally identical CNN can be
constructed by arranging internally symmetric operations in parallel with the
same transformation family that includes a flatten layer with weights sharing
among their corresponding transformation elements. Other transformationally
identical CNNs can be constructed by averaging transformed input vectors of the
family at the input layer followed by an ordinary CNN process or by a set of
symmetric operations. Interestingly, we found that both types of
transformationally identical CNN systems are mathematically equivalent by
either applying an averaging operation to corresponding elements of all
sub-channels before the activation function or without using a non-linear
activation function.
|
Robust Student Network Learning
|
Deep neural networks bring in impressive accuracy in various applications,
but the success often relies on the heavy network architecture. Taking
well-trained heavy networks as teachers, classical teacher-student learning
paradigm aims to learn a student network that is lightweight yet accurate. In
this way, a portable student network with significantly fewer parameters can
achieve a considerable accuracy which is comparable to that of teacher network.
However, beyond accuracy, robustness of the learned student network against
perturbation is also essential for practical uses. Existing teacher-student
learning frameworks mainly focus on accuracy and compression ratios, but ignore
the robustness. In this paper, we make the student network produce more
confident predictions with the help of the teacher network, and analyze the
lower bound of the perturbation that will destroy the confidence of the student
network. Two important objectives regarding prediction scores and gradients of
examples are developed to maximize this lower bound, so as to enhance the
robustness of the student network without sacrificing the performance.
Experiments on benchmark datasets demonstrate the efficiency of the proposed
approach to learn robust student networks which have satisfying accuracy and
compact sizes.
|
Lead Sheet Generation and Arrangement by Conditional Generative
Adversarial Network
|
Research on automatic music generation has seen great progress due to the
development of deep neural networks. However, the generation of
multi-instrument music of arbitrary genres still remains a challenge. Existing
research either works on lead sheets or multi-track piano-rolls found in MIDIs,
but both musical notations have their limits. In this work, we propose a new
task called lead sheet arrangement to avoid such limits. A new recurrent
convolutional generative model for the task is proposed, along with three new
symbolic-domain harmonic features to facilitate learning from unpaired lead
sheets and MIDIs. Our model can generate lead sheets and their arrangements of
eight-bar long. Audio samples of the generated result can be found at
https://drive.google.com/open?id=1c0FfODTpudmLvuKBbc23VBCgQizY6-Rk
|
A Group-Theoretic Approach to Computational Abstraction: Symmetry-Driven
Hierarchical Clustering
|
Abstraction plays a key role in concept learning and knowledge discovery;
this paper is concerned with computational abstraction. In particular, we study
the nature of abstraction through a group-theoretic approach, formalizing it as
symmetry-driven---as opposed to data-driven---hierarchical clustering. Thus,
the resulting clustering framework is data-free, feature-free, similarity-free,
and globally hierarchical---the four key features that distinguish it from
common data clustering models such as $k$-means. Beyond a theoretical
foundation for abstraction, we also present a top-down and a bottom-up approach
to establish an algorithmic foundation for practical abstraction-generating
methods. Lastly, via both a theoretical explanation and a real-world
application, we illustrate that further coupling of our abstraction framework
with statistics realizes Shannon's information lattice and even further, brings
learning into the picture. This not only presents one use case of our proposed
computational abstraction, but also gives a first step towards a principled and
cognitive way of automatic concept learning and knowledge discovery.
|
Online Learning with an Almost Perfect Expert
|
We study the multiclass online learning problem where a forecaster makes a
sequence of predictions using the advice of $n$ experts. Our main contribution
is to analyze the regime where the best expert makes at most $b$ mistakes and
to show that when $b = o(\log_4{n})$, the expected number of mistakes made by
the optimal forecaster is at most $\log_4{n} + o(\log_4{n})$. We also describe
an adversary strategy showing that this bound is tight and that the worst case
is attained for binary prediction.
|
Highly Scalable Deep Learning Training System with Mixed-Precision:
Training ImageNet in Four Minutes
|
Synchronized stochastic gradient descent (SGD) optimizers with data
parallelism are widely used in training large-scale deep neural networks.
Although using larger mini-batch sizes can improve the system scalability by
reducing the communication-to-computation ratio, it may hurt the generalization
ability of the models. To this end, we build a highly scalable deep learning
training system for dense GPU clusters with three main contributions: (1) We
propose a mixed-precision training method that significantly improves the
training throughput of a single GPU without losing accuracy. (2) We propose an
optimization approach for extremely large mini-batch size (up to 64k) that can
train CNN models on the ImageNet dataset without losing accuracy. (3) We
propose highly optimized all-reduce algorithms that achieve up to 3x and 11x
speedup on AlexNet and ResNet-50 respectively than NCCL-based training on a
cluster with 1024 Tesla P40 GPUs. On training ResNet-50 with 90 epochs, the
state-of-the-art GPU-based system with 1024 Tesla P100 GPUs spent 15 minutes
and achieved 74.9\% top-1 test accuracy, and another KNL-based system with 2048
Intel KNLs spent 20 minutes and achieved 75.4\% accuracy. Our training system
can achieve 75.8\% top-1 test accuracy in only 6.6 minutes using 2048 Tesla P40
GPUs. When training AlexNet with 95 epochs, our system can achieve 58.7\% top-1
test accuracy within 4 minutes, which also outperforms all other existing
systems.
|
Predicting Conversion of Mild Cognitive Impairments to Alzheimer's
Disease and Exploring Impact of Neuroimaging
|
Nowadays, a lot of scientific efforts are concentrated on the diagnosis of
Alzheimer's Disease (AD) applying deep learning methods to neuroimaging data.
Even for 2017, there were published more than a hundred papers dedicated to AD
diagnosis, whereas only a few works considered a problem of mild cognitive
impairments (MCI) conversion to the AD. However, the conversion prediction is
an important problem since approximately 15% of patients with MCI converges to
the AD every year. In the current work, we are focusing on the conversion
prediction using brain Magnetic Resonance Imaging and clinical data, such as
demographics, cognitive assessments, genetic, and biochemical markers. First of
all, we applied state-of-the-art deep learning algorithms on the neuroimaging
data and compared these results with two machine learning algorithms that we
fit using the clinical data. As a result, the models trained on the clinical
data outperform the deep learning algorithms applied to the MR images. To
explore the impact of neuroimaging further, we trained a deep feed-forward
embedding using similarity learning with Histogram loss on all available MRIs
and obtained 64-dimensional vector representation of neuroimaging data. The use
of learned representation from the deep embedding allowed to increase the
quality of prediction based on the neuroimaging. Finally, the current results
on this dataset show that the neuroimaging does affect conversion prediction,
however, cannot noticeably increase the quality of the prediction. The best
results of predicting MCI-to-AD conversion are provided by XGBoost algorithm
trained on the clinical and embedding data. The resulting accuracy is 0.76 +-
0.01 and the area under the ROC curve - 0.86 +- 0.01.
|
Semantic Labeling in Very High Resolution Images via a Self-Cascaded
Convolutional Neural Network
|
Semantic labeling for very high resolution (VHR) images in urban areas, is of
significant importance in a wide range of remote sensing applications. However,
many confusing manmade objects and intricate fine-structured objects make it
very difficult to obtain both coherent and accurate labeling results. For this
challenging task, we propose a novel deep model with convolutional neural
networks (CNNs), i.e., an end-to-end self-cascaded network (ScasNet).
Specifically, for confusing manmade objects, ScasNet improves the labeling
coherence with sequential global-to-local contexts aggregation. Technically,
multi-scale contexts are captured on the output of a CNN encoder, and then they
are successively aggregated in a self-cascaded manner. Meanwhile, for
fine-structured objects, ScasNet boosts the labeling accuracy with a
coarse-to-fine refinement strategy. It progressively refines the target objects
using the low-level features learned by CNN's shallow layers. In addition, to
correct the latent fitting residual caused by multi-feature fusion inside
ScasNet, a dedicated residual correction scheme is proposed. It greatly
improves the effectiveness of ScasNet. Extensive experimental results on three
public datasets, including two challenging benchmarks, show that ScasNet
achieves the state-of-the-art performance.
|
Uncertainty Quantification in CNN-Based Surface Prediction Using Shape
Priors
|
Surface reconstruction is a vital tool in a wide range of areas of medical
image analysis and clinical research. Despite the fact that many methods have
proposed solutions to the reconstruction problem, most, due to their
deterministic nature, do not directly address the issue of quantifying
uncertainty associated with their predictions. We remedy this by proposing a
novel probabilistic deep learning approach capable of simultaneous surface
reconstruction and associated uncertainty prediction. The method incorporates
prior shape information in the form of a principal component analysis (PCA)
model. Experiments using the UK Biobank data show that our probabilistic
approach outperforms an analogous deterministic PCA-based method in the task of
2D organ delineation and quantifies uncertainty by formulating distributions
over predicted surface vertex positions.
|
Kernel Density Estimation-Based Markov Models with Hidden State
|
We consider Markov models of stochastic processes where the next-step
conditional distribution is defined by a kernel density estimator (KDE),
similar to Markov forecast densities and certain time-series bootstrap schemes.
The KDE Markov models (KDE-MMs) we discuss are nonlinear, nonparametric, fully
probabilistic representations of stationary processes, based on techniques with
strong asymptotic consistency properties. The models generate new data by
concatenating points from the training data sequences in a context-sensitive
manner, together with some additive driving noise. We present novel EM-type
maximum-likelihood algorithms for data-driven bandwidth selection in KDE-MMs.
Additionally, we augment the KDE-MMs with a hidden state, yielding a new model
class, KDE-HMMs. The added state variable captures non-Markovian long memory
and signal structure (e.g., slow oscillations), complementing the short-range
dependences described by the Markov process. The resulting joint Markov and
hidden-Markov structure is appealing for modelling complex real-world processes
such as speech signals. We present guaranteed-ascent EM-update equations for
model parameters in the case of Gaussian kernels, as well as relaxed update
formulas that greatly accelerate training in practice. Experiments demonstrate
increased held-out set probability for KDE-HMMs on several challenging natural
and synthetic data series, compared to traditional techniques such as
autoregressive models, HMMs, and their combinations.
|
Dropout-GAN: Learning from a Dynamic Ensemble of Discriminators
|
We propose to incorporate adversarial dropout in generative multi-adversarial
networks, by omitting or dropping out, the feedback of each discriminator in
the framework with some probability at the end of each batch. Our approach
forces the single generator not to constrain its output to satisfy a single
discriminator, but, instead, to satisfy a dynamic ensemble of discriminators.
We show that this leads to a more generalized generator, promoting variety in
the generated samples and avoiding the common mode collapse problem commonly
experienced with generative adversarial networks (GANs). We further provide
evidence that the proposed framework, named Dropout-GAN, promotes sample
diversity both within and across epochs, eliminating mode collapse and
stabilizing training.
|
Weakly-Supervised Deep Learning of Heat Transport via Physics Informed
Loss
|
In typical machine learning tasks and applications, it is necessary to obtain
or create large labeled datasets in order to to achieve high performance.
Unfortunately, large labeled datasets are not always available and can be
expensive to source, creating a bottleneck towards more widely applicable
machine learning. The paradigm of weak supervision offers an alternative that
allows for integration of domain-specific knowledge by enforcing constraints
that a correct solution to the learning problem will obey over the output
space. In this work, we explore the application of this paradigm to 2-D
physical systems governed by non-linear differential equations. We demonstrate
that knowledge of the partial differential equations governing a system can be
encoded into the loss function of a neural network via an appropriately chosen
convolutional kernel. We demonstrate this by showing that the steady-state
solution to the 2-D heat equation can be learned directly from initial
conditions by a convolutional neural network, in the absence of labeled
training data. We also extend recent work in the progressive growing of fully
convolutional networks to achieve high accuracy (< 1.5% error) at multiple
scales of the heat-flow problem, including at the very large scale (1024x1024).
Finally, we demonstrate that this method can be used to speed up exact
calculation of the solution to the differential equations via finite
difference.
|
Making Classifier Chains Resilient to Class Imbalance
|
Class imbalance is an intrinsic characteristic of multi-label data. Most of
the labels in multi-label data sets are associated with a small number of
training examples, much smaller compared to the size of the data set. Class
imbalance poses a key challenge that plagues most multi-label learning methods.
Ensemble of Classifier Chains (ECC), one of the most prominent multi-label
learning methods, is no exception to this rule, as each of the binary models it
builds is trained from all positive and negative examples of a label. To make
ECC resilient to class imbalance, we first couple it with random undersampling.
We then present two extensions of this basic approach, where we build a varying
number of binary models per label and construct chains of different sizes, in
order to improve the exploitation of majority examples with approximately the
same computational budget. Experimental results on 16 multi-label datasets
demonstrate the effectiveness of the proposed approaches in a variety of
evaluation metrics.
|
Preference-based Online Learning with Dueling Bandits: A Survey
|
In machine learning, the notion of multi-armed bandits refers to a class of
online learning problems, in which an agent is supposed to simultaneously
explore and exploit a given set of choice alternatives in the course of a
sequential decision process. In the standard setting, the agent learns from
stochastic feedback in the form of real-valued rewards. In many applications,
however, numerical reward signals are not readily available -- instead, only
weaker information is provided, in particular relative preferences in the form
of qualitative comparisons between pairs of alternatives. This observation has
motivated the study of variants of the multi-armed bandit problem, in which
more general representations are used both for the type of feedback to learn
from and the target of prediction. The aim of this paper is to provide a survey
of the state of the art in this field, referred to as preference-based
multi-armed bandits or dueling bandits. To this end, we provide an overview of
problems that have been considered in the literature as well as methods for
tackling them. Our taxonomy is mainly based on the assumptions made by these
methods about the data-generating process and, related to this, the properties
of the preference-based feedback.
|
HybridNet: Classification and Reconstruction Cooperation for
Semi-Supervised Learning
|
In this paper, we introduce a new model for leveraging unlabeled data to
improve generalization performances of image classifiers: a two-branch
encoder-decoder architecture called HybridNet. The first branch receives
supervision signal and is dedicated to the extraction of invariant
class-related representations. The second branch is fully unsupervised and
dedicated to model information discarded by the first branch to reconstruct
input data. To further support the expected behavior of our model, we propose
an original training objective. It favors stability in the discriminative
branch and complementarity between the learned representations in the two
branches. HybridNet is able to outperform state-of-the-art results on CIFAR-10,
SVHN and STL-10 in various semi-supervised settings. In addition,
visualizations and ablation studies validate our contributions and the behavior
of the model on both CIFAR-10 and STL-10 datasets.
|
Local Linear Forests
|
Random forests are a powerful method for non-parametric regression, but are
limited in their ability to fit smooth signals, and can show poor predictive
performance in the presence of strong, smooth effects. Taking the perspective
of random forests as an adaptive kernel method, we pair the forest kernel with
a local linear regression adjustment to better capture smoothness. The
resulting procedure, local linear forests, enables us to improve on asymptotic
rates of convergence for random forests with smooth signals, and provides
substantial gains in accuracy on both real and simulated data. We prove a
central limit theorem valid under regularity conditions on the forest and
smoothness constraints, and propose a computationally efficient construction
for confidence intervals. Moving to a causal inference application, we discuss
the merits of local regression adjustments for heterogeneous treatment effect
estimation, and give an example on a dataset exploring the effect word choice
has on attitudes to the social safety net. Last, we include simulation results
on real and generated data.
|
Faster Convergence & Generalization in DNNs
|
Deep neural networks have gained tremendous popularity in last few years.
They have been applied for the task of classification in almost every domain.
Despite the success, deep networks can be incredibly slow to train for even
moderate sized models on sufficiently large datasets. Additionally, these
networks require large amounts of data to be able to generalize. The importance
of speeding up convergence, and generalization in deep networks can not be
overstated. In this work, we develop an optimization algorithm based on
generalized-optimal updates derived from minibatches that lead to faster
convergence. Towards the end, we demonstrate on two benchmark datasets that the
proposed method achieves two orders of magnitude speed up over traditional
back-propagation, and is more robust to noise/over-fitting.
|
High-dimensional estimation via sum-of-squares proofs
|
Estimation is the computational task of recovering a hidden parameter $x$
associated with a distribution $D_x$, given a measurement $y$ sampled from the
distribution. High dimensional estimation problems arise naturally in
statistics, machine learning, and complexity theory.
Many high dimensional estimation problems can be formulated as systems of
polynomial equations and inequalities, and thus give rise to natural
probability distributions over polynomial systems. Sum-of-squares proofs
provide a powerful framework to reason about polynomial systems, and further
there exist efficient algorithms to search for low-degree sum-of-squares
proofs.
Understanding and characterizing the power of sum-of-squares proofs for
estimation problems has been a subject of intense study in recent years. On one
hand, there is a growing body of work utilizing sum-of-squares proofs for
recovering solutions to polynomial systems when the system is feasible. On the
other hand, a general technique referred to as pseudocalibration has been
developed towards showing lower bounds on the degree of sum-of-squares proofs.
Finally, the existence of sum-of-squares refutations of a polynomial system has
been shown to be intimately connected to the existence of spectral algorithms.
In this article we survey these developments.
|
Kalman Filter-based Heuristic Ensemble (KFHE): A new perspective on
multi-class ensemble classification using Kalman filters
|
This paper introduces a new perspective on multi-class ensemble
classification that considers training an ensemble as a state estimation
problem. The new perspective considers the final ensemble classifier model as a
static state, which can be estimated using a Kalman filter that combines noisy
estimates made by individual classifier models. A new algorithm based on this
perspective, the Kalman Filter-based Heuristic Ensemble (KFHE), is also
presented in this paper which shows the practical applicability of the new
perspective. Experiments performed on 30 datasets compare KFHE with
state-of-the-art multi-class ensemble classification algorithms and show the
potential and effectiveness of the new perspective and algorithm. Existing
ensemble approaches trade off classification accuracy against robustness to
class label noise, but KFHE is shown to be significantly better or at least as
good as the state-of-the-art algorithms for datasets both with and without
class label noise.
|
Leveraging Motion Priors in Videos for Improving Human Segmentation
|
Despite many advances in deep-learning based semantic segmentation,
performance drop due to distribution mismatch is often encountered in the real
world. Recently, a few domain adaptation and active learning approaches have
been proposed to mitigate the performance drop. However, very little attention
has been made toward leveraging information in videos which are naturally
captured in most camera systems. In this work, we propose to leverage "motion
prior" in videos for improving human segmentation in a weakly-supervised active
learning setting. By extracting motion information using optical flow in
videos, we can extract candidate foreground motion segments (referred to as
motion prior) potentially corresponding to human segments. We propose to learn
a memory-network-based policy model to select strong candidate segments
(referred to as strong motion prior) through reinforcement learning. The
selected segments have high precision and are directly used to finetune the
model. In a newly collected surveillance camera dataset and a publicly
available UrbanStreet dataset, our proposed method improves the performance of
human segmentation across multiple scenes and modalities (i.e., RGB to Infrared
(IR)). Last but not least, our method is empirically complementary to existing
domain adaptation approaches such that additional performance gain is achieved
by combining our weakly-supervised active learning approach with domain
adaptation approaches.
|
Comparator Networks
|
The objective of this work is set-based verification, e.g. to decide if two
sets of images of a face are of the same person or not. The traditional
approach to this problem is to learn to generate a feature vector per image,
aggregate them into one vector to represent the set, and then compute the
cosine similarity between sets. Instead, we design a neural network
architecture that can directly learn set-wise verification. Our contributions
are: (i) We propose a Deep Comparator Network (DCN) that can ingest a pair of
sets (each may contain a variable number of images) as inputs, and compute a
similarity between the pair--this involves attending to multiple discriminative
local regions (landmarks), and comparing local descriptors between pairs of
faces; (ii) To encourage high-quality representations for each set, internal
competition is introduced for recalibration based on the landmark score; (iii)
Inspired by image retrieval, a novel hard sample mining regime is proposed to
control the sampling process, such that the DCN is complementary to the
standard image classification models. Evaluations on the IARPA Janus face
recognition benchmarks show that the comparator networks outperform the
previous state-of-the-art results by a large margin.
|
Improving Transferability of Deep Neural Networks
|
Learning from small amounts of labeled data is a challenge in the area of
deep learning. This is currently addressed by Transfer Learning where one
learns the small data set as a transfer task from a larger source dataset.
Transfer Learning can deliver higher accuracy if the hyperparameters and source
dataset are chosen well. One of the important parameters is the learning rate
for the layers of the neural network. We show through experiments on the
ImageNet22k and Oxford Flowers datasets that improvements in accuracy in range
of 127% can be obtained by proper choice of learning rates. We also show that
the images/label parameter for a dataset can potentially be used to determine
optimal learning rates for the layers to get the best overall accuracy. We
additionally validate this method on a sample of real-world image
classification tasks from a public visual recognition API.
|
Deep Encoder-Decoder Models for Unsupervised Learning of Controllable
Speech Synthesis
|
Generating versatile and appropriate synthetic speech requires control over
the output expression separate from the spoken text. Important non-textual
speech variation is seldom annotated, in which case output control must be
learned in an unsupervised fashion. In this paper, we perform an in-depth study
of methods for unsupervised learning of control in statistical speech
synthesis. For example, we show that popular unsupervised training heuristics
can be interpreted as variational inference in certain autoencoder models. We
additionally connect these models to VQ-VAEs, another, recently-proposed class
of deep variational autoencoders, which we show can be derived from a very
similar mathematical argument. The implications of these new probabilistic
interpretations are discussed. We illustrate the utility of the various
approaches with an application to acoustic modelling for emotional speech
synthesis, where the unsupervised methods for learning expression control
(without access to emotional labels) are found to give results that in many
aspects match or surpass the previous best supervised approach.
|
Estimating Failure in Brittle Materials using Graph Theory
|
In brittle fracture applications, failure paths, regions where the failure
occurs and damage statistics, are some of the key quantities of interest (QoI).
High-fidelity models for brittle failure that accurately predict these QoI
exist but are highly computationally intensive, making them infeasible to
incorporate in upscaling and uncertainty quantification frameworks. The goal of
this paper is to provide a fast heuristic to reasonably estimate quantities
such as failure path and damage in the process of brittle failure. Towards this
goal, we first present a method to predict failure paths under tensile loading
conditions and low-strain rates. The method uses a $k$-nearest neighbors
algorithm built on fracture process zone theory, and identifies the set of all
possible pre-existing cracks that are likely to join early to form a large
crack. The method then identifies zone of failure and failure paths using
weighted graphs algorithms. We compare these failure paths to those computed
with a high-fidelity model called the Hybrid Optimization Software Simulation
Suite (HOSS). A probabilistic evolution model for average damage in a system is
also developed that is trained using 150 HOSS simulations and tested on 40
simulations. A non-parametric approach based on confidence intervals is used to
determine the damage evolution over time along the dominant failure path. For
upscaling, damage is the key QoI needed as an input by the continuum models.
This needs to be informed accurately by the surrogate models for calculating
effective modulii at continuum-scale. We show that for the proposed average
damage evolution model, the prediction accuracy on the test data is more than
90\%. In terms of the computational time, the proposed models are $\approx
\mathcal{O}(10^6)$ times faster compared to high-fidelity HOSS.
|
Call Detail Records Driven Anomaly Detection and Traffic Prediction in
Mobile Cellular Networks
|
Mobile networks possess information about the users as well as the network.
Such information is useful for making the network end-to-end visible and
intelligent. Big data analytics can efficiently analyze user and network
information, unearth meaningful insights with the help of machine learning
tools. Utilizing big data analytics and machine learning, this work contributes
in three ways. First, we utilize the call detail records (CDR) data to detect
anomalies in the network. For authentication and verification of anomalies, we
use k-means clustering, an unsupervised machine learning algorithm. Through
effective detection of anomalies, we can proceed to suitable design for
resource distribution as well as fault detection and avoidance. Second, we
prepare anomaly-free data by removing anomalous activities and train a neural
network model. By passing anomaly and anomaly-free data through this model, we
observe the effect of anomalous activities in training of the model and also
observe mean square error of anomaly and anomaly free data. Lastly, we use an
autoregressive integrated moving average (ARIMA) model to predict future
traffic for a user. Through simple visualization, we show that anomaly free
data better generalizes the learning models and performs better on prediction
task.
|
Efficient Gauss-Newton-Krylov momentum conservation constrained
PDE-LDDMM using the band-limited vector field parameterization
|
The class of non-rigid registration methods proposed in the framework of
PDE-constrained Large Deformation Diffeomorphic Metric Mapping is a
particularly interesting family of physically meaningful diffeomorphic
registration methods. PDE-constrained LDDMM methods are formulated as
constrained variational problems, where the different physical models are
imposed using the associated partial differential equations as hard
constraints. Inexact Newton-Krylov optimization has shown an excellent
numerical accuracy and an extraordinarily fast convergence rate in this
framework. However, the Galerkin representation of the non-stationary velocity
fields does not provide proper geodesic paths. In a previous work, we proposed
a method for PDE-constrained LDDMM parameterized in the space of initial
velocity fields under the EPDiff equation. The proposed method provided
geodesics in the framework of PDE-constrained LDDMM, and it showed performance
competitive to benchmark PDE-constrained LDDMM and EPDiff-LDDMM methods.
However, the major drawback of this method was the large memory load inherent
to PDE-constrained LDDMM methods and the increased computational time with
respect to the benchmark methods. In this work we optimize the computational
complexity of the method using the band-limited vector field parameterization
closing the loop with our previous works.
|
State-of-the-art and gaps for deep learning on limited training data in
remote sensing
|
Deep learning usually requires big data, with respect to both volume and
variety. However, most remote sensing applications only have limited training
data, of which a small subset is labeled. Herein, we review three
state-of-the-art approaches in deep learning to combat this challenge. The
first topic is transfer learning, in which some aspects of one domain, e.g.,
features, are transferred to another domain. The next is unsupervised learning,
e.g., autoencoders, which operate on unlabeled data. The last is generative
adversarial networks, which can generate realistic looking data that can fool
the likes of both a deep learning network and human. The aim of this article is
to raise awareness of this dilemma, to direct the reader to existing work and
to highlight current gaps that need solving.
|
A Hierarchical Approach to Neural Context-Aware Modeling
|
We present a new recurrent neural network topology to enhance
state-of-the-art machine learning systems by incorporating a broader context.
Our approach overcomes recent limitations with extended narratives through a
multi-layered computational approach to generate an abstract context
representation. Therefore, the developed system captures the narrative on
word-level, sentence-level, and context-level. Through the hierarchical set-up,
our proposed model summarizes the most salient information on each level and
creates an abstract representation of the extended context. We subsequently use
this representation to enhance neural language processing systems on the task
of semantic error detection. To show the potential of the newly introduced
topology, we compare the approach against a context-agnostic set-up including a
standard neural language model and a supervised binary classification network.
The performance measures on the error detection task show the advantage of the
hierarchical context-aware topologies, improving the baseline by 12.75%
relative for unsupervised models and 20.37% relative for supervised models.
|
K-medoids Clustering of Data Sequences with Composite Distributions
|
This paper studies clustering of data sequences using the k-medoids
algorithm. All the data sequences are assumed to be generated from
\emph{unknown} continuous distributions, which form clusters with each cluster
containing a composite set of closely located distributions (based on a certain
distance metric between distributions). The maximum intra-cluster distance is
assumed to be smaller than the minimum inter-cluster distance, and both values
are assumed to be known. The goal is to group the data sequences together if
their underlying generative distributions (which are unknown) belong to one
cluster. Distribution distance metrics based k-medoids algorithms are proposed
for known and unknown number of distribution clusters. Upper bounds on the
error probability and convergence results in the large sample regime are also
provided. It is shown that the error probability decays exponentially fast as
the number of samples in each data sequence goes to infinity. The error
exponent has a simple form regardless of the distance metric applied when
certain conditions are satisfied. In particular, the error exponent is
characterized when either the Kolmogrov-Smirnov distance or the maximum mean
discrepancy are used as the distance metric. Simulation results are provided to
validate the analysis.
|
Count-Based Exploration with the Successor Representation
|
In this paper we introduce a simple approach for exploration in reinforcement
learning (RL) that allows us to develop theoretically justified algorithms in
the tabular case but that is also extendable to settings where function
approximation is required. Our approach is based on the successor
representation (SR), which was originally introduced as a representation
defining state generalization by the similarity of successor states. Here we
show that the norm of the SR, while it is being learned, can be used as a
reward bonus to incentivize exploration. In order to better understand this
transient behavior of the norm of the SR we introduce the substochastic
successor representation (SSR) and we show that it implicitly counts the number
of times each state (or feature) has been observed. We use this result to
introduce an algorithm that performs as well as some theoretically
sample-efficient approaches. Finally, we extend these ideas to a deep RL
algorithm and show that it achieves state-of-the-art performance in Atari 2600
games when in a low sample-complexity regime.
|
MnasNet: Platform-Aware Neural Architecture Search for Mobile
|
Designing convolutional neural networks (CNN) for mobile devices is
challenging because mobile models need to be small and fast, yet still
accurate. Although significant efforts have been dedicated to design and
improve mobile CNNs on all dimensions, it is very difficult to manually balance
these trade-offs when there are so many architectural possibilities to
consider. In this paper, we propose an automated mobile neural architecture
search (MNAS) approach, which explicitly incorporate model latency into the
main objective so that the search can identify a model that achieves a good
trade-off between accuracy and latency. Unlike previous work, where latency is
considered via another, often inaccurate proxy (e.g., FLOPS), our approach
directly measures real-world inference latency by executing the model on mobile
phones. To further strike the right balance between flexibility and search
space size, we propose a novel factorized hierarchical search space that
encourages layer diversity throughout the network. Experimental results show
that our approach consistently outperforms state-of-the-art mobile CNN models
across multiple vision tasks. On the ImageNet classification task, our MnasNet
achieves 75.2% top-1 accuracy with 78ms latency on a Pixel phone, which is 1.8x
faster than MobileNetV2 [29] with 0.5% higher accuracy and 2.3x faster than
NASNet [36] with 1.2% higher accuracy. Our MnasNet also achieves better mAP
quality than MobileNets for COCO object detection. Code is at
https://github.com/tensorflow/tpu/tree/master/models/official/mnasnet
|
Composable Core-sets for Determinant Maximization Problems via Spectral
Spanners
|
We study a spectral generalization of classical combinatorial graph spanners
to the spectral setting. Given a set of vectors $V\subseteq \Re^d$, we say a
set $U\subseteq V$ is an $\alpha$-spectral spanner if for all $v\in V$ there is
a probability distribution $\mu_v$ supported on $U$ such that $$vv^\intercal
\preceq \alpha\cdot\mathbb{E}_{u\sim\mu_v} uu^\intercal.$$ We show that any set
$V$ has an $\tilde{O}(d)$-spectral spanner of size $\tilde{O}(d)$ and this
bound is almost optimal in the worst case.
We use spectral spanners to study composable core-sets for spectral problems.
We show that for many objective functions one can use a spectral spanner,
independent of the underlying functions, as a core-set and obtain almost
optimal composable core-sets. For example, for the determinant maximization
problem we obtain an $\tilde{O}(k)^k$-composable core-set and we show that this
is almost optimal in the worst case.
Our algorithm is a spectral analogue of the classical greedy algorithm for
finding (combinatorial) spanners in graphs. We expect that our spanners find
many other applications in distributed or parallel models of computation. Our
proof is spectral. As a side result of our techniques, we show that the rank of
diagonally dominant lower-triangular matrices are robust under `small
perturbations' which could be of independent interests.
|
Security and Privacy Issues in Deep Learning
|
To promote secure and private artificial intelligence (SPAI), we review
studies on the model security and data privacy of DNNs. Model security allows
system to behave as intended without being affected by malicious external
influences that can compromise its integrity and efficiency. Security attacks
can be divided based on when they occur: if an attack occurs during training,
it is known as a poisoning attack, and if it occurs during inference (after
training) it is termed an evasion attack. Poisoning attacks compromise the
training process by corrupting the data with malicious examples, while evasion
attacks use adversarial examples to disrupt entire classification process.
Defenses proposed against such attacks include techniques to recognize and
remove malicious data, train a model to be insensitive to such data, and mask
the model's structure and parameters to render attacks more challenging to
implement. Furthermore, the privacy of the data involved in model training is
also threatened by attacks such as the model-inversion attack, or by dishonest
service providers of AI applications. To maintain data privacy, several
solutions that combine existing data-privacy techniques have been proposed,
including differential privacy and modern cryptography techniques. In this
paper, we describe the notions of some of methods, e.g., homomorphic
encryption, and review their advantages and challenges when implemented in
deep-learning models.
|
Deep Belief Networks Based Feature Generation and Regression for
Predicting Wind Power
|
Wind energy forecasting helps to manage power production, and hence, reduces
energy cost. Deep Neural Networks (DNN) mimics hierarchical learning in the
human brain and thus possesses hierarchical, distributed, and multi-task
learning capabilities. Based on aforementioned characteristics, we report Deep
Belief Network (DBN) based forecast engine for wind power prediction because of
its good generalization and unsupervised pre-training attributes. The proposed
DBN-WP forecast engine, which exhibits stochastic feature generation
capabilities and is composed of multiple Restricted Boltzmann Machines,
generates suitable features for wind power prediction using atmospheric
properties as input. DBN-WP, due to its unsupervised pre-training of RBM layers
and generalization capabilities, is able to learn the fluctuations in the
meteorological properties and thus is able to perform effective mapping of the
wind power. In the deep network, a regression layer is appended at the end to
predict sort-term wind power. It is experimentally shown that the deep learning
and unsupervised pre-training capabilities of DBN based model has comparable
and in some cases better results than hybrid and complex learning techniques
proposed for wind power prediction. The proposed prediction system based on
DBN, achieves mean values of RMSE, MAE and SDE as 0.124, 0.083 and 0.122,
respectively. Statistical analysis of several independent executions of the
proposed DBN-WP wind power prediction system demonstrates the stability of the
system. The proposed DBN-WP architecture is easy to implement and offers
generalization as regards the change in location of the wind farm is concerned.
|
Spectrum concentration in deep residual learning: a free probability
approach
|
We revisit the initialization of deep residual networks (ResNets) by
introducing a novel analytical tool in free probability to the community of
deep learning. This tool deals with non-Hermitian random matrices, rather than
their conventional Hermitian counterparts in the literature. As a consequence,
this new tool enables us to evaluate the singular value spectrum of the
input-output Jacobian of a fully-connected deep ResNet for both linear and
nonlinear cases. With the powerful tool of free probability, we conduct an
asymptotic analysis of the spectrum on the single-layer case, and then extend
this analysis to the multi-layer case of an arbitrary number of layers. In
particular, we propose to rescale the classical random initialization by the
number of residual units, so that the spectrum has the order of $O(1)$, when
compared with the large width and depth of the network. We empirically
demonstrate that the proposed initialization scheme learns at a speed of orders
of magnitudes faster than the classical ones, and thus attests a strong
practical relevance of this investigation.
|
Multimodal Deep Domain Adaptation
|
Typically a classifier trained on a given dataset (source domain) does not
performs well if it is tested on data acquired in a different setting (target
domain). This is the problem that domain adaptation (DA) tries to overcome and,
while it is a well explored topic in computer vision, it is largely ignored in
robotic vision where usually visual classification methods are trained and
tested in the same domain. Robots should be able to deal with unknown
environments, recognize objects and use them in the correct way, so it is
important to explore the domain adaptation scenario also in this context. The
goal of the project is to define a benchmark and a protocol for multi-modal
domain adaptation that is valuable for the robot vision community. With this
purpose some of the state-of-the-art DA methods are selected: Deep Adaptation
Network (DAN), Domain Adversarial Training of Neural Network (DANN), Automatic
Domain Alignment Layers (AutoDIAL) and Adversarial Discriminative Domain
Adaptation (ADDA). Evaluations have been done using different data types: RGB
only, depth only and RGB-D over the following datasets, designed for the
robotic community: RGB-D Object Dataset (ROD), Web Object Dataset (WOD),
Autonomous Robot Indoor Dataset (ARID), Big Berkeley Instance Recognition
Dataset (BigBIRD) and Active Vision Dataset. Although progresses have been made
on the formulation of effective adaptation algorithms and more realistic object
datasets are available, the results obtained show that, training a sufficiently
good object classifier, especially in the domain adaptation scenario, is still
an unsolved problem. Also the best way to combine depth with RGB informations
to improve the performance is a point that needs to be investigated more.
|
Rank and Rate: Multi-task Learning for Recommender Systems
|
The two main tasks in the Recommender Systems domain are the ranking and
rating prediction tasks. The rating prediction task aims at predicting to what
extent a user would like any given item, which would enable to recommend the
items with the highest predicted scores. The ranking task on the other hand
directly aims at recommending the most valuable items for the user. Several
previous approaches proposed learning user and item representations to optimize
both tasks simultaneously in a multi-task framework. In this work we propose a
novel multi-task framework that exploits the fact that a user does a two-phase
decision process - first decides to interact with an item (ranking task) and
only afterward to rate it (rating prediction task). We evaluated our framework
on two benchmark datasets, on two different configurations and showed its
superiority over state-of-the-art methods.
|
Feature Grouping as a Stochastic Regularizer for High-Dimensional
Structured Data
|
In many applications where collecting data is expensive, for example
neuroscience or medical imaging, the sample size is typically small compared to
the feature dimension. It is challenging in this setting to train expressive,
non-linear models without overfitting. These datasets call for intelligent
regularization that exploits known structure, such as correlations between the
features arising from the measurement device. However, existing structured
regularizers need specially crafted solvers, which are difficult to apply to
complex models. We propose a new regularizer specifically designed to leverage
structure in the data in a way that can be applied efficiently to complex
models. Our approach relies on feature grouping, using a fast clustering
algorithm inside a stochastic gradient descent loop: given a family of feature
groupings that capture feature covariations, we randomly select these groups at
each iteration. We show that this approach amounts to enforcing a denoising
regularizer on the solution. The method is easy to implement in many model
architectures, such as fully connected neural networks, and has a linear
computational cost. We apply this regularizer to a real-world fMRI dataset and
the Olivetti Faces datasets. Experiments on both datasets demonstrate that the
proposed approach produces models that generalize better than those trained
with conventional regularizers, and also improves convergence speed.
|
Multi-Speaker DOA Estimation Using Deep Convolutional Networks Trained
with Noise Signals
|
Supervised learning based methods for source localization, being data driven,
can be adapted to different acoustic conditions via training and have been
shown to be robust to adverse acoustic environments. In this paper, a
convolutional neural network (CNN) based supervised learning method for
estimating the direction-of-arrival (DOA) of multiple speakers is proposed.
Multi-speaker DOA estimation is formulated as a multi-class multi-label
classification problem, where the assignment of each DOA label to the input
feature is treated as a separate binary classification problem. The phase
component of the short-time Fourier transform (STFT) coefficients of the
received microphone signals are directly fed into the CNN, and the features for
DOA estimation are learnt during training. Utilizing the assumption of disjoint
speaker activity in the STFT domain, a novel method is proposed to train the
CNN with synthesized noise signals. Through experimental evaluation with both
simulated and measured acoustic impulse responses, the ability of the proposed
DOA estimation approach to adapt to unseen acoustic conditions and its
robustness to unseen noise type is demonstrated. Through additional empirical
investigation, it is also shown that with an array of M microphones our
proposed framework yields the best localization performance with M-1
convolution layers. The ability of the proposed method to accurately localize
speakers in a dynamic acoustic scenario with varying number of sources is also
shown.
|
Practical Constrained Optimization of Auction Mechanisms in E-Commerce
Sponsored Search Advertising
|
Sponsored search in E-commerce platforms such as Amazon, Taobao and Tmall
provides sellers an effective way to reach potential buyers with most relevant
purpose. In this paper, we study the auction mechanism optimization problem in
sponsored search on Alibaba's mobile E-commerce platform. Besides generating
revenue, we are supposed to maintain an efficient marketplace with plenty of
quality users, guarantee a reasonable return on investment (ROI) for
advertisers, and meanwhile, facilitate a pleasant shopping experience for the
users. These requirements essentially pose a constrained optimization problem.
Directly optimizing over auction parameters yields a discontinuous, non-convex
problem that denies effective solutions. One of our major contribution is a
practical convex optimization formulation of the original problem. We devise a
novel re-parametrization of auction mechanism with discrete sets of
representative instances. To construct the optimization problem, we build an
auction simulation system which estimates the resulted business indicators of
the selected parameters by replaying the auctions recorded from real online
requests. We summarized the experiments on real search traffics to analyze the
effects of fidelity of auction simulation, the efficacy under various
constraint targets and the influence of regularization. The experiment results
show that with proper entropy regularization, we are able to maximize revenue
while constraining other business indicators within given ranges.
|
Disaster Monitoring using Unmanned Aerial Vehicles and Deep Learning
|
Monitoring of disasters is crucial for mitigating their effects on the
environment and human population, and can be facilitated by the use of unmanned
aerial vehicles (UAV), equipped with camera sensors that produce aerial photos
of the areas of interest. A modern technique for recognition of events based on
aerial photos is deep learning. In this paper, we present the state of the art
work related to the use of deep learning techniques for disaster
identification. We demonstrate the potential of this technique in identifying
disasters with high accuracy, by means of a relatively simple deep learning
model. Based on a dataset of 544 images (containing disaster images such as
fires, earthquakes, collapsed buildings, tsunami and flooding, as well as
non-disaster scenes), our results show an accuracy of 91% achieved, indicating
that deep learning, combined with UAV equipped with camera sensors, have the
potential to predict disasters with high accuracy.
|
Deep learning in agriculture: A survey
|
Deep learning constitutes a recent, modern technique for image processing and
data analysis, with promising results and large potential. As deep learning has
been successfully applied in various domains, it has recently entered also the
domain of agriculture. In this paper, we perform a survey of 40 research
efforts that employ deep learning techniques, applied to various agricultural
and food production challenges. We examine the particular agricultural problems
under study, the specific models and frameworks employed, the sources, nature
and pre-processing of data used, and the overall performance achieved according
to the metrics used at each work under study. Moreover, we study comparisons of
deep learning with other existing popular techniques, in respect to differences
in classification or regression performance. Our findings indicate that deep
learning provides high accuracy, outperforming existing commonly used image
processing techniques.
|
t-SNE-CUDA: GPU-Accelerated t-SNE and its Applications to Modern Data
|
Modern datasets and models are notoriously difficult to explore and analyze
due to their inherent high dimensionality and massive numbers of samples.
Existing visualization methods which employ dimensionality reduction to two or
three dimensions are often inefficient and/or ineffective for these datasets.
This paper introduces t-SNE-CUDA, a GPU-accelerated implementation of
t-distributed Symmetric Neighbor Embedding (t-SNE) for visualizing datasets and
models. t-SNE-CUDA significantly outperforms current implementations with
50-700x speedups on the CIFAR-10 and MNIST datasets. These speedups enable, for
the first time, visualization of the neural network activations on the entire
ImageNet dataset - a feat that was previously computationally intractable. We
also demonstrate visualization performance in the NLP domain by visualizing the
GloVe embedding vectors. From these visualizations, we can draw interesting
conclusions about using the L2 metric in these embedding spaces. t-SNE-CUDA is
publicly available athttps://github.com/CannyLab/tsne-cuda
|
Inferring the ground truth through crowdsourcing
|
Universally valid ground truth is almost impossible to obtain or would come
at a very high cost. For supervised learning without universally valid ground
truth, a recommended approach is applying crowdsourcing: Gathering a large data
set annotated by multiple individuals of varying possibly expertise levels and
inferring the ground truth data to be used as labels to train the classifier.
Nevertheless, due to the sensitivity of the problem at hand (e.g. mitosis
detection in breast cancer histology images), the obtained data needs
verification and proper assessment before being used for classifier training.
Even in the context of organic computing systems, an indisputable ground truth
might not always exist. Therefore, it should be inferred through the
aggregation and verification of the local knowledge of each autonomous agent.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.