title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Obfuscated Gradients Give a False Sense of Security: Circumventing
Defenses to Adversarial Examples | We identify obfuscated gradients, a kind of gradient masking, as a phenomenon
that leads to a false sense of security in defenses against adversarial
examples. While defenses that cause obfuscated gradients appear to defeat
iterative optimization-based attacks, we find defenses relying on this effect
can be circumvented. We describe characteristic behaviors of defenses
exhibiting the effect, and for each of the three types of obfuscated gradients
we discover, we develop attack techniques to overcome it. In a case study,
examining non-certified white-box-secure defenses at ICLR 2018, we find
obfuscated gradients are a common occurrence, with 7 of 9 defenses relying on
obfuscated gradients. Our new attacks successfully circumvent 6 completely, and
1 partially, in the original threat model each paper considers.
|
Nearly Optimal Dynamic $k$-Means Clustering for High-Dimensional Data | We consider the $k$-means clustering problem in the dynamic streaming
setting, where points from a discrete Euclidean space $\{1, 2, \ldots,
\Delta\}^d$ can be dynamically inserted to or deleted from the dataset. For
this problem, we provide a one-pass coreset construction algorithm using space
$\tilde{O}(k\cdot \mathrm{poly}(d, \log\Delta))$, where $k$ is the target
number of centers. To our knowledge, this is the first dynamic geometric data
stream algorithm for $k$-means using space polynomial in dimension and nearly
optimal (linear) in $k$.
|
Analysis of Fast Alternating Minimization for Structured Dictionary
Learning | Methods exploiting sparsity have been popular in imaging and signal
processing applications including compression, denoising, and imaging inverse
problems. Data-driven approaches such as dictionary learning and transform
learning enable one to discover complex image features from datasets and
provide promising performance over analytical models. Alternating minimization
algorithms have been particularly popular in dictionary or transform learning.
In this work, we study the properties of alternating minimization for
structured (unitary) sparsifying operator learning. While the algorithm
converges to the stationary points of the non-convex problem in general, we
prove rapid local linear convergence to the underlying generative model under
mild assumptions. Our experiments show that the unitary operator learning
algorithm is robust to initialization.
|
Scalable L\'evy Process Priors for Spectral Kernel Learning | Gaussian processes are rich distributions over functions, with generalization
properties determined by a kernel function. When used for long-range
extrapolation, predictions are particularly sensitive to the choice of kernel
parameters. It is therefore critical to account for kernel uncertainty in our
predictive distributions. We propose a distribution over kernels formed by
modelling a spectral mixture density with a L\'evy process. The resulting
distribution has support for all stationary covariances--including the popular
RBF, periodic, and Mat\'ern kernels--combined with inductive biases which
enable automatic and data efficient learning, long-range extrapolation, and
state of the art predictive performance. The proposed model also presents an
approach to spectral regularization, as the L\'evy process introduces a
sparsity-inducing prior over mixture components, allowing automatic selection
over model order and pruning of extraneous components. We exploit the algebraic
structure of the proposed process for $\mathcal{O}(n)$ training and
$\mathcal{O}(1)$ predictions. We perform extrapolations having reasonable
uncertainty estimates on several benchmarks, show that the proposed model can
recover flexible ground truth covariances and that it is robust to errors in
initialization.
|
Causal Learning and Explanation of Deep Neural Networks via Autoencoded
Activations | Deep neural networks are complex and opaque. As they enter application in a
variety of important and safety critical domains, users seek methods to explain
their output predictions. We develop an approach to explaining deep neural
networks by constructing causal models on salient concepts contained in a CNN.
We develop methods to extract salient concepts throughout a target network by
using autoencoders trained to extract human-understandable representations of
network activations. We then build a bayesian causal model using these
extracted concepts as variables in order to explain image classification.
Finally, we use this causal model to identify and visualize features with
significant causal influence on final classification.
|
Modeling polypharmacy side effects with graph convolutional networks | The use of drug combinations, termed polypharmacy, is common to treat
patients with complex diseases and co-existing conditions. However, a major
consequence of polypharmacy is a much higher risk of adverse side effects for
the patient. Polypharmacy side effects emerge because of drug-drug
interactions, in which activity of one drug may change if taken with another
drug. The knowledge of drug interactions is limited because these complex
relationships are rare, and are usually not observed in relatively small
clinical testing. Discovering polypharmacy side effects thus remains an
important challenge with significant implications for patient mortality. Here,
we present Decagon, an approach for modeling polypharmacy side effects. The
approach constructs a multimodal graph of protein-protein interactions,
drug-protein target interactions, and the polypharmacy side effects, which are
represented as drug-drug interactions, where each side effect is an edge of a
different type. Decagon is developed specifically to handle such multimodal
graphs with a large number of edge types. Our approach develops a new graph
convolutional neural network for multirelational link prediction in multimodal
networks. Decagon predicts the exact side effect, if any, through which a given
drug combination manifests clinically. Decagon accurately predicts polypharmacy
side effects, outperforming baselines by up to 69%. We find that it
automatically learns representations of side effects indicative of
co-occurrence of polypharmacy in patients. Furthermore, Decagon models
particularly well side effects with a strong molecular basis, while on
predominantly non-molecular side effects, it achieves good performance because
of effective sharing of model parameters across edge types. Decagon creates
opportunities to use large pharmacogenomic and patient data to flag and
prioritize side effects for follow-up analysis.
|
Interpretable Deep Convolutional Neural Networks via Meta-learning | Model interpretability is a requirement in many applications in which crucial
decisions are made by users relying on a model's outputs. The recent movement
for "algorithmic fairness" also stipulates explainability, and therefore
interpretability of learning models. And yet the most successful contemporary
Machine Learning approaches, the Deep Neural Networks, produce models that are
highly non-interpretable. We attempt to address this challenge by proposing a
technique called CNN-INTE to interpret deep Convolutional Neural Networks (CNN)
via meta-learning. In this work, we interpret a specific hidden layer of the
deep CNN model on the MNIST image dataset. We use a clustering algorithm in a
two-level structure to find the meta-level training data and Random Forest as
base learning algorithms to generate the meta-level test data. The
interpretation results are displayed visually via diagrams, which clearly
indicates how a specific test instance is classified. Our method achieves
global interpretation for all the test instances without sacrificing the
accuracy obtained by the original deep CNN model. This means our model is
faithful to the deep CNN model, which leads to reliable interpretations.
|
Dual Memory Neural Computer for Asynchronous Two-view Sequential
Learning | One of the core tasks in multi-view learning is to capture relations among
views. For sequential data, the relations not only span across views, but also
extend throughout the view length to form long-term intra-view and inter-view
interactions. In this paper, we present a new memory augmented neural network
model that aims to model these complex interactions between two asynchronous
sequential views. Our model uses two encoders for reading from and writing to
two external memories for encoding input views. The intra-view interactions and
the long-term dependencies are captured by the use of memories during this
encoding process. There are two modes of memory accessing in our system:
late-fusion and early-fusion, corresponding to late and early inter-view
interactions. In the late-fusion mode, the two memories are separated,
containing only view-specific contents. In the early-fusion mode, the two
memories share the same addressing space, allowing cross-memory accessing. In
both cases, the knowledge from the memories will be combined by a decoder to
make predictions over the output space. The resulting dual memory neural
computer is demonstrated on a comprehensive set of experiments, including a
synthetic task of summing two sequences and the tasks of drug prescription and
disease progression in healthcare. The results demonstrate competitive
performance over both traditional algorithms and deep learning methods designed
for multi-view problems.
|
Representation Learning for Resource Usage Prediction | Creating a model of a computer system that can be used for tasks such as
predicting future resource usage and detecting anomalies is a challenging
problem. Most current systems rely on heuristics and overly simplistic
assumptions about the workloads and system statistics. These heuristics are
typically a one-size-fits-all solution so as to be applicable in a wide range
of applications and systems environments.
With this paper, we present our ongoing work of integrating systems telemetry
ranging from standard resource usage statistics to kernel and library calls of
applications into a machine learning model. Intuitively, such a ML model
approximates, at any point in time, the state of a system and allows us to
solve tasks such as resource usage prediction and anomaly detection. To achieve
this goal, we leverage readily-available information that does not require any
changes to the applications run on the system. We train recurrent neural
networks to learn a model of the system under consideration. As a proof of
concept, we train models specifically to predict future resource usage of
running applications.
|
A Generative Model for Natural Sounds Based on Latent Force Modelling | Recent advances in analysis of subband amplitude envelopes of natural sounds
have resulted in convincing synthesis, showing subband amplitudes to be a
crucial component of perception. Probabilistic latent variable analysis is
particularly revealing, but existing approaches don't incorporate prior
knowledge about the physical behaviour of amplitude envelopes, such as
exponential decay and feedback. We use latent force modelling, a probabilistic
learning paradigm that incorporates physical knowledge into Gaussian process
regression, to model correlation across spectral subband envelopes. We augment
the standard latent force model approach by explicitly modelling correlations
over multiple time steps. Incorporating this prior knowledge strengthens the
interpretation of the latent functions as the source that generated the signal.
We examine this interpretation via an experiment which shows that sounds
generated by sampling from our probabilistic model are perceived to be more
realistic than those generated by similar models based on nonnegative matrix
factorisation, even in cases where our model is outperformed from a
reconstruction error perspective.
|
Hierarchical Aggregation Approach for Distributed clustering of spatial
datasets | In this paper, we present a new approach of distributed clustering for
spatial datasets, based on an innovative and efficient aggregation technique.
This distributed approach consists of two phases: 1) local clustering phase,
where each node performs a clustering on its local data, 2) aggregation phase,
where the local clusters are aggregated to produce global clusters. This
approach is characterised by the fact that the local clusters are represented
in a simple and efficient way. And The aggregation phase is designed in such a
way that the final clusters are compact and accurate while the overall process
is efficient in both response time and memory allocation. We evaluated the
approach with different datasets and compared it to well-known clustering
techniques. The experimental results show that our approach is very promising
and outperforms all those algorithms
|
Short-term Memory of Deep RNN | The extension of deep learning towards temporal data processing is gaining an
increasing research interest. In this paper we investigate the properties of
state dynamics developed in successive levels of deep recurrent neural networks
(RNNs) in terms of short-term memory abilities. Our results reveal interesting
insights that shed light on the nature of layering as a factor of RNN design.
Noticeably, higher layers in a hierarchically organized RNN architecture
results to be inherently biased towards longer memory spans even prior to
training of the recurrent connections. Moreover, in the context of Reservoir
Computing framework, our analysis also points out the benefit of a layered
recurrent organization as an efficient approach to improve the memory skills of
reservoir models.
|
Deep Learning for Genomics: A Concise Overview | Advancements in genomic research such as high-throughput sequencing
techniques have driven modern genomic studies into "big data" disciplines. This
data explosion is constantly challenging conventional methods used in genomics.
In parallel with the urgent demand for robust algorithms, deep learning has
succeeded in a variety of fields such as vision, speech, and text processing.
Yet genomics entails unique challenges to deep learning since we are expecting
from deep learning a superhuman intelligence that explores beyond our knowledge
to interpret the genome. A powerful deep learning model should rely on
insightful utilization of task-specific knowledge. In this paper, we briefly
discuss the strengths of different deep learning models from a genomic
perspective so as to fit each particular task with a proper deep architecture,
and remark on practical considerations of developing modern deep learning
architectures for genomics. We also provide a concise review of deep learning
applications in various aspects of genomic research, as well as pointing out
potential opportunities and obstacles for future genomics applications.
|
VIBNN: Hardware Acceleration of Bayesian Neural Networks | Bayesian Neural Networks (BNNs) have been proposed to address the problem of
model uncertainty in training and inference. By introducing weights associated
with conditioned probability distributions, BNNs are capable of resolving the
overfitting issue commonly seen in conventional neural networks and allow for
small-data training, through the variational inference process. Frequent usage
of Gaussian random variables in this process requires a properly optimized
Gaussian Random Number Generator (GRNG). The high hardware cost of conventional
GRNG makes the hardware implementation of BNNs challenging.
In this paper, we propose VIBNN, an FPGA-based hardware accelerator design
for variational inference on BNNs. We explore the design space for massive
amount of Gaussian variable sampling tasks in BNNs. Specifically, we introduce
two high performance Gaussian (pseudo) random number generators: the RAM-based
Linear Feedback Gaussian Random Number Generator (RLF-GRNG), which is inspired
by the properties of binomial distribution and linear feedback logics; and the
Bayesian Neural Network-oriented Wallace Gaussian Random Number Generator. To
achieve high scalability and efficient memory access, we propose a deep
pipelined accelerator architecture with fast execution and good hardware
utilization. Experimental results demonstrate that the proposed VIBNN
implementations on an FPGA can achieve throughput of 321,543.4 Images/s and
energy efficiency upto 52,694.8 Images/J while maintaining similar accuracy as
its software counterpart.
|
Intriguing Properties of Randomly Weighted Networks: Generalizing While
Learning Next to Nothing | Training deep neural networks results in strong learned representations that
show good generalization capabilities. In most cases, training involves
iterative modification of all weights inside the network via back-propagation.
In Extreme Learning Machines, it has been suggested to set the first layer of a
network to fixed random values instead of learning it. In this paper, we
propose to take this approach a step further and fix almost all layers of a
deep convolutional neural network, allowing only a small portion of the weights
to be learned. As our experiments show, fixing even the majority of the
parameters of the network often results in performance which is on par with the
performance of learning all of them. The implications of this intriguing
property of deep neural networks are discussed and we suggest ways to harness
it to create more robust representations.
|
Deep UQ: Learning deep neural network surrogate models for high
dimensional uncertainty quantification | State-of-the-art computer codes for simulating real physical systems are
often characterized by a vast number of input parameters. Performing
uncertainty quantification (UQ) tasks with Monte Carlo (MC) methods is almost
always infeasible because of the need to perform hundreds of thousands or even
millions of forward model evaluations in order to obtain convergent statistics.
One, thus, tries to construct a cheap-to-evaluate surrogate model to replace
the forward model solver. For systems with large numbers of input parameters,
one has to deal with the curse of dimensionality - the exponential increase in
the volume of the input space, as the number of parameters increases linearly.
In this work, we demonstrate the use of deep neural networks (DNN) to construct
surrogate models for numerical simulators. We parameterize the structure of the
DNN in a manner that lends the DNN surrogate the interpretation of recovering a
low dimensional nonlinear manifold. The model response is a parameterized
nonlinear function of the low dimensional projections of the input. We think of
this low dimensional manifold as a nonlinear generalization of the notion of
the active subspace. Our approach is demonstrated with a problem on uncertainty
propagation in a stochastic elliptic partial differential equation (SPDE) with
uncertain diffusion coefficient. We deviate from traditional formulations of
the SPDE problem by not imposing a specific covariance structure on the random
diffusion coefficient. Instead, we attempt to solve a more challenging problem
of learning a map between an arbitrary snapshot of the diffusion field and the
response.
|
Bayesian Renewables Scenario Generation via Deep Generative Networks | We present a method to generate renewable scenarios using Bayesian
probabilities by implementing the Bayesian generative adversarial
network~(Bayesian GAN), which is a variant of generative adversarial networks
based on two interconnected deep neural networks. By using a Bayesian
formulation, generators can be constructed and trained to produce scenarios
that capture different salient modes in the data, allowing for better diversity
and more accurate representation of the underlying physical process. Compared
to conventional statistical models that are often hard to scale or sample from,
this method is model-free and can generate samples extremely efficiently. For
validation, we use wind and solar times-series data from NREL integration data
sets to train the Bayesian GAN. We demonstrate that proposed method is able to
generate clusters of wind scenarios with different variance and mean value, and
is able to distinguish and generate wind and solar scenarios simultaneously
even if the historical data are intentionally mixed.
|
Joint Binary Neural Network for Multi-label Learning with Applications
to Emotion Classification | Recently the deep learning techniques have achieved success in multi-label
classification due to its automatic representation learning ability and the
end-to-end learning framework. Existing deep neural networks in multi-label
classification can be divided into two kinds: binary relevance neural network
(BRNN) and threshold dependent neural network (TDNN). However, the former needs
to train a set of isolate binary networks which ignore dependencies between
labels and have heavy computational load, while the latter needs an additional
threshold function mechanism to transform the multi-class probabilities to
multi-label outputs. In this paper, we propose a joint binary neural network
(JBNN), to address these shortcomings. In JBNN, the representation of the text
is fed to a set of logistic functions instead of a softmax function, and the
multiple binary classifications are carried out synchronously in one neural
network framework. Moreover, the relations between labels are captured via
training on a joint binary cross entropy (JBCE) loss. To better meet
multi-label emotion classification, we further proposed to incorporate the
prior label relations into the JBCE loss. The experimental results on the
benchmark dataset show that our model performs significantly better than the
state-of-the-art multi-label emotion classification methods, in both
classification performance and computational efficiency.
|
Learning Parametric Closed-Loop Policies for Markov Potential Games | Multiagent systems where agents interact among themselves and with a
stochastic environment can be formalized as stochastic games. We study a
subclass named Markov potential games (MPGs) that appear often in economic and
engineering applications when the agents share a common resource. We consider
MPGs with continuous state-action variables, coupled constraints and nonconvex
rewards. Previous analysis followed a variational approach that is only valid
for very simple cases (convex rewards, invertible dynamics, and no coupled
constraints); or considered deterministic dynamics and provided open-loop (OL)
analysis, studying strategies that consist in predefined action sequences,
which are not optimal for stochastic environments. We present a closed-loop
(CL) analysis for MPGs and consider parametric policies that depend on the
current state. We provide easily verifiable, sufficient and necessary
conditions for a stochastic game to be an MPG, even for complex parametric
functions (e.g., deep neural networks); and show that a closed-loop Nash
equilibrium (NE) can be found (or at least approximated) by solving a related
optimal control problem (OCP). This is useful since solving an OCP--which is a
single-objective problem--is usually much simpler than solving the original set
of coupled OCPs that form the game--which is a multiobjective control problem.
This is a considerable improvement over the previously standard approach for
the CL analysis of MPGs, which gives no approximate solution if no NE belongs
to the chosen parametric family, and which is practical only for simple
parametric forms. We illustrate the theoretical contributions with an example
by applying our approach to a noncooperative communications engineering game.
We then solve the game with a deep reinforcement learning algorithm that learns
policies that closely approximates an exact variational NE of the game.
|
GeniePath: Graph Neural Networks with Adaptive Receptive Paths | We present, GeniePath, a scalable approach for learning adaptive receptive
fields of neural networks defined on permutation invariant graph data. In
GeniePath, we propose an adaptive path layer consists of two complementary
functions designed for breadth and depth exploration respectively, where the
former learns the importance of different sized neighborhoods, while the latter
extracts and filters signals aggregated from neighbors of different hops away.
Our method works in both transductive and inductive settings, and extensive
experiments compared with competitive methods show that our approaches yield
state-of-the-art results on large graphs.
|
Active, Continual Fine Tuning of Convolutional Neural Networks for
Reducing Annotation Efforts | The splendid success of convolutional neural networks (CNNs) in computer
vision is largely attributable to the availability of massive annotated
datasets, such as ImageNet and Places. However, in medical imaging, it is
challenging to create such large annotated datasets, as annotating medical
images is not only tedious, laborious, and time consuming, but it also demands
costly, specialty-oriented skills, which are not easily accessible. To
dramatically reduce annotation cost, this paper presents a novel method to
naturally integrate active learning and transfer learning (fine-tuning) into a
single framework, which starts directly with a pre-trained CNN to seek "worthy"
samples for annotation and gradually enhances the (fine-tuned) CNN via
continual fine-tuning. We have evaluated our method using three distinct
medical imaging applications, demonstrating that it can reduce annotation
efforts by at least half compared with random selection.
|
Multi-attention Recurrent Network for Human Communication Comprehension | Human face-to-face communication is a complex multimodal signal. We use words
(language modality), gestures (vision modality) and changes in tone (acoustic
modality) to convey our intentions. Humans easily process and understand
face-to-face communication, however, comprehending this form of communication
remains a significant challenge for Artificial Intelligence (AI). AI must
understand each modality and the interactions between them that shape human
communication. In this paper, we present a novel neural architecture for
understanding human communication called the Multi-attention Recurrent Network
(MARN). The main strength of our model comes from discovering interactions
between modalities through time using a neural component called the
Multi-attention Block (MAB) and storing them in the hybrid memory of a
recurrent component called the Long-short Term Hybrid Memory (LSTHM). We
perform extensive comparisons on six publicly available datasets for multimodal
sentiment analysis, speaker trait recognition and emotion recognition. MARN
shows state-of-the-art performance on all the datasets.
|
Multimodal Sentiment Analysis with Word-Level Fusion and Reinforcement
Learning | With the increasing popularity of video sharing websites such as YouTube and
Facebook, multimodal sentiment analysis has received increasing attention from
the scientific community. Contrary to previous works in multimodal sentiment
analysis which focus on holistic information in speech segments such as bag of
words representations and average facial expression intensity, we develop a
novel deep architecture for multimodal sentiment analysis that performs
modality fusion at the word level. In this paper, we propose the Gated
Multimodal Embedding LSTM with Temporal Attention (GME-LSTM(A)) model that is
composed of 2 modules. The Gated Multimodal Embedding alleviates the
difficulties of fusion when there are noisy modalities. The LSTM with Temporal
Attention performs word level fusion at a finer fusion resolution between input
modalities and attends to the most important time steps. As a result, the
GME-LSTM(A) is able to better model the multimodal structure of speech through
time and perform better sentiment comprehension. We demonstrate the
effectiveness of this approach on the publicly-available Multimodal Corpus of
Sentiment Intensity and Subjectivity Analysis (CMU-MOSI) dataset by achieving
state-of-the-art sentiment classification and regression results. Qualitative
analysis on our model emphasizes the importance of the Temporal Attention Layer
in sentiment prediction because the additional acoustic and visual modalities
are noisy. We also demonstrate the effectiveness of the Gated Multimodal
Embedding in selectively filtering these noisy modalities out. Our results and
analysis open new areas in the study of sentiment analysis in human
communication and provide new models for multimodal fusion.
|
Memory Fusion Network for Multi-view Sequential Learning | Multi-view sequential learning is a fundamental problem in machine learning
dealing with multi-view sequences. In a multi-view sequence, there exists two
forms of interactions between different views: view-specific interactions and
cross-view interactions. In this paper, we present a new neural architecture
for multi-view sequential learning called the Memory Fusion Network (MFN) that
explicitly accounts for both interactions in a neural architecture and
continuously models them through time. The first component of the MFN is called
the System of LSTMs, where view-specific interactions are learned in isolation
through assigning an LSTM function to each view. The cross-view interactions
are then identified using a special attention mechanism called the Delta-memory
Attention Network (DMAN) and summarized through time with a Multi-view Gated
Memory. Through extensive experimentation, MFN is compared to various proposed
approaches for multi-view sequential learning on multiple publicly available
benchmark datasets. MFN outperforms all the existing multi-view approaches.
Furthermore, MFN outperforms all current state-of-the-art models, setting new
state-of-the-art results for these multi-view datasets.
|
Mixed Precision Training of Convolutional Neural Networks using Integer
Operations | The state-of-the-art (SOTA) for mixed precision training is dominated by
variants of low precision floating point operations, and in particular, FP16
accumulating into FP32 Micikevicius et al. (2017). On the other hand, while a
lot of research has also happened in the domain of low and mixed-precision
Integer training, these works either present results for non-SOTA networks (for
instance only AlexNet for ImageNet-1K), or relatively small datasets (like
CIFAR-10). In this work, we train state-of-the-art visual understanding neural
networks on the ImageNet-1K dataset, with Integer operations on General Purpose
(GP) hardware. In particular, we focus on Integer Fused-Multiply-and-Accumulate
(FMA) operations which take two pairs of INT16 operands and accumulate results
into an INT32 output.We propose a shared exponent representation of tensors and
develop a Dynamic Fixed Point (DFP) scheme suitable for common neural network
operations. The nuances of developing an efficient integer convolution kernel
is examined, including methods to handle overflow of the INT32 accumulator. We
implement CNN training for ResNet-50, GoogLeNet-v1, VGG-16 and AlexNet; and
these networks achieve or exceed SOTA accuracy within the same number of
iterations as their FP32 counterparts without any change in hyper-parameters
and with a 1.8X improvement in end-to-end training throughput. To the best of
our knowledge these results represent the first INT16 training results on GP
hardware for ImageNet-1K dataset using SOTA CNNs and achieve highest reported
accuracy using half-precision
|
Contextual Bandit with Adaptive Feature Extraction | We consider an online decision making setting known as contextual bandit
problem, and propose an approach for improving contextual bandit performance by
using an adaptive feature extraction (representation learning) based on online
clustering. Our approach starts with an off-line pre-training on unlabeled
history of contexts (which can be exploited by our approach, but not by the
standard contextual bandit), followed by an online selection and adaptation of
encoders. Specifically, given an input sample (context), the proposed approach
selects the most appropriate encoding function to extract a feature vector
which becomes an input for a contextual bandit, and updates both the bandit and
the encoding function based on the context and on the feedback (reward). Our
experiments on a variety of datasets, and both in stationary and non-stationary
environments of several kinds demonstrate clear advantages of the proposed
adaptive representation learning over the standard contextual bandit based on
"raw" input contexts.
|
Fairness and Accountability Design Needs for Algorithmic Support in
High-Stakes Public Sector Decision-Making | Calls for heightened consideration of fairness and accountability in
algorithmically-informed public decisions---like taxation, justice, and child
protection---are now commonplace. How might designers support such human
values? We interviewed 27 public sector machine learning practitioners across 5
OECD countries regarding challenges understanding and imbuing public values
into their work. The results suggest a disconnect between organisational and
institutional realities, constraints and needs, and those addressed by current
research into usable, transparent and 'discrimination-aware' machine
learning---absences likely to undermine practical initiatives unless addressed.
We see design opportunities in this disconnect, such as in supporting the
tracking of concept drift in secondary data sources, and in building usable
transparency tools to identify risks and incorporate domain knowledge, aimed
both at managers and at the 'street-level bureaucrats' on the frontlines of
public service. We conclude by outlining ethical challenges and future
directions for collaboration in these high-stakes applications.
|
Multi-task Learning for Continuous Control | Reliable and effective multi-task learning is a prerequisite for the
development of robotic agents that can quickly learn to accomplish related,
everyday tasks. However, in the reinforcement learning domain, multi-task
learning has not exhibited the same level of success as in other domains, such
as computer vision. In addition, most reinforcement learning research on
multi-task learning has been focused on discrete action spaces, which are not
used for robotic control in the real-world. In this work, we apply multi-task
learning methods to continuous action spaces and benchmark their performance on
a series of simulated continuous control tasks. Most notably, we show that
multi-task learning outperforms our baselines and alternative knowledge sharing
methods.
|
Deep Temporal Clustering : Fully Unsupervised Learning of Time-Domain
Features | Unsupervised learning of time series data, also known as temporal clustering,
is a challenging problem in machine learning. Here we propose a novel
algorithm, Deep Temporal Clustering (DTC), to naturally integrate
dimensionality reduction and temporal clustering into a single end-to-end
learning framework, fully unsupervised. The algorithm utilizes an autoencoder
for temporal dimensionality reduction and a novel temporal clustering layer for
cluster assignment. Then it jointly optimizes the clustering objective and the
dimensionality reduction objec tive. Based on requirement and application, the
temporal clustering layer can be customized with any temporal similarity
metric. Several similarity metrics and state-of-the-art algorithms are
considered and compared. To gain insight into temporal features that the
network has learned for its clustering, we apply a visualization method that
generates a region of interest heatmap for the time series. The viability of
the algorithm is demonstrated using time series data from diverse domains,
ranging from earthquakes to spacecraft sensor data. In each case, we show that
the proposed algorithm outperforms traditional methods. The superior
performance is attributed to the fully integrated temporal dimensionality
reduction and clustering criterion.
|
Hierarchical Adversarially Learned Inference | We propose a novel hierarchical generative model with a simple Markovian
structure and a corresponding inference model. Both the generative and
inference model are trained using the adversarial learning paradigm. We
demonstrate that the hierarchical structure supports the learning of
progressively more abstract representations as well as providing semantically
meaningful reconstructions with different levels of fidelity. Furthermore, we
show that minimizing the Jensen-Shanon divergence between the generative and
inference network is enough to minimize the reconstruction error. The resulting
semantically meaningful hierarchical latent structure discovery is exemplified
on the CelebA dataset. There, we show that the features learned by our model in
an unsupervised way outperform the best handcrafted features. Furthermore, the
extracted features remain competitive when compared to several recent deep
supervised approaches on an attribute prediction task on CelebA. Finally, we
leverage the model's inference network to achieve state-of-the-art performance
on a semi-supervised variant of the MNIST digit classification task.
|
Software Engineers vs. Machine Learning Algorithms: An Empirical Study
Assessing Performance and Reuse Tasks | Several papers have recently contained reports on applying machine learning
(ML) to the automation of software engineering (SE) tasks, such as project
management, modeling and development. However, there appear to be no approaches
comparing how software engineers fare against machine-learning algorithms as
applied to specific software development tasks. Such a comparison is essential
to gain insight into which tasks are better performed by humans and which by
machine learning and how cooperative work or human-in-the-loop processes can be
implemented more effectively. In this paper, we present an empirical study that
compares how software engineers and machine-learning algorithms perform and
reuse tasks. The empirical study involves the synthesis of the control
structure of an autonomous streetlight application. Our approach consists of
four steps. First, we solved the problem using machine learning to determine
specific performance and reuse tasks. Second, we asked software engineers with
different domain knowledge levels to provide a solution to the same tasks.
Third, we compared how software engineers fare against machine-learning
algorithms when accomplishing the performance and reuse tasks based on criteria
such as energy consumption and safety. Finally, we analyzed the results to
understand which tasks are better performed by either humans or algorithms so
that they can work together more effectively. Such an understanding and the
resulting human-in-the-loop approaches, which take into account the strengths
and weaknesses of humans and machine-learning algorithms, are fundamental not
only to provide a basis for cooperative work in support of software
engineering, but also, in other areas.
|
Non-Gaussian information from weak lensing data via deep learning | Weak lensing maps contain information beyond two-point statistics on small
scales. Much recent work has tried to extract this information through a range
of different observables or via nonlinear transformations of the lensing field.
Here we train and apply a 2D convolutional neural network to simulated
noiseless lensing maps covering 96 different cosmological models over a range
of {$\Omega_m,\sigma_8$}. Using the area of the confidence contour in the
{$\Omega_m,\sigma_8$} plane as a figure-of-merit, derived from simulated
convergence maps smoothed on a scale of 1.0 arcmin, we show that the neural
network yields $\approx 5 \times$ tighter constraints than the power spectrum,
and $\approx 4 \times$ tighter than the lensing peaks. Such gains illustrate
the extent to which weak lensing data encode cosmological information not
accessible to the power spectrum or even other, non-Gaussian statistics such as
lensing peaks.
|
Learning Compact Neural Networks with Regularization | Proper regularization is critical for speeding up training, improving
generalization performance, and learning compact models that are cost
efficient. We propose and analyze regularized gradient descent algorithms for
learning shallow neural networks. Our framework is general and covers
weight-sharing (convolutional networks), sparsity (network pruning), and
low-rank constraints among others. We first introduce covering dimension to
quantify the complexity of the constraint set and provide insights on the
generalization properties. Then, we show that proposed algorithms become
well-behaved and local linear convergence occurs once the amount of data
exceeds the covering dimension. Overall, our results demonstrate that
near-optimal sample complexity is sufficient for efficient learning and
illustrate how regularization can be beneficial to learn over-parameterized
networks.
|
Counting and Sampling from Markov Equivalent DAGs Using Clique Trees | A directed acyclic graph (DAG) is the most common graphical model for
representing causal relationships among a set of variables. When restricted to
using only observational data, the structure of the ground truth DAG is
identifiable only up to Markov equivalence, based on conditional independence
relations among the variables. Therefore, the number of DAGs equivalent to the
ground truth DAG is an indicator of the causal complexity of the underlying
structure--roughly speaking, it shows how many interventions or how much
additional information is further needed to recover the underlying DAG. In this
paper, we propose a new technique for counting the number of DAGs in a Markov
equivalence class. Our approach is based on the clique tree representation of
chordal graphs. We show that in the case of bounded degree graphs, the proposed
algorithm is polynomial time. We further demonstrate that this technique can be
utilized for uniform sampling from a Markov equivalence class, which provides a
stochastic way to enumerate DAGs in the equivalence class and may be needed for
finding the best DAG or for causal inference given the equivalence class as
input. We also extend our counting and sampling method to the case where prior
knowledge about the underlying DAG is available, and present applications of
this extension in causal experiment design and estimating the causal effect of
joint interventions.
|
Enhancing Multi-Class Classification of Random Forest using Random
Vector Functional Neural Network and Oblique Decision Surfaces | Both neural networks and decision trees are popular machine learning methods
and are widely used to solve problems from diverse domains. These two
classifiers are commonly used base classifiers in an ensemble framework. In
this paper, we first present a new variant of oblique decision tree based on a
linear classifier, then construct an ensemble classifier based on the fusion of
a fast neural network, random vector functional link network and oblique
decision trees. Random Vector Functional Link Network has an elegant closed
form solution with extremely short training time. The neural network partitions
each training bag (obtained using bagging) at the root level into C subsets
where C is the number of classes in the dataset and subsequently, C oblique
decision trees are trained on such partitions. The proposed method provides a
rich insight into the data by grouping the confusing or hard to classify
samples for each class and thus, provides an opportunity to employ fine-grained
classification rule over the data. The performance of the ensemble classifier
is evaluated on several multi-class datasets where it demonstrates a superior
performance compared to other state-of- the-art classifiers.
|
Task-Aware Compressed Sensing with Generative Adversarial Networks | In recent years, neural network approaches have been widely adopted for
machine learning tasks, with applications in computer vision. More recently,
unsupervised generative models based on neural networks have been successfully
applied to model data distributions via low-dimensional latent spaces. In this
paper, we use Generative Adversarial Networks (GANs) to impose structure in
compressed sensing problems, replacing the usual sparsity constraint. We
propose to train the GANs in a task-aware fashion, specifically for
reconstruction tasks. We also show that it is possible to train our model
without using any (or much) non-compressed data. Finally, we show that the
latent space of the GAN carries discriminative information and can further be
regularized to generate input features for general inference tasks. We
demonstrate the effectiveness of our method on a variety of reconstruction and
classification problems.
|
Online Compact Convexified Factorization Machine | Factorization Machine (FM) is a supervised learning approach with a powerful
capability of feature engineering. It yields state-of-the-art performance in
various batch learning tasks where all the training data is made available
prior to the training. However, in real-world applications where the data
arrives sequentially in a streaming manner, the high cost of re-training with
batch learning algorithms has posed formidable challenges in the online
learning scenario. The initial challenge is that no prior formulations of FM
could fulfill the requirements in Online Convex Optimization (OCO) -- the
paramount framework for online learning algorithm design. To address the
aforementioned challenge, we invent a new convexification scheme leading to a
Compact Convexified FM (CCFM) that seamlessly meets the requirements in OCO.
However for learning Compact Convexified FM (CCFM) in the online learning
setting, most existing algorithms suffer from expensive projection operations.
To address this subsequent challenge, we follow the general projection-free
algorithmic framework of Online Conditional Gradient and propose an Online
Compact Convex Factorization Machine (OCCFM) algorithm that eschews the
projection operation with efficient linear optimization steps. In support of
the proposed OCCFM in terms of its theoretical foundation, we prove that the
developed algorithm achieves a sub-linear regret bound. To evaluate the
empirical performance of OCCFM, we conduct extensive experiments on 6
real-world datasets for online recommendation and binary classification tasks.
The experimental results show that OCCFM outperforms the state-of-art online
learning algorithms.
|
To understand deep learning we need to understand kernel learning | Generalization performance of classifiers in deep learning has recently
become a subject of intense study. Deep models, typically over-parametrized,
tend to fit the training data exactly. Despite this "overfitting", they perform
well on test data, a phenomenon not yet fully understood.
The first point of our paper is that strong performance of overfitted
classifiers is not a unique feature of deep learning. Using six real-world and
two synthetic datasets, we establish experimentally that kernel machines
trained to have zero classification or near zero regression error perform very
well on test data, even when the labels are corrupted with a high level of
noise. We proceed to give a lower bound on the norm of zero loss solutions for
smooth kernels, showing that they increase nearly exponentially with data size.
We point out that this is difficult to reconcile with the existing
generalization bounds. Moreover, none of the bounds produce non-trivial results
for interpolating solutions.
Second, we show experimentally that (non-smooth) Laplacian kernels easily fit
random labels, a finding that parallels results for ReLU neural networks. In
contrast, fitting noisy data requires many more epochs for smooth Gaussian
kernels. Similar performance of overfitted Laplacian and Gaussian classifiers
on test, suggests that generalization is tied to the properties of the kernel
function rather than the optimization process.
Certain key phenomena of deep learning are manifested similarly in kernel
methods in the modern "overfitted" regime. The combination of the experimental
and theoretical results presented in this paper indicates a need for new
theoretical ideas for understanding properties of classical kernel methods. We
argue that progress on understanding deep learning will be difficult until more
tractable "shallow" kernel methods are better understood.
|
A Learning-based Approach to Joint Content Caching and Recommendation at
Base Stations | Recommendation system is able to shape user demands, which can be used for
boosting caching gain. In this paper, we jointly optimize content caching and
recommendation at base stations to maximize the caching gain meanwhile not
compromising the user preference. We first propose a model to capture the
impact of recommendation on user demands, which is controlled by a
user-specific psychological threshold. We then formulate a joint caching and
recommendation problem maximizing the successful offloading probability, which
is a mixed integer programming problem. We develop a hierarchical iterative
algorithm to solve the problem when the threshold is known. Since the user
threshold is unknown in practice, we proceed to propose an $\varepsilon$-greedy
algorithm to find the solution by learning the threshold via interactions with
users. Simulation results show that the proposed algorithms improve the
successful offloading probability compared with prior works with/without
recommendation. The $\varepsilon$-greedy algorithm learns the user threshold
quickly, and achieves more than $1-\varepsilon$ of the performance obtained by
the algorithm with known threshold.
|
First-order Adversarial Vulnerability of Neural Networks and Input
Dimension | Over the past few years, neural networks were proven vulnerable to
adversarial images: targeted but imperceptible image perturbations lead to
drastically different predictions. We show that adversarial vulnerability
increases with the gradients of the training objective when viewed as a
function of the inputs. Surprisingly, vulnerability does not depend on network
topology: for many standard network architectures, we prove that at
initialization, the $\ell_1$-norm of these gradients grows as the square root
of the input dimension, leaving the networks increasingly vulnerable with
growing image size. We empirically show that this dimension dependence persists
after either usual or robust training, but gets attenuated with higher
regularization.
|
Interactive Grounded Language Acquisition and Generalization in a 2D
World | We build a virtual agent for learning language in a 2D maze-like world. The
agent sees images of the surrounding environment, listens to a virtual teacher,
and takes actions to receive rewards. It interactively learns the teacher's
language from scratch based on two language use cases: sentence-directed
navigation and question answering. It learns simultaneously the visual
representations of the world, the language, and the action control. By
disentangling language grounding from other computational routines and sharing
a concept detection function between language grounding and prediction, the
agent reliably interpolates and extrapolates to interpret sentences that
contain new word combinations or new words missing from training sentences. The
new words are transferred from the answers of language prediction. Such a
language ability is trained and evaluated on a population of over 1.6 million
distinct sentences consisting of 119 object words, 8 color words, 9
spatial-relation words, and 50 grammatical words. The proposed model
significantly outperforms five comparison methods for interpreting zero-shot
sentences. In addition, we demonstrate human-interpretable intermediate outputs
of the model in the appendix.
|
Hardening Deep Neural Networks via Adversarial Model Cascades | Deep neural networks (DNNs) are vulnerable to malicious inputs crafted by an
adversary to produce erroneous outputs. Works on securing neural networks
against adversarial examples achieve high empirical robustness on simple
datasets such as MNIST. However, these techniques are inadequate when
empirically tested on complex data sets such as CIFAR-10 and SVHN. Further,
existing techniques are designed to target specific attacks and fail to
generalize across attacks. We propose the Adversarial Model Cascades (AMC) as a
way to tackle the above inadequacies. Our approach trains a cascade of models
sequentially where each model is optimized to be robust towards a mixture of
multiple attacks. Ultimately, it yields a single model which is secure against
a wide range of attacks; namely FGSM, Elastic, Virtual Adversarial
Perturbations and Madry. On an average, AMC increases the model's empirical
robustness against various attacks simultaneously, by a significant margin (of
6.225% for MNIST, 5.075% for SVHN and 2.65% for CIFAR10). At the same time, the
model's performance on non-adversarial inputs is comparable to the
state-of-the-art models.
|
Explicit Inductive Bias for Transfer Learning with Convolutional
Networks | In inductive transfer learning, fine-tuning pre-trained convolutional
networks substantially outperforms training from scratch. When using
fine-tuning, the underlying assumption is that the pre-trained model extracts
generic features, which are at least partially relevant for solving the target
task, but would be difficult to extract from the limited amount of data
available on the target task. However, besides the initialization with the
pre-trained model and the early stopping, there is no mechanism in fine-tuning
for retaining the features learned on the source task. In this paper, we
investigate several regularization schemes that explicitly promote the
similarity of the final solution with the initial model. We show the benefit of
having an explicit inductive bias towards the initial model, and we eventually
recommend a simple $L^2$ penalty with the pre-trained model being a reference
as the baseline of penalty for transfer learning tasks.
|
Linear Convergence of the Primal-Dual Gradient Method for Convex-Concave
Saddle Point Problems without Strong Convexity | We consider the convex-concave saddle point problem $\min_{x}\max_{y}
f(x)+y^\top A x-g(y)$ where $f$ is smooth and convex and $g$ is smooth and
strongly convex. We prove that if the coupling matrix $A$ has full column rank,
the vanilla primal-dual gradient method can achieve linear convergence even if
$f$ is not strongly convex. Our result generalizes previous work which either
requires $f$ and $g$ to be quadratic functions or requires proximal mappings
for both $f$ and $g$. We adopt a novel analysis technique that in each
iteration uses a "ghost" update as a reference, and show that the iterates in
the primal-dual gradient method converge to this "ghost" sequence. Using the
same technique we further give an analysis for the primal-dual stochastic
variance reduced gradient (SVRG) method for convex-concave saddle point
problems with a finite-sum structure.
|
Background subtraction using the factored 3-way restricted Boltzmann
machines | In this paper, we proposed a method for reconstructing the 3D model based on
continuous sensory input. The robot can draw on extremely large data from the
real world using various sensors. However, the sensory inputs are usually too
noisy and high-dimensional data. It is very difficult and time consuming for
robot to process using such raw data when the robot tries to construct 3D
model. Hence, there needs to be a method that can extract useful information
from such sensory inputs. To address this problem our method utilizes the
concept of Object Semantic Hierarchy (OSH). Different from the previous work
that used this hierarchy framework, we extract the motion information using the
Deep Belief Network technique instead of applying classical computer vision
approaches. We have trained on two large sets of random dot images (10,000)
which are translated and rotated, respectively, and have successfully extracted
several bases that explain the translation and rotation motion. Based on this
translation and rotation bases, background subtraction have become possible
using Object Semantic Hierarchy.
|
The Matrix Calculus You Need For Deep Learning | This paper is an attempt to explain all the matrix calculus you need in order
to understand the training of deep neural networks. We assume no math knowledge
beyond what you learned in calculus 1, and provide links to help you refresh
the necessary math where needed. Note that you do not need to understand this
material before you start learning to train and use deep learning in practice;
rather, this material is for those who are already familiar with the basics of
neural networks, and wish to deepen their understanding of the underlying math.
Don't worry if you get stuck at some point along the way---just go back and
reread the previous section, and try writing down and working through some
examples. And if you're still stuck, we're happy to answer your questions in
the Theory category at forums.fast.ai. Note: There is a reference section at
the end of the paper summarizing all the key matrix calculus rules and
terminology discussed here. See related articles at http://explained.ai
|
Real-time Prediction of Intermediate-Horizon Automotive Collision Risk | Advanced collision avoidance and driver hand-off systems can benefit from the
ability to accurately predict, in real time, the probability a vehicle will be
involved in a collision within an intermediate horizon of 10 to 20 seconds. The
rarity of collisions in real-world data poses a significant challenge to
developing this capability because, as we demonstrate empirically,
intermediate-horizon risk prediction depends heavily on high-dimensional driver
behavioral features. As a result, a large amount of data is required to fit an
effective predictive model. In this paper, we assess whether simulated data can
help alleviate this issue. Focusing on highway driving, we present a three-step
approach for generating data and fitting a predictive model capable of
real-time prediction. First, high-risk automotive scenes are generated using
importance sampling on a learned Bayesian network scene model. Second,
collision risk is estimated through Monte Carlo simulation. Third, a neural
network domain adaptation model is trained on real and simulated data to
address discrepancies between the two domains. Experiments indicate that
simulated data can mitigate issues resulting from collision rarity, thereby
improving risk prediction in real-world data.
|
Blind Pre-Processing: A Robust Defense Method Against Adversarial
Examples | Deep learning algorithms and networks are vulnerable to perturbed inputs
which is known as the adversarial attack. Many defense methodologies have been
investigated to defend against such adversarial attack. In this work, we
propose a novel methodology to defend the existing powerful attack model. We
for the first time introduce a new attacking scheme for the attacker and set a
practical constraint for white box attack. Under this proposed attacking
scheme, we present the best defense ever reported against some of the recent
strong attacks. It consists of a set of nonlinear function to process the input
data which will make it more robust over the adversarial attack. However, we
make this processing layer completely hidden from the attacker. Blind
pre-processing improves the white box attack accuracy of MNIST from 94.3\% to
98.7\%. Even with increasing defense when others defenses completely fail,
blind pre-processing remains one of the strongest ever reported. Another
strength of our defense is that it eliminates the need for adversarial training
as it can significantly increase the MNIST accuracy without adversarial
training as well. Additionally, blind pre-processing can also increase the
inference accuracy in the face of a powerful attack on CIFAR-10 and SVHN data
set as well without much sacrificing clean data accuracy.
|
One-Shot Imitation from Observing Humans via Domain-Adaptive
Meta-Learning | Humans and animals are capable of learning a new behavior by observing others
perform the skill just once. We consider the problem of allowing a robot to do
the same -- learning from a raw video pixels of a human, even when there is
substantial domain shift in the perspective, environment, and embodiment
between the robot and the observed human. Prior approaches to this problem have
hand-specified how human and robot actions correspond and often relied on
explicit human pose detection systems. In this work, we present an approach for
one-shot learning from a video of a human by using human and robot
demonstration data from a variety of previous tasks to build up prior knowledge
through meta-learning. Then, combining this prior knowledge and only a single
video demonstration from a human, the robot can perform the task that the human
demonstrated. We show experiments on both a PR2 arm and a Sawyer arm,
demonstrating that after meta-learning, the robot can learn to place, push, and
pick-and-place new objects using just one video of a human performing the
manipulation.
|
IMPALA: Scalable Distributed Deep-RL with Importance Weighted
Actor-Learner Architectures | In this work we aim to solve a large collection of tasks using a single
reinforcement learning agent with a single set of parameters. A key challenge
is to handle the increased amount of data and extended training time. We have
developed a new distributed agent IMPALA (Importance Weighted Actor-Learner
Architecture) that not only uses resources more efficiently in single-machine
training but also scales to thousands of machines without sacrificing data
efficiency or resource utilisation. We achieve stable learning at high
throughput by combining decoupled acting and learning with a novel off-policy
correction method called V-trace. We demonstrate the effectiveness of IMPALA
for multi-task reinforcement learning on DMLab-30 (a set of 30 tasks from the
DeepMind Lab environment (Beattie et al., 2016)) and Atari-57 (all available
Atari games in Arcade Learning Environment (Bellemare et al., 2013a)). Our
results show that IMPALA is able to achieve better performance than previous
agents with less data, and crucially exhibits positive transfer between tasks
as a result of its multi-task approach.
|
Selective Sampling and Mixture Models in Generative Adversarial Networks | In this paper, we propose a multi-generator extension to the adversarial
training framework, in which the objective of each generator is to represent a
unique component of a target mixture distribution. In the training phase, the
generators cooperate to represent, as a mixture, the target distribution while
maintaining distinct manifolds. As opposed to traditional generative models,
inference from a particular generator after training resembles selective
sampling from a unique component in the target distribution. We demonstrate the
feasibility of the proposed architecture both analytically and with basic
Multi-Layer Perceptron (MLP) models trained on the MNIST dataset.
|
Alleviating catastrophic forgetting using context-dependent gating and
synaptic stabilization | Humans and most animals can learn new tasks without forgetting old ones.
However, training artificial neural networks (ANNs) on new tasks typically
cause it to forget previously learned tasks. This phenomenon is the result of
"catastrophic forgetting", in which training an ANN disrupts connection weights
that were important for solving previous tasks, degrading task performance.
Several recent studies have proposed methods to stabilize connection weights of
ANNs that are deemed most important for solving a task, which helps alleviate
catastrophic forgetting. Here, drawing inspiration from algorithms that are
believed to be implemented in vivo, we propose a complementary method: adding a
context-dependent gating signal, such that only sparse, mostly non-overlapping
patterns of units are active for any one task. This method is easy to
implement, requires little computational overhead, and allows ANNs to maintain
high performance across large numbers of sequentially presented tasks when
combined with weight stabilization. This work provides another example of how
neuroscience-inspired algorithms can benefit ANN design and capability.
|
MotifNet: a motif-based Graph Convolutional Network for directed graphs | Deep learning on graphs and in particular, graph convolutional neural
networks, have recently attracted significant attention in the machine learning
community. Many of such techniques explore the analogy between the graph
Laplacian eigenvectors and the classical Fourier basis, allowing to formulate
the convolution as a multiplication in the spectral domain. One of the key
drawback of spectral CNNs is their explicit assumption of an undirected graph,
leading to a symmetric Laplacian matrix with orthogonal eigendecomposition. In
this work we propose MotifNet, a graph CNN capable of dealing with directed
graphs by exploiting local graph motifs. We present experimental evidence
showing the advantage of our approach on real data.
|
Re-Weighted Learning for Sparsifying Deep Neural Networks | This paper addresses the topic of sparsifying deep neural networks (DNN's).
While DNN's are powerful models that achieve state-of-the-art performance on a
large number of tasks, the large number of model parameters poses serious
storage and computational challenges. To combat these difficulties, a growing
line of work focuses on pruning network weights without sacrificing
performance. We propose a general affine scaling transformation (AST) algorithm
to sparsify DNN's. Our approach follows in the footsteps of popular sparse
recovery techniques, which have yet to be explored in the context of DNN's. We
describe a principled framework for transforming densely connected DNN's into
sparsely connected ones without sacrificing network performance. Unlike
existing methods, our approach is able to learn sparse connections at each
layer simultaneously, and achieves comparable pruning results on the
architecture tested.
|
Deep Learning with a Rethinking Structure for Multi-label Classification | Multi-label classification (MLC) is an important class of machine learning
problems that come with a wide spectrum of applications, each demanding a
possibly different evaluation criterion. When solving the MLC problems, we
generally expect the learning algorithm to take the hidden correlation of the
labels into account to improve the prediction performance. Extracting the
hidden correlation is generally a challenging task. In this work, we propose a
novel deep learning framework to better extract the hidden correlation with the
help of the memory structure within recurrent neural networks. The memory
stores the temporary guesses on the labels and effectively allows the framework
to rethink about the goodness and correlation of the guesses before making the
final prediction. Furthermore, the rethinking process makes it easy to adapt to
different evaluation criteria to match real-world application needs. In
particular, the framework can be trained in an end-to-end style with respect to
any given MLC evaluation criteria. The end-to-end design can be seamlessly
combined with other deep learning techniques to conquer challenging MLC
problems like image tagging. Experimental results across many real-world data
sets justify that the rethinking framework indeed improves MLC performance
across different evaluation criteria and leads to superior performance over
state-of-the-art MLC algorithms.
|
Weakly-supervised Dictionary Learning | We present a probabilistic modeling and inference framework for
discriminative analysis dictionary learning under a weak supervision setting.
Dictionary learning approaches have been widely used for tasks such as
low-level signal denoising and restoration as well as high-level classification
tasks, which can be applied to audio and image analysis. Synthesis dictionary
learning aims at jointly learning a dictionary and corresponding sparse
coefficients to provide accurate data representation. This approach is useful
for denoising and signal restoration, but may lead to sub-optimal
classification performance. By contrast, analysis dictionary learning provides
a transform that maps data to a sparse discriminative representation suitable
for classification. We consider the problem of analysis dictionary learning for
time-series data under a weak supervision setting in which signals are assigned
with a global label instead of an instantaneous label signal. We propose a
discriminative probabilistic model that incorporates both label information and
sparsity constraints on the underlying latent instantaneous label signal using
cardinality control. We present the expectation maximization (EM) procedure for
maximum likelihood estimation (MLE) of the proposed model. To facilitate a
computationally efficient E-step, we propose both a chain and a novel tree
graph reformulation of the graphical model. The performance of the proposed
model is demonstrated on both synthetic and real-world data.
|
Bayesian Coreset Construction via Greedy Iterative Geodesic Ascent | Coherent uncertainty quantification is a key strength of Bayesian methods.
But modern algorithms for approximate Bayesian posterior inference often
sacrifice accurate posterior uncertainty estimation in the pursuit of
scalability. This work shows that previous Bayesian coreset construction
algorithms---which build a small, weighted subset of the data that approximates
the full dataset---are no exception. We demonstrate that these algorithms scale
the coreset log-likelihood suboptimally, resulting in underestimated posterior
uncertainty. To address this shortcoming, we develop greedy iterative geodesic
ascent (GIGA), a novel algorithm for Bayesian coreset construction that scales
the coreset log-likelihood optimally. GIGA provides geometric decay in
posterior approximation error as a function of coreset size, and maintains the
fast running time of its predecessors. The paper concludes with validation of
GIGA on both synthetic and real datasets, demonstrating that it reduces
posterior approximation error by orders of magnitude compared with previous
coreset constructions.
|
Shared Autonomy via Deep Reinforcement Learning | In shared autonomy, user input is combined with semi-autonomous control to
achieve a common goal. The goal is often unknown ex-ante, so prior work enables
agents to infer the goal from user input and assist with the task. Such methods
tend to assume some combination of knowledge of the dynamics of the
environment, the user's policy given their goal, and the set of possible goals
the user might target, which limits their application to real-world scenarios.
We propose a deep reinforcement learning framework for model-free shared
autonomy that lifts these assumptions. We use human-in-the-loop reinforcement
learning with neural network function approximation to learn an end-to-end
mapping from environmental observation and user input to agent action values,
with task reward as the only form of supervision. This approach poses the
challenge of following user commands closely enough to provide the user with
real-time action feedback and thereby ensure high-quality user input, but also
deviating from the user's actions when they are suboptimal. We balance these
two needs by discarding actions whose values fall below some threshold, then
selecting the remaining action closest to the user's input. Controlled studies
with users (n = 12) and synthetic pilots playing a video game, and a pilot
study with users (n = 4) flying a real quadrotor, demonstrate the ability of
our algorithm to assist users with real-time control tasks in which the agent
cannot directly access the user's private information through observations, but
receives a reward signal and user input that both depend on the user's intent.
The agent learns to assist the user without access to this private information,
implicitly inferring it from the user's input. This paper is a proof of concept
that illustrates the potential for deep reinforcement learning to enable
flexible and practical assistive systems.
|
Near-Optimal Coresets of Kernel Density Estimates | We construct near-optimal coresets for kernel density estimates for points in
$\mathbb{R}^d$ when the kernel is positive definite. Specifically we show a
polynomial time construction for a coreset of size $O(\sqrt{d}/\varepsilon\cdot
\sqrt{\log 1/\varepsilon} )$, and we show a near-matching lower bound of size
$\Omega(\min\{\sqrt{d}/\varepsilon, 1/\varepsilon^2\})$. When $d\geq
1/\varepsilon^2$, it is known that the size of coreset can be
$O(1/\varepsilon^2)$. The upper bound is a polynomial-in-$(1/\varepsilon)$
improvement when $d \in [3,1/\varepsilon^2)$ and the lower bound is the first
known lower bound to depend on $d$ for this problem. Moreover, the upper bound
restriction that the kernel is positive definite is significant in that it
applies to a wide-variety of kernels, specifically those most important for
machine learning. This includes kernels for information distances and the sinc
kernel which can be negative.
|
Training Generative Adversarial Networks via Primal-Dual Subgradient
Methods: A Lagrangian Perspective on GAN | We relate the minimax game of generative adversarial networks (GANs) to
finding the saddle points of the Lagrangian function for a convex optimization
problem, where the discriminator outputs and the distribution of generator
outputs play the roles of primal variables and dual variables, respectively.
This formulation shows the connection between the standard GAN training process
and the primal-dual subgradient methods for convex optimization. The inherent
connection does not only provide a theoretical convergence proof for training
GANs in the function space, but also inspires a novel objective function for
training. The modified objective function forces the distribution of generator
outputs to be updated along the direction according to the primal-dual
subgradient methods. A toy example shows that the proposed method is able to
resolve mode collapse, which in this case cannot be avoided by the standard GAN
or Wasserstein GAN. Experiments on both Gaussian mixture synthetic data and
real-world image datasets demonstrate the performance of the proposed method on
generating diverse samples.
|
Decomposition Methods with Deep Corrections for Reinforcement Learning | Decomposition methods have been proposed to approximate solutions to large
sequential decision making problems. In contexts where an agent interacts with
multiple entities, utility decomposition can be used to separate the global
objective into local tasks considering each individual entity independently. An
arbitrator is then responsible for combining the individual utilities and
selecting an action in real time to solve the global problem. Although these
techniques can perform well empirically, they rely on strong assumptions of
independence between the local tasks and sacrifice the optimality of the global
solution. This paper proposes an approach that improves upon such approximate
solutions by learning a correction term represented by a neural network. We
demonstrate this approach on a fisheries management problem where multiple
boats must coordinate to maximize their catch over time as well as on a
pedestrian avoidance problem for autonomous driving. In each problem,
decomposition methods can scale to multiple boats or pedestrians by using
strategies involving one entity. We verify empirically that the proposed
correction method significantly improves the decomposition method and
outperforms a policy trained on the full scale problem without utility
decomposition.
|
Mixed Link Networks | Basing on the analysis by revealing the equivalence of modern networks, we
find that both ResNet and DenseNet are essentially derived from the same "dense
topology", yet they only differ in the form of connection -- addition (dubbed
"inner link") vs. concatenation (dubbed "outer link"). However, both two forms
of connections have the superiority and insufficiency. To combine their
advantages and avoid certain limitations on representation learning, we present
a highly efficient and modularized Mixed Link Network (MixNet) which is
equipped with flexible inner link and outer link modules. Consequently, ResNet,
DenseNet and Dual Path Network (DPN) can be regarded as a special case of
MixNet, respectively. Furthermore, we demonstrate that MixNets can achieve
superior efficiency in parameter over the state-of-the-art architectures on
many competitive datasets like CIFAR-10/100, SVHN and ImageNet.
|
Decoding-History-Based Adaptive Control of Attention for Neural Machine
Translation | Attention-based sequence-to-sequence model has proved successful in Neural
Machine Translation (NMT). However, the attention without consideration of
decoding history, which includes the past information in the decoder and the
attention mechanism, often causes much repetition. To address this problem, we
propose the decoding-history-based Adaptive Control of Attention (ACA) for the
NMT model. ACA learns to control the attention by keeping track of the decoding
history and the current information with a memory vector, so that the model can
take the translated contents and the current information into consideration.
Experiments on Chinese-English translation and the English-Vietnamese
translation have demonstrated that our model significantly outperforms the
strong baselines. The analysis shows that our model is capable of generating
translation with less repetition and higher accuracy. The code will be
available at https://github.com/lancopku
|
Texygen: A Benchmarking Platform for Text Generation Models | We introduce Texygen, a benchmarking platform to support research on
open-domain text generation models. Texygen has not only implemented a majority
of text generation models, but also covered a set of metrics that evaluate the
diversity, the quality and the consistency of the generated texts. The Texygen
platform could help standardize the research on text generation and facilitate
the sharing of fine-tuned open-source implementations among researchers for
their work. As a consequence, this would help in improving the reproductivity
and reliability of future research work in text generation.
|
The steerable graph Laplacian and its application to filtering image
data-sets | In recent years, improvements in various image acquisition techniques gave
rise to the need for adaptive processing methods, aimed particularly for large
datasets corrupted by noise and deformations. In this work, we consider
datasets of images sampled from a low-dimensional manifold (i.e. an
image-valued manifold), where the images can assume arbitrary planar rotations.
To derive an adaptive and rotation-invariant framework for processing such
datasets, we introduce a graph Laplacian (GL)-like operator over the dataset,
termed ${\textit{steerable graph Laplacian}}$. Essentially, the steerable GL
extends the standard GL by accounting for all (infinitely-many) planar
rotations of all images. As it turns out, similarly to the standard GL, a
properly normalized steerable GL converges to the Laplace-Beltrami operator on
the low-dimensional manifold. However, the steerable GL admits an improved
convergence rate compared to the GL, where the improved convergence behaves as
if the intrinsic dimension of the underlying manifold is lower by one.
Moreover, it is shown that the steerable GL admits eigenfunctions of the form
of Fourier modes (along the orbits of the images' rotations) multiplied by
eigenvectors of certain matrices, which can be computed efficiently by the FFT.
For image datasets corrupted by noise, we employ a subset of these
eigenfunctions to "filter" the dataset via a Fourier-like filtering scheme,
essentially using all images and their rotations simultaneously. We demonstrate
our filtering framework by de-noising simulated single-particle cryo-EM image
datasets.
|
A Survey Of Methods For Explaining Black Box Models | In the last years many accurate decision support systems have been
constructed as black boxes, that is as systems that hide their internal logic
to the user. This lack of explanation constitutes both a practical and an
ethical issue. The literature reports many approaches aimed at overcoming this
crucial weakness sometimes at the cost of scarifying accuracy for
interpretability. The applications in which black box decision systems can be
used are various, and each approach is typically developed to provide a
solution for a specific problem and, as a consequence, delineating explicitly
or implicitly its own definition of interpretability and explanation. The aim
of this paper is to provide a classification of the main problems addressed in
the literature with respect to the notion of explanation and the type of black
box system. Given a problem definition, a black box type, and a desired
explanation this survey should help the researcher to find the proposals more
useful for his own work. The proposed classification of approaches to open
black box models should also be useful for putting the many research open
questions in perspective.
|
Improving Variational Encoder-Decoders in Dialogue Generation | Variational encoder-decoders (VEDs) have shown promising results in dialogue
generation. However, the latent variable distributions are usually approximated
by a much simpler model than the powerful RNN structure used for encoding and
decoding, yielding the KL-vanishing problem and inconsistent training
objective. In this paper, we separate the training step into two phases: The
first phase learns to autoencode discrete texts into continuous embeddings,
from which the second phase learns to generalize latent representations by
reconstructing the encoded embedding. In this case, latent variables are
sampled by transforming Gaussian noise through multi-layer perceptrons and are
trained with a separate VED model, which has the potential of realizing a much
more flexible distribution. We compare our model with current popular models
and the experiment demonstrates substantial improvement in both metric-based
and human evaluations.
|
Neural Network Detection of Data Sequences in Communication Systems | We consider detection based on deep learning, and show it is possible to
train detectors that perform well without any knowledge of the underlying
channel models. Moreover, when the channel model is known, we demonstrate that
it is possible to train detectors that do not require channel state information
(CSI). In particular, a technique we call a sliding bidirectional recurrent
neural network (SBRNN) is proposed for detection where, after training, the
detector estimates the data in real-time as the signal stream arrives at the
receiver. We evaluate this algorithm, as well as other neural network (NN)
architectures, using the Poisson channel model, which is applicable to both
optical and molecular communication systems. In addition, we also evaluate the
performance of this detection method applied to data sent over a molecular
communication platform, where the channel model is difficult to model
analytically. We show that SBRNN is computationally efficient, and can perform
detection under various channel conditions without knowing the underlying
channel model. We also demonstrate that the bit error rate (BER) performance of
the proposed SBRNN detector is better than that of a Viterbi detector with
imperfect CSI as well as that of other NN detectors that have been previously
proposed. Finally, we show that the SBRNN can perform well in rapidly changing
channels, where the coherence time is on the order of a single symbol duration.
|
Syst\`eme de traduction automatique statistique Anglais-Arabe | Machine translation (MT) is the process of translating text written in a
source language into text in a target language. In this article, we present our
English-Arabic statistical machine translation system. First, we present the
general process for setting up a statistical machine translation system, then
we describe the tools as well as the different corpora we used to build our MT
system. Our system was evaluated in terms of the BLUE score (24.51%)
|
On the Feasibility of Generic Deep Disaggregation for Single-Load
Extraction | Recently, and with the growing development of big energy datasets,
data-driven learning techniques began to represent a potential solution to the
energy disaggregation problem outperforming engineered and hand-crafted models.
However, most proposed deep disaggregation models are load-dependent in the
sense that either expert knowledge or a hyper-parameter optimization stage is
required prior to training and deployment (normally for each load category)
even upon acquisition and cleansing of aggregate and sub-metered data. In this
paper, we present a feasibility study on the development of a generic
disaggregation model based on data-driven learning. Specifically, we present a
generic deep disaggregation model capable of achieving state-of-art performance
in load monitoring for a variety of load categories. The developed model is
evaluated on the publicly available UK-DALE dataset with a moderately low
sampling frequency and various domestic loads.
|
DeepTravel: a Neural Network Based Travel Time Estimation Model with
Auxiliary Supervision | Estimating the travel time of a path is of great importance to smart urban
mobility. Existing approaches are either based on estimating the time cost of
each road segment which are not able to capture many cross-segment complex
factors, or designed heuristically in a non-learning-based way which fail to
utilize the existing abundant temporal labels of the data, i.e., the time stamp
of each trajectory point. In this paper, we leverage on new development of deep
neural networks and propose a novel auxiliary supervision model, namely
DeepTravel, that can automatically and effectively extract different features,
as well as make full use of the temporal labels of the trajectory data. We have
conducted comprehensive experiments on real datasets to demonstrate the
out-performance of DeepTravel over existing approaches.
|
FastNet | Inception and the Resnet family of Convolutional Neural Network
archi-tectures have broken records in the past few years, but recent state of
the art models have also incurred very high computational cost in terms of
training, inference and model size. Making the deployment of these models on
Edge devices, impractical. In light of this, we present a new novel
architecture that is designed for high computational efficiency on both GPUs
and CPUs, and is highly suited for deployment on Mobile Applications, Smart
Cameras, Iot devices and controllers as well as low cost drones. Our
architecture boasts competitive accuracies on standard Datasets even
out-performing the original Resnet. We present below the motivation for this
research, the architecture of the network, single test accuracies on CIFAR 10
and CIFAR 100 , a detailed comparison with other well-known architectures and
link to an implementation in Keras.
|
Granger-causal Attentive Mixtures of Experts: Learning Important
Features with Neural Networks | Knowledge of the importance of input features towards decisions made by
machine-learning models is essential to increase our understanding of both the
models and the underlying data. Here, we present a new approach to estimating
feature importance with neural networks based on the idea of distributing the
features of interest among experts in an attentive mixture of experts (AME).
AMEs use attentive gating networks trained with a Granger-causal objective to
learn to jointly produce accurate predictions as well as estimates of feature
importance in a single model. Our experiments show (i) that the feature
importance estimates provided by AMEs compare favourably to those provided by
state-of-the-art methods, (ii) that AMEs are significantly faster at estimating
feature importance than existing methods, and (iii) that the associations
discovered by AMEs are consistent with those reported by domain experts.
|
Automatic construction of Chinese herbal prescription from tongue image
via CNNs and auxiliary latent therapy topics | The tongue image provides important physical information of humans. It is of
great importance for diagnoses and treatments in clinical medicine. Herbal
prescriptions are simple, noninvasive and have low side effects. Thus, they are
widely applied in China. Studies on the automatic construction technology of
herbal prescriptions based on tongue images have great significance for deep
learning to explore the relevance of tongue images for herbal prescriptions, it
can be applied to healthcare services in mobile medical systems. In order to
adapt to the tongue image in a variety of photographic environments and
construct herbal prescriptions, a neural network framework for prescription
construction is designed. It includes single/double convolution channels and
fully connected layers. Furthermore, it proposes the auxiliary therapy topic
loss mechanism to model the therapy of Chinese doctors and alleviate the
interference of sparse output labels on the diversity of results. The
experiment use the real world tongue images and the corresponding prescriptions
and the results can generate prescriptions that are close to the real samples,
which verifies the feasibility of the proposed method for the automatic
construction of herbal prescriptions from tongue images. Also, it provides a
reference for automatic herbal prescription construction from more physical
information.
|
Automated dataset generation for image recognition using the example of
taxonomy | This master thesis addresses the subject of automatically generating a
dataset for image recognition, which takes a lot of time when being done
manually. As the thesis was written with motivation from the context of the
biodiversity workgroup at the City University of Applied Sciences Bremen, the
classification of taxonomic entries was chosen as an exemplary use case. In
order to automate the dataset creation, a prototype was conceptualized and
implemented after working out knowledge basics and analyzing requirements for
it. It makes use of an pre-trained abstract artificial intelligence which is
able to sort out images that do not contain the desired content. Subsequent to
the implementation and the automated dataset creation resulting from it, an
evaluation was performed. Other, manually collected datasets were compared to
the one the prototype produced in means of specifications and accuracy. The
results were more than satisfactory and showed that automatically generating a
dataset for image recognition is not only possible, but also might be a decent
alternative to spending time and money in doing this task manually. At the very
end of this work, an idea of how to use the principle of employing abstract
artificial intelligences for step-by-step classification of deeper taxonomic
layers in a productive system is presented and discussed.
|
Classification and Disease Localization in Histopathology Using Only
Global Labels: A Weakly-Supervised Approach | Analysis of histopathology slides is a critical step for many diagnoses, and
in particular in oncology where it defines the gold standard. In the case of
digital histopathological analysis, highly trained pathologists must review
vast whole-slide-images of extreme digital resolution ($100,000^2$ pixels)
across multiple zoom levels in order to locate abnormal regions of cells, or in
some cases single cells, out of millions. The application of deep learning to
this problem is hampered not only by small sample sizes, as typical datasets
contain only a few hundred samples, but also by the generation of ground-truth
localized annotations for training interpretable classification and
segmentation models. We propose a method for disease localization in the
context of weakly supervised learning, where only image-level labels are
available during training. Even without pixel-level annotations, we are able to
demonstrate performance comparable with models trained with strong annotations
on the Camelyon-16 lymph node metastases detection challenge. We accomplish
this through the use of pre-trained deep convolutional networks, feature
embedding, as well as learning via top instances and negative evidence, a
multiple instance learning technique from the field of semantic segmentation
and object detection.
|
A machine learning approach to reconstruction of heart surface
potentials from body surface potentials | Invasive cardiac catheterisation is a common procedure that is carried out
before surgical intervention. Yet, invasive cardiac diagnostics are full of
risks, especially for young children. Decades of research has been conducted on
the so called inverse problem of electrocardiography, which can be used to
reconstruct Heart Surface Potentials (HSPs) from Body Surface Potentials
(BSPs), for non-invasive diagnostics. State of the art solutions to the inverse
problem are unsatisfactory, since the inverse problem is known to be ill-posed.
In this paper we propose a novel approach to reconstructing HSPs from BSPs
using a Time-Delay Artificial Neural Network (TDANN). We first design the TDANN
architecture, and then develop an iterative search space algorithm to find the
parameters of the TDANN, which results in the best overall HSP prediction. We
use real-world recorded BSPs and HSPs from individuals suffering from serious
cardiac conditions to validate our TDANN. The results are encouraging, in that
coefficients obtained by correlating the predicted HSP with the recorded
patient' HSP approach ideal values.
|
Seismic-Net: A Deep Densely Connected Neural Network to Detect Seismic
Events | One of the risks of large-scale geologic carbon sequestration is the
potential migration of fluids out of the storage formations. Accurate and fast
detection of this fluids migration is not only important but also challenging,
due to the large subsurface uncertainty and complex governing physics.
Traditional leakage detection and monitoring techniques rely on geophysical
observations including seismic. However, the resulting accuracy of these
methods is limited because of indirect information they provide requiring
expert interpretation, therefore yielding in-accurate estimates of leakage
rates and locations. In this work, we develop a novel machine-learning
detection package, named "Seismic-Net", which is based on the deep densely
connected neural network. To validate the performance of our proposed leakage
detection method, we employ our method to a natural analog site at Chimay\'o,
New Mexico. The seismic events in the data sets are generated because of the
eruptions of geysers, which is due to the leakage of $\mathrm{CO}_\mathrm{2}$.
In particular, we demonstrate the efficacy of our Seismic-Net by formulating
our detection problem as an event detection problem with time series data. A
fixed-length window is slid throughout the time series data and we build a deep
densely connected network to classify each window to determine if a geyser
event is included. Through our numerical tests, we show that our model achieves
precision/recall as high as 0.889/0.923. Therefore, our Seismic-Net has a great
potential for detection of $\mathrm{CO}_\mathrm{2}$ leakage.
|
Universal Deep Neural Network Compression | In this paper, we investigate lossy compression of deep neural networks
(DNNs) by weight quantization and lossless source coding for memory-efficient
deployment. Whereas the previous work addressed non-universal scalar
quantization and entropy coding of DNN weights, we for the first time introduce
universal DNN compression by universal vector quantization and universal source
coding. In particular, we examine universal randomized lattice quantization of
DNNs, which randomizes DNN weights by uniform random dithering before lattice
quantization and can perform near-optimally on any source without relying on
knowledge of its probability distribution. Moreover, we present a method of
fine-tuning vector quantized DNNs to recover the performance loss after
quantization. Our experimental results show that the proposed universal DNN
compression scheme compresses the 32-layer ResNet (trained on CIFAR-10) and the
AlexNet (trained on ImageNet) with compression ratios of $47.1$ and $42.5$,
respectively.
|
From Game-theoretic Multi-agent Log Linear Learning to Reinforcement
Learning | The main focus of this paper is on enhancement of two types of game-theoretic
learning algorithms: log-linear learning and reinforcement learning. The
standard analysis of log-linear learning needs a highly structured environment,
i.e. strong assumptions about the game from an implementation perspective. In
this paper, we introduce a variant of log-linear learning that provides
asymptotic guarantees while relaxing the structural assumptions to include
synchronous updates and limitations in information available to the players. On
the other hand, model-free reinforcement learning is able to perform even under
weaker assumptions on players' knowledge about the environment and other
players' strategies. We propose a reinforcement algorithm that uses a
double-aggregation scheme in order to deepen players' insight about the
environment and constant learning step-size which achieves a higher convergence
rate. Numerical experiments are conducted to verify each algorithm's robustness
and performance.
|
Spectral Image Visualization Using Generative Adversarial Networks | Spectral images captured by satellites and radio-telescopes are analyzed to
obtain information about geological compositions distributions, distant asters
as well as undersea terrain. Spectral images usually contain tens to hundreds
of continuous narrow spectral bands and are widely used in various fields. But
the vast majority of those image signals are beyond the visible range, which
calls for special visualization technique. The visualizations of spectral
images shall convey as much information as possible from the original signal
and facilitate image interpretation. However, most of the existing visualizatio
methods display spectral images in false colors, which contradict with human's
experience and expectation. In this paper, we present a novel visualization
generative adversarial network (GAN) to display spectral images in natural
colors. To achieve our goal, we propose a loss function which consists of an
adversarial loss and a structure loss. The adversarial loss pushes our solution
to the natural image distribution using a discriminator network that is trained
to differentiate between false-color images and natural-color images. We also
use a cycle loss as the structure constraint to guarantee structure
consistency. Experimental results show that our method is able to generate
structure-preserved and natural-looking visualizations.
|
Machine Learning-Based Prototyping of Graphical User Interfaces for
Mobile Apps | It is common practice for developers of user-facing software to transform a
mock-up of a graphical user interface (GUI) into code. This process takes place
both at an application's inception and in an evolutionary context as GUI
changes keep pace with evolving features. Unfortunately, this practice is
challenging and time-consuming. In this paper, we present an approach that
automates this process by enabling accurate prototyping of GUIs via three
tasks: detection, classification, and assembly. First, logical components of a
GUI are detected from a mock-up artifact using either computer vision
techniques or mock-up metadata. Then, software repository mining, automated
dynamic analysis, and deep convolutional neural networks are utilized to
accurately classify GUI-components into domain-specific types (e.g.,
toggle-button). Finally, a data-driven, K-nearest-neighbors algorithm generates
a suitable hierarchical GUI structure from which a prototype application can be
automatically assembled. We implemented this approach for Android in a system
called ReDraw. Our evaluation illustrates that ReDraw achieves an average
GUI-component classification accuracy of 91% and assembles prototype
applications that closely mirror target mock-ups in terms of visual affinity
while exhibiting reasonable code structure. Interviews with industrial
practitioners illustrate ReDraw's potential to improve real development
workflows.
|
Improved Oracle Complexity of Variance Reduced Methods for Nonsmooth
Convex Stochastic Composition Optimization | We consider the nonsmooth convex composition optimization problem where the
objective is a composition of two finite-sum functions and analyze stochastic
compositional variance reduced gradient (SCVRG) methods for them. SCVRG and its
variants have recently drawn much attention given their edge over stochastic
compositional gradient descent (SCGD); but the theoretical analysis exclusively
assumes strong convexity of the objective, which excludes several important
examples such as Lasso, logistic regression, principle component analysis and
deep neural nets. In contrast, we prove non-asymptotic incremental first-order
oracle (IFO) complexity of SCVRG or its novel variants for nonsmooth convex
composition optimization and show that they are provably faster than SCGD and
gradient descent. More specifically, our method achieves the total IFO
complexity of $O\left((m+n)\log\left(1/\epsilon\right)+1/\epsilon^3\right)$
which improves that of $O\left(1/\epsilon^{3.5}\right)$ and
$O\left((m+n)/\sqrt{\epsilon}\right)$ obtained by SCGD and accelerated gradient
descent (AGD) respectively. Experimental results confirm that our methods
outperform several existing methods, e.g., SCGD and AGD, on sparse
mean-variance optimization problem.
|
Stochastic Deconvolutional Neural Network Ensemble Training on
Generative Pseudo-Adversarial Networks | The training of Generative Adversarial Networks is a difficult task mainly
due to the nature of the networks. One such issue is when the generator and
discriminator start oscillating, rather than converging to a fixed point.
Another case can be when one agent becomes more adept than the other which
results in the decrease of the other agent's ability to learn, reducing the
learning capacity of the system as a whole. Additionally, there exists the
problem of Mode Collapse which involves the generators output collapsing to a
single sample or a small set of similar samples. To train GANs a careful
selection of the architecture that is used along with a variety of other
methods to improve training. Even when applying these methods there is low
stability of training in relation to the parameters that are chosen. Stochastic
ensembling is suggested as a method for improving the stability while training
GANs.
|
Spectral Learning of Binomial HMMs for DNA Methylation Data | We consider learning parameters of Binomial Hidden Markov Models, which may
be used to model DNA methylation data. The standard algorithm for the problem
is EM, which is computationally expensive for sequences of the scale of the
mammalian genome. Recently developed spectral algorithms can learn parameters
of latent variable models via tensor decomposition, and are highly efficient
for large data. However, these methods have only been applied to categorial
HMMs, and the main challenge is how to extend them to Binomial HMMs while still
retaining computational efficiency. We address this challenge by introducing a
new feature-map based approach that exploits specific properties of Binomial
HMMs. We provide theoretical performance guarantees for our algorithm and
evaluate it on real DNA methylation data.
|
Cadre Modeling: Simultaneously Discovering Subpopulations and Predictive
Models | We consider the problem in regression analysis of identifying subpopulations
that exhibit different patterns of response, where each subpopulation requires
a different underlying model. Unlike statistical cohorts, these subpopulations
are not known a priori; thus, we refer to them as cadres. When the cadres and
their associated models are interpretable, modeling leads to insights about the
subpopulations and their associations with the regression target. We introduce
a discriminative model that simultaneously learns cadre assignment and
target-prediction rules. Sparsity-inducing priors are placed on the model
parameters, under which independent feature selection is performed for both the
cadre assignment and target-prediction processes. We learn models using
adaptive step size stochastic gradient descent, and we assess cadre quality
with bootstrapped sample analysis. We present simulated results showing that,
when the true clustering rule does not depend on the entire set of features,
our method significantly outperforms methods that learn subpopulation-discovery
and target-prediction rules separately. In a materials-by-design case study,
our model provides state-of-the-art prediction of polymer glass transition
temperature. Importantly, the method identifies cadres of polymers that respond
differently to structural perturbations, thus providing design insight for
targeting or avoiding specific transition temperature ranges. It identifies
chemically meaningful cadres, each with interpretable models. Further
experimental results show that cadre methods have generalization that is
competitive with linear and nonlinear regression models and can identify robust
subpopulations.
|
DeepHeart: Semi-Supervised Sequence Learning for Cardiovascular Risk
Prediction | We train and validate a semi-supervised, multi-task LSTM on 57,675
person-weeks of data from off-the-shelf wearable heart rate sensors, showing
high accuracy at detecting multiple medical conditions, including diabetes
(0.8451), high cholesterol (0.7441), high blood pressure (0.8086), and sleep
apnea (0.8298). We compare two semi-supervised train- ing methods,
semi-supervised sequence learning and heuristic pretraining, and show they
outperform hand-engineered biomarkers from the medical literature. We believe
our work suggests a new approach to patient risk stratification based on
cardiovascular risk scores derived from popular wearables such as Fitbit, Apple
Watch, or Android Wear.
|
A Spatial Mapping Algorithm with Applications in Deep Learning-Based
Structure Classification | Convolutional Neural Network (CNN)-based machine learning systems have made
breakthroughs in feature extraction and image recognition tasks in two
dimensions (2D). Although there is significant ongoing work to apply CNN
technology to domains involving complex 3D data, the success of such efforts
has been constrained, in part, by limitations in data representation
techniques. Most current approaches rely upon low-resolution 3D models,
strategic limitation of scope in the 3D space, or the application of lossy
projection techniques to allow for the use of 2D CNNs. To address this issue,
we present a mapping algorithm that converts 3D structures to 2D and 1D data
grids by mapping a traversal of a 3D space-filling curve to the traversal of
corresponding 2D and 1D curves. We explore the performance of 2D and 1D CNNs
trained on data encoded with our method versus comparable volumetric CNNs
operating upon raw 3D data from a popular benchmarking dataset. Our experiments
demonstrate that both 2D and 1D representations of 3D data generated via our
method preserve a significant proportion of the 3D data's features in forms
learnable by CNNs. Furthermore, we demonstrate that our method of encoding 3D
data into lower-dimensional representations allows for decreased CNN training
time cost, increased original 3D model rendering resolutions, and supports
increased numbers of data channels when compared to purely volumetric
approaches. This demonstration is accomplished in the context of a structural
biology classification task wherein we train 3D, 2D, and 1D CNNs on examples of
two homologous branches within the Ras protein family. The essential
contribution of this paper is the introduction of a dimensionality-reduction
method that may ease the application of powerful deep learning tools to domains
characterized by complex structural data.
|
Directly and Efficiently Optimizing Prediction Error and AUC of Linear
Classifiers | The predictive quality of machine learning models is typically measured in
terms of their (approximate) expected prediction error or the so-called Area
Under the Curve (AUC) for a particular data distribution. However, when the
models are constructed by the means of empirical risk minimization, surrogate
functions such as the logistic loss are optimized instead. This is done because
the empirical approximations of the expected error and AUC functions are
nonconvex and nonsmooth, and more importantly have zero derivative almost
everywhere. In this work, we show that in the case of linear predictors, and
under the assumption that the data has normal distribution, the expected error
and the expected AUC are not only smooth, but have closed form expressions,
which depend on the first and second moments of the normal distribution. Hence,
we derive derivatives of these two functions and use these derivatives in an
optimization algorithm to directly optimize the expected error and the AUC. In
the case of real data sets, the derivatives can be approximated using empirical
moments. We show that even when data is not normally distributed, computed
derivatives are sufficiently useful to render an efficient optimization method
and high quality solutions. Thus, we propose a gradient-based optimization
method for direct optimization of the prediction error and AUC. Moreover, the
per-iteration complexity of the proposed algorithm has no dependence on the
size of the data set, unlike those for optimizing logistic regression and all
other well known empirical risk minimization problems.
|
Learning One Convolutional Layer with Overlapping Patches | We give the first provably efficient algorithm for learning a one hidden
layer convolutional network with respect to a general class of (potentially
overlapping) patches. Additionally, our algorithm requires only mild conditions
on the underlying distribution. We prove that our framework captures commonly
used schemes from computer vision, including one-dimensional and
two-dimensional "patch and stride" convolutions.
Our algorithm-- $Convotron$ -- is inspired by recent work applying isotonic
regression to learning neural networks. Convotron uses a simple, iterative
update rule that is stochastic in nature and tolerant to noise (requires only
that the conditional mean function is a one layer convolutional network, as
opposed to the realizable setting). In contrast to gradient descent, Convotron
requires no special initialization or learning-rate tuning to converge to the
global optimum.
We also point out that learning one hidden convolutional layer with respect
to a Gaussian distribution and just $one$ disjoint patch $P$ (the other patches
may be arbitrary) is $easy$ in the following sense: Convotron can efficiently
recover the hidden weight vector by updating $only$ in the direction of $P$.
|
Predicting Hurricane Trajectories using a Recurrent Neural Network | Hurricanes are cyclones circulating about a defined center whose closed wind
speeds exceed 75 mph originating over tropical and subtropical waters. At
landfall, hurricanes can result in severe disasters. The accuracy of predicting
their trajectory paths is critical to reduce economic loss and save human
lives. Given the complexity and nonlinearity of weather data, a recurrent
neural network (RNN) could be beneficial in modeling hurricane behavior. We
propose the application of a fully connected RNN to predict the trajectory of
hurricanes. We employed the RNN over a fine grid to reduce typical truncation
errors. We utilized their latitude, longitude, wind speed, and pressure
publicly provided by the National Hurricane Center (NHC) to predict the
trajectory of a hurricane at 6-hour intervals. Results show that this proposed
technique is competitive to methods currently employed by the NHC and can
predict up to approximately 120 hours of hurricane path.
|
Semi-Amortized Variational Autoencoders | Amortized variational inference (AVI) replaces instance-specific local
inference with a global inference network. While AVI has enabled efficient
training of deep generative models such as variational autoencoders (VAE),
recent empirical work suggests that inference networks can produce suboptimal
variational parameters. We propose a hybrid approach, to use AVI to initialize
the variational parameters and run stochastic variational inference (SVI) to
refine them. Crucially, the local SVI procedure is itself differentiable, so
the inference network and generative model can be trained end-to-end with
gradient-based optimization. This semi-amortized approach enables the use of
rich generative models without experiencing the posterior-collapse phenomenon
common in training VAEs for problems like text generation. Experiments show
this approach outperforms strong autoregressive and variational baselines on
standard text and image datasets.
|
Intentional Control of Type I Error over Unconscious Data Distortion: a
Neyman-Pearson Approach to Text Classification | This paper addresses the challenges in classifying textual data obtained from
open online platforms, which are vulnerable to distortion. Most existing
classification methods minimize the overall classification error and may yield
an undesirably large type I error (relevant textual messages are classified as
irrelevant), particularly when available data exhibit an asymmetry between
relevant and irrelevant information. Data distortion exacerbates this situation
and often leads to fallacious prediction. To deal with inestimable data
distortion, we propose the use of the Neyman-Pearson (NP) classification
paradigm, which minimizes type II error under a user-specified type I error
constraint. Theoretically, we show that the NP oracle is unaffected by data
distortion when the class conditional distributions remain the same.
Empirically, we study a case of classifying posts about worker strikes obtained
from a leading Chinese microblogging platform, which are frequently prone to
extensive, unpredictable and inestimable censorship. We demonstrate that, even
though the training and test data are susceptible to different distortion and
therefore potentially follow different distributions, our proposed NP methods
control the type I error on test data at the targeted level. The methods and
implementation pipeline proposed in our case study are applicable to many other
problems involving data distortion.
|
Applying Cooperative Machine Learning to Speed Up the Annotation of
Social Signals in Large Multi-modal Corpora | Scientific disciplines, such as Behavioural Psychology, Anthropology and
recently Social Signal Processing are concerned with the systematic exploration
of human behaviour. A typical work-flow includes the manual annotation (also
called coding) of social signals in multi-modal corpora of considerable size.
For the involved annotators this defines an exhausting and time-consuming task.
In the article at hand we present a novel method and also provide the tools to
speed up the coding procedure. To this end, we suggest and evaluate the use of
Cooperative Machine Learning (CML) techniques to reduce manual labelling
efforts by combining the power of computational capabilities and human
intelligence. The proposed CML strategy starts with a small number of labelled
instances and concentrates on predicting local parts first. Afterwards, a
session-independent classification model is created to finish the remaining
parts of the database. Confidence values are computed to guide the manual
inspection and correction of the predictions. To bring the proposed approach
into application we introduce NOVA - an open-source tool for collaborative and
machine-aided annotations. In particular, it gives labellers immediate access
to CML strategies and directly provides visual feedback on the results. Our
experiments show that the proposed method has the potential to significantly
reduce human labelling efforts.
|
VISER: Visual Self-Regularization | In this work, we propose the use of large set of unlabeled images as a source
of regularization data for learning robust visual representation. Given a
visual model trained by a labeled dataset in a supervised fashion, we augment
our training samples by incorporating large number of unlabeled data and train
a semi-supervised model. We demonstrate that our proposed learning approach
leverages an abundance of unlabeled images and boosts the visual recognition
performance which alleviates the need to rely on large labeled datasets for
learning robust representation. To increment the number of image instances
needed to learn robust visual models in our approach, each labeled image
propagates its label to its nearest unlabeled image instances. These retrieved
unlabeled images serve as local perturbations of each labeled image to perform
Visual Self-Regularization (VISER). To retrieve such visual self regularizers,
we compute the cosine similarity in a semantic space defined by the penultimate
layer in a fully convolutional neural network. We use the publicly available
Yahoo Flickr Creative Commons 100M dataset as the source of our unlabeled image
set and propose a distributed approximate nearest neighbor algorithm to make
retrieval practical at that scale. Using the labeled instances and their
regularizer samples we show that we significantly improve object categorization
and localization performance on the MS COCO and Visual Genome datasets where
objects appear in context.
|
Recognition of Acoustic Events Using Masked Conditional Neural Networks | Automatic feature extraction using neural networks has accomplished
remarkable success for images, but for sound recognition, these models are
usually modified to fit the nature of the multi-dimensional temporal
representation of the audio signal in spectrograms. This may not efficiently
harness the time-frequency representation of the signal. The ConditionaL Neural
Network (CLNN) takes into consideration the interrelation between the temporal
frames, and the Masked ConditionaL Neural Network (MCLNN) extends upon the CLNN
by forcing a systematic sparseness over the network's weights using a binary
mask. The masking allows the network to learn about frequency bands rather than
bins, mimicking a filterbank used in signal transformations such as MFCC.
Additionally, the Mask is designed to consider various combinations of
features, which automates the feature hand-crafting process. We applied the
MCLNN for the Environmental Sound Recognition problem using the Urbansound8k,
YorNoise, ESC-10 and ESC-50 datasets. The MCLNN have achieved competitive
performance compared to state-of-the-art Convolutional Neural Networks and
hand-crafted attempts.
|
Tight Lower Bounds for Locally Differentially Private Selection | We prove a tight lower bound (up to constant factors) on the sample
complexity of any non-interactive local differentially private protocol for
optimizing a linear function over the simplex. This lower bound also implies a
tight lower bound (again, up to constant factors) on the sample complexity of
any non-interactive local differentially private protocol implementing the
exponential mechanism. These results reveal that any local protocol for these
problems has exponentially worse dependence on the dimension than corresponding
algorithms in the central model. Previously, Kasiviswanathan et al. (FOCS 2008)
proved an exponential separation between local and central model algorithms for
PAC learning the class of parity functions. In contrast, our lower bound are
quantitatively tight, apply to a simple and natural class of linear
optimization problems, and our techniques are arguably simpler.
|
Gradient conjugate priors and multi-layer neural networks | The paper deals with learning probability distributions of observed data by
artificial neural networks. We suggest a so-called gradient conjugate prior
(GCP) update appropriate for neural networks, which is a modification of the
classical Bayesian update for conjugate priors. We establish a connection
between the gradient conjugate prior update and the maximization of the
log-likelihood of the predictive distribution. Unlike for the Bayesian neural
networks, we use deterministic weights of neural networks, but rather assume
that the ground truth distribution is normal with unknown mean and variance and
learn by the neural networks the parameters of a prior (normal-gamma
distribution) for these unknown mean and variance. The update of the parameters
is done, using the gradient that, at each step, directs towards minimizing the
Kullback--Leibler divergence from the prior to the posterior distribution (both
being normal-gamma). We obtain a corresponding dynamical system for the prior's
parameters and analyze its properties. In particular, we study the limiting
behavior of all the prior's parameters and show how it differs from the case of
the classical full Bayesian update. The results are validated on synthetic and
real world data sets.
|
Geometry Score: A Method For Comparing Generative Adversarial Networks | One of the biggest challenges in the research of generative adversarial
networks (GANs) is assessing the quality of generated samples and detecting
various levels of mode collapse. In this work, we construct a novel measure of
performance of a GAN by comparing geometrical properties of the underlying data
manifold and the generated one, which provides both qualitative and
quantitative means for evaluation. Our algorithm can be applied to datasets of
an arbitrary nature and is not limited to visual data. We test the obtained
metric on various real-life models and datasets and demonstrate that our method
provides new insights into properties of GANs.
|
Biological Mechanisms for Learning: A Computational Model of Olfactory
Learning in the Manduca sexta Moth, with Applications to Neural Nets | The insect olfactory system, which includes the antennal lobe (AL), mushroom
body (MB), and ancillary structures, is a relatively simple neural system
capable of learning. Its structural features, which are widespread in
biological neural systems, process olfactory stimuli through a cascade of
networks where large dimension shifts occur from stage to stage and where
sparsity and randomness play a critical role in coding. Learning is partly
enabled by a neuromodulatory reward mechanism of octopamine stimulation of the
AL, whose increased activity induces rewiring of the MB through Hebbian
plasticity. Enforced sparsity in the MB focuses Hebbian growth on neurons that
are the most important for the representation of the learned odor. Based upon
current biophysical knowledge, we have constructed an end-to-end computational
model of the Manduca sexta moth olfactory system which includes the interaction
of the AL and MB under octopamine stimulation. Our model is able to robustly
learn new odors, and our simulations of integrate-and-fire neurons match the
statistical features of in-vivo firing rate data. From a biological
perspective, the model provides a valuable tool for examining the role of
neuromodulators, like octopamine, in learning, and gives insight into critical
interactions between sparsity, Hebbian growth, and stimulation during learning.
Our simulations also inform predictions about structural details of the
olfactory system that are not currently well-characterized. From a machine
learning perspective, the model yields bio-inspired mechanisms that are
potentially useful in constructing neural nets for rapid learning from very few
samples. These mechanisms include high-noise layers, sparse layers as noise
filters, and a biologically-plausible optimization method to train the network
based on octopamine stimulation, sparse layers, and Hebbian growth.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.