title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
End-to-End Offline Goal-Oriented Dialog Policy Learning via Policy
Gradient | Learning a goal-oriented dialog policy is generally performed offline with
supervised learning algorithms or online with reinforcement learning (RL).
Additionally, as companies accumulate massive quantities of dialog transcripts
between customers and trained human agents, encoder-decoder methods have gained
popularity as agent utterances can be directly treated as supervision without
the need for utterance-level annotations. However, one potential drawback of
such approaches is that they myopically generate the next agent utterance
without regard for dialog-level considerations. To resolve this concern, this
paper describes an offline RL method for learning from unannotated corpora that
can optimize a goal-oriented policy at both the utterance and dialog level. We
introduce a novel reward function and use both on-policy and off-policy policy
gradient to learn a policy offline without requiring online user interaction or
an explicit state space definition.
|
Per-Pixel Feedback for improving Semantic Segmentation | Semantic segmentation is the task of assigning a label to each pixel in the
image.In recent years, deep convolutional neural networks have been driving
advances in multiple tasks related to cognition. Although, DCNNs have resulted
in unprecedented visual recognition performances, they offer little
transparency. To understand how DCNN based models work at the task of semantic
segmentation, we try to analyze the DCNN models in semantic segmentation. We
try to find the importance of global image information for labeling pixels.
Based on the experiments on discriminative regions, and modeling of
fixations, we propose a set of new training loss functions for fine-tuning DCNN
based models. The proposed training regime has shown improvement in performance
of DeepLab Large FOV(VGG-16) Segmentation model for PASCAL VOC 2012 dataset.
However, further test remains to conclusively evaluate the benefits due to the
proposed loss functions across models, and data-sets.
Submitted in part fulfillment of the requirements for the degree of
Integrated Masters of Science in Applied Mathematics.
Update: Further Experiment showed minimal benefits.
Code Available [here](https://github.com/BardOfCodes/Seg-Unravel).
|
Columnar Database Techniques for Creating AI Features | Recent advances with in-memory columnar database techniques have increased
the performance of analytical queries on very large databases and data
warehouses. At the same time, advances in artificial intelligence (AI)
algorithms have increased the ability to analyze data. We use the term AI to
encompass both Deep Learning (DL or neural network) and Machine Learning (ML
aka Big Data analytics). Our exploration of the AI full stack has led us to a
cross-stack columnar database innovation that efficiently creates features for
AI analytics. The innovation is to create Augmented Dictionary Values (ADVs) to
add to existing columnar database dictionaries in order to increase the
efficiency of featurization by minimizing data movement and data duplication.
We show how various forms of featurization (feature selection, feature
extraction, and feature creation) can be efficiently calculated in a columnar
database. The full stack AI investigation has also led us to propose an
integrated columnar database and AI architecture. This architecture has
information flows and feedback loops to improve the whole analytics cycle
during multiple iterations of extracting data from the data sources,
featurization, and analysis.
|
Blind Multiclass Ensemble Classification | The rising interest in pattern recognition and data analytics has spurred the
development of innovative machine learning algorithms and tools. However, as
each algorithm has its strengths and limitations, one is motivated to
judiciously fuse multiple algorithms in order to find the "best" performing
one, for a given dataset. Ensemble learning aims at such high-performance
meta-algorithm, by combining the outputs from multiple algorithms. The present
work introduces a blind scheme for learning from ensembles of classifiers,
using a moment matching method that leverages joint tensor and matrix
factorization. Blind refers to the combiner who has no knowledge of the
ground-truth labels that each classifier has been trained on. A rigorous
performance analysis is derived and the proposed scheme is evaluated on
synthetic and real datasets.
|
CycleGAN, a Master of Steganography | CycleGAN (Zhu et al. 2017) is one recent successful approach to learn a
transformation between two image distributions. In a series of experiments, we
demonstrate an intriguing property of the model: CycleGAN learns to "hide"
information about a source image into the images it generates in a nearly
imperceptible, high-frequency signal. This trick ensures that the generator can
recover the original sample and thus satisfy the cyclic consistency
requirement, while the generated image remains realistic. We connect this
phenomenon with adversarial attacks by viewing CycleGAN's training procedure as
training a generator of adversarial examples and demonstrate that the cyclic
consistency loss causes CycleGAN to be especially vulnerable to adversarial
attacks.
|
Nonconvex Sparse Spectral Clustering by Alternating Direction Method of
Multipliers and Its Convergence Analysis | Spectral Clustering (SC) is a widely used data clustering method which first
learns a low-dimensional embedding $U$ of data by computing the eigenvectors of
the normalized Laplacian matrix, and then performs k-means on $U^\top$ to get
the final clustering result. The Sparse Spectral Clustering (SSC) method
extends SC with a sparse regularization on $UU^\top$ by using the block
diagonal structure prior of $UU^\top$ in the ideal case. However, encouraging
$UU^\top$ to be sparse leads to a heavily nonconvex problem which is
challenging to solve and the work (Lu, Yan, and Lin 2016) proposes a convex
relaxation in the pursuit of this aim indirectly. However, the convex
relaxation generally leads to a loose approximation and the quality of the
solution is not clear. This work instead considers to solve the nonconvex
formulation of SSC which directly encourages $UU^\top$ to be sparse. We propose
an efficient Alternating Direction Method of Multipliers (ADMM) to solve the
nonconvex SSC and provide the convergence guarantee. In particular, we prove
that the sequences generated by ADMM always exist a limit point and any limit
point is a stationary point. Our analysis does not impose any assumptions on
the iterates and thus is practical. Our proposed ADMM for nonconvex problems
allows the stepsize to be increasing but upper bounded, and this makes it very
efficient in practice. Experimental analysis on several real data sets verifies
the effectiveness of our method.
|
Coordinate Descent with Bandit Sampling | Coordinate descent methods usually minimize a cost function by updating a
random decision variable (corresponding to one coordinate) at a time. Ideally,
we would update the decision variable that yields the largest decrease in the
cost function. However, finding this coordinate would require checking all of
them, which would effectively negate the improvement in computational
tractability that coordinate descent is intended to afford. To address this, we
propose a new adaptive method for selecting a coordinate. First, we find a
lower bound on the amount the cost function decreases when a coordinate is
updated. We then use a multi-armed bandit algorithm to learn which coordinates
result in the largest lower bound by interleaving this learning with
conventional coordinate descent updates except that the coordinate is selected
proportionately to the expected decrease. We show that our approach improves
the convergence of coordinate descent methods both theoretically and
experimentally.
|
Artificial Neural Networks that Learn to Satisfy Logic Constraints | Logic-based problems such as planning, theorem proving, or puzzles, typically
involve combinatoric search and structured knowledge representation. Artificial
neural networks are very successful statistical learners, however, for many
years, they have been criticized for their weaknesses in representing and in
processing complex structured knowledge which is crucial for combinatoric
search and symbol manipulation. Two neural architectures are presented, which
can encode structured relational knowledge in neural activation, and store
bounded First Order Logic constraints in connection weights. Both architectures
learn to search for a solution that satisfies the constraints. Learning is done
by unsupervised practicing on problem instances from the same domain, in a way
that improves the network-solving speed. No teacher exists to provide answers
for the problem instances of the training and test sets. However, the domain
constraints are provided as prior knowledge to a loss function that measures
the degree of constraint violations. Iterations of activation calculation and
learning are executed until a solution that maximally satisfies the constraints
emerges on the output units. As a test case, block-world planning problems are
used to train networks that learn to plan in that domain, but the techniques
proposed could be used more generally as in integrating prior symbolic
knowledge with statistical learning
|
DeepWear: Adaptive Local Offloading for On-Wearable Deep Learning | Due to their on-body and ubiquitous nature, wearables can generate a wide
range of unique sensor data creating countless opportunities for deep learning
tasks. We propose DeepWear, a deep learning (DL) framework for wearable devices
to improve the performance and reduce the energy footprint. DeepWear
strategically offloads DL tasks from a wearable device to its paired handheld
device through local network. Compared to the remote-cloud-based offloading,
DeepWear requires no Internet connectivity, consumes less energy, and is robust
to privacy breach. DeepWear provides various novel techniques such as
context-aware offloading, strategic model partition, and pipelining support to
efficiently utilize the processing capacity from nearby paired handhelds.
Deployed as a user-space library, DeepWear offers developer-friendly APIs that
are as simple as those in traditional DL libraries such as TensorFlow. We have
implemented DeepWear on the Android OS and evaluated it on COTS smartphones and
smartwatches with real DL models. DeepWear brings up to 5.08X and 23.0X
execution speedup, as well as 53.5% and 85.5% energy saving compared to
wearable-only and handheld-only strategies, respectively.
|
A Class of Logistic Functions for Approximating State-Inclusive Koopman
Operators | An outstanding challenge in nonlinear systems theory is identification or
learning of a given nonlinear system's Koopman operator directly from data or
models. Advances in extended dynamic mode decomposition approaches and machine
learning methods have enabled data-driven discovery of Koopman operators, for
both continuous and discrete-time systems. Since Koopman operators are often
infinite-dimensional, they are approximated in practice using
finite-dimensional systems. The fidelity and convergence of a given
finite-dimensional Koopman approximation is a subject of ongoing research. In
this paper we introduce a class of Koopman observable functions that confer an
approximate closure property on their corresponding finite-dimensional
approximations of the Koopman operator. We derive error bounds for the fidelity
of this class of observable functions, as well as identify two key learning
parameters which can be used to tune performance. We illustrate our approach on
two classical nonlinear system models: the Van Der Pol oscillator and the
bistable toggle switch.
|
On Adaptive Estimation for Dynamic Bernoulli Bandits | The multi-armed bandit (MAB) problem is a classic example of the
exploration-exploitation dilemma. It is concerned with maximising the total
rewards for a gambler by sequentially pulling an arm from a multi-armed slot
machine where each arm is associated with a reward distribution. In static
MABs, the reward distributions do not change over time, while in dynamic MABs,
each arm's reward distribution can change, and the optimal arm can switch over
time. Motivated by many real applications where rewards are binary, we focus on
dynamic Bernoulli bandits. Standard methods like $\epsilon$-Greedy and Upper
Confidence Bound (UCB), which rely on the sample mean estimator, often fail to
track changes in the underlying reward for dynamic problems. In this paper, we
overcome the shortcoming of slow response to change by deploying adaptive
estimation in the standard methods and propose a new family of algorithms,
which are adaptive versions of $\epsilon$-Greedy, UCB, and Thompson sampling.
These new methods are simple and easy to implement. Moreover, they do not
require any prior knowledge about the dynamic reward process, which is
important for real applications. We examine the new algorithms numerically in
different scenarios and the results show solid improvements of our algorithms
in dynamic environments.
|
Wild Patterns: Ten Years After the Rise of Adversarial Machine Learning | Learning-based pattern classifiers, including deep networks, have shown
impressive performance in several application domains, ranging from computer
vision to cybersecurity. However, it has also been shown that adversarial input
perturbations carefully crafted either at training or at test time can easily
subvert their predictions. The vulnerability of machine learning to such wild
patterns (also referred to as adversarial examples), along with the design of
suitable countermeasures, have been investigated in the research field of
adversarial machine learning. In this work, we provide a thorough overview of
the evolution of this research area over the last ten years and beyond,
starting from pioneering, earlier work on the security of non-deep learning
algorithms up to more recent work aimed to understand the security properties
of deep learning algorithms, in the context of computer vision and
cybersecurity tasks. We report interesting connections between these
apparently-different lines of work, highlighting common misconceptions related
to the security evaluation of machine-learning algorithms. We review the main
threat models and attacks defined to this end, and discuss the main limitations
of current work, along with the corresponding future challenges towards the
design of more secure learning algorithms.
|
Music Transcription by Deep Learning with Data and "Artificial Semantic"
Augmentation | In this progress paper the previous results of the single note recognition by
deep learning are presented. The several ways for data augmentation and
"artificial semantic" augmentation are proposed to enhance efficiency of deep
learning approaches for monophonic and polyphonic note recognition by increase
of dimensions of training data, their lossless and lossy transformations.
|
Neumann Optimizer: A Practical Optimization Algorithm for Deep Neural
Networks | Progress in deep learning is slowed by the days or weeks it takes to train
large models. The natural solution of using more hardware is limited by
diminishing returns, and leads to inefficient use of additional resources. In
this paper, we present a large batch, stochastic optimization algorithm that is
both faster than widely used algorithms for fixed amounts of computation, and
also scales up substantially better as more computational resources become
available. Our algorithm implicitly computes the inverse Hessian of each
mini-batch to produce descent directions; we do so without either an explicit
approximation to the Hessian or Hessian-vector products. We demonstrate the
effectiveness of our algorithm by successfully training large ImageNet models
(Inception-V3, Resnet-50, Resnet-101 and Inception-Resnet-V2) with mini-batch
sizes of up to 32000 with no loss in validation error relative to current
baselines, and no increase in the total number of steps. At smaller mini-batch
sizes, our optimizer improves the validation error in these models by 0.8-0.9%.
Alternatively, we can trade off this accuracy to reduce the number of training
steps needed by roughly 10-30%. Our work is practical and easily usable by
others -- only one hyperparameter (learning rate) needs tuning, and
furthermore, the algorithm is as computationally cheap as the commonly used
Adam optimizer.
|
Assumed Density Filtering Q-learning | While off-policy temporal difference (TD) methods have widely been used in
reinforcement learning due to their efficiency and simple implementation, their
Bayesian counterparts have not been utilized as frequently. One reason is that
the non-linear max operation in the Bellman optimality equation makes it
difficult to define conjugate distributions over the value functions. In this
paper, we introduce a novel Bayesian approach to off-policy TD methods, called
as ADFQ, which updates beliefs on state-action values, Q, through an online
Bayesian inference method known as Assumed Density Filtering. We formulate an
efficient closed-form solution for the value update by approximately estimating
analytic parameters of the posterior of the Q-beliefs. Uncertainty measures in
the beliefs not only are used in exploration but also provide a natural
regularization for the value update considering all next available actions.
ADFQ converges to Q-learning as the uncertainty measures of the Q-beliefs
decrease and improves common drawbacks of other Bayesian RL algorithms such as
computational complexity. We extend ADFQ with a neural network. Our empirical
results demonstrate that ADFQ outperforms comparable algorithms on various
Atari 2600 games, with drastic improvements in highly stochastic domains or
domains with a large action space.
|
Bayesian Joint Matrix Decomposition for Data Integration with
Heterogeneous Noise | Matrix decomposition is a popular and fundamental approach in machine
learning and data mining. It has been successfully applied into various fields.
Most matrix decomposition methods focus on decomposing a data matrix from one
single source. However, it is common that data are from different sources with
heterogeneous noise. A few of matrix decomposition methods have been extended
for such multi-view data integration and pattern discovery. While only few
methods were designed to consider the heterogeneity of noise in such multi-view
data for data integration explicitly. To this end, we propose a joint matrix
decomposition framework (BJMD), which models the heterogeneity of noise by
Gaussian distribution in a Bayesian framework. We develop two algorithms to
solve this model: one is a variational Bayesian inference algorithm, which
makes full use of the posterior distribution; and another is a maximum a
posterior algorithm, which is more scalable and can be easily paralleled.
Extensive experiments on synthetic and real-world datasets demonstrate that
BJMD considering the heterogeneity of noise is superior or competitive to the
state-of-the-art methods.
|
Variational auto-encoding of protein sequences | Proteins are responsible for the most diverse set of functions in biology.
The ability to extract information from protein sequences and to predict the
effects of mutations is extremely valuable in many domains of biology and
medicine. However the mapping between protein sequence and function is complex
and poorly understood. Here we present an embedding of natural protein
sequences using a Variational Auto-Encoder and use it to predict how mutations
affect protein function. We use this unsupervised approach to cluster natural
variants and learn interactions between sets of positions within a protein.
This approach generally performs better than baseline methods that consider no
interactions within sequences, and in some cases better than the
state-of-the-art approaches that use the inverse-Potts model. This generative
model can be used to computationally guide exploration of protein sequence
space and to better inform rational and automatic protein design.
|
Peephole: Predicting Network Performance Before Training | The quest for performant networks has been a significant force that drives
the advancements of deep learning in recent years. While rewarding, improving
network design has never been an easy journey. The large design space combined
with the tremendous cost required for network training poses a major obstacle
to this endeavor. In this work, we propose a new approach to this problem,
namely, predicting the performance of a network before training, based on its
architecture. Specifically, we develop a unified way to encode individual
layers into vectors and bring them together to form an integrated description
via LSTM. Taking advantage of the recurrent network's strong expressive power,
this method can reliably predict the performances of various network
architectures. Our empirical studies showed that it not only achieved accurate
predictions but also produced consistent rankings across datasets -- a key
desideratum in performance prediction.
|
NAG: Network for Adversary Generation | Adversarial perturbations can pose a serious threat for deploying machine
learning systems. Recent works have shown existence of image-agnostic
perturbations that can fool classifiers over most natural images. Existing
methods present optimization approaches that solve for a fooling objective with
an imperceptibility constraint to craft the perturbations. However, for a given
classifier, they generate one perturbation at a time, which is a single
instance from the manifold of adversarial perturbations. Also, in order to
build robust models, it is essential to explore the manifold of adversarial
perturbations. In this paper, we propose for the first time, a generative
approach to model the distribution of adversarial perturbations. The
architecture of the proposed model is inspired from that of GANs and is trained
using fooling and diversity objectives. Our trained generator network attempts
to capture the distribution of adversarial perturbations for a given classifier
and readily generates a wide variety of such perturbations. Our experimental
evaluation demonstrates that perturbations crafted by our model (i) achieve
state-of-the-art fooling rates, (ii) exhibit wide variety and (iii) deliver
excellent cross model generalizability. Our work can be deemed as an important
step in the process of inferring about the complex manifolds of adversarial
perturbations.
|
Cost-Sensitive Approach to Batch Size Adaptation for Gradient Descent | In this paper, we propose a novel approach to automatically determine the
batch size in stochastic gradient descent methods. The choice of the batch size
induces a trade-off between the accuracy of the gradient estimate and the cost
in terms of samples of each update. We propose to determine the batch size by
optimizing the ratio between a lower bound to a linear or quadratic Taylor
approximation of the expected improvement and the number of samples used to
estimate the gradient. The performance of the proposed approach is empirically
compared with related methods on popular classification tasks.
The work was presented at the NIPS workshop on Optimizing the Optimizers.
Barcelona, Spain, 2016.
|
Reinforced dynamics for enhanced sampling in large atomic and molecular
systems | A new approach for efficiently exploring the configuration space and
computing the free energy of large atomic and molecular systems is proposed,
motivated by an analogy with reinforcement learning. There are two major
components in this new approach. Like metadynamics, it allows for an efficient
exploration of the configuration space by adding an adaptively computed biasing
potential to the original dynamics. Like deep reinforcement learning, this
biasing potential is trained on the fly using deep neural networks, with data
collected judiciously from the exploration and an uncertainty indicator from
the neural network model playing the role of the reward function.
Parameterization using neural networks makes it feasible to handle cases with a
large set of collective variables. This has the potential advantage that
selecting precisely the right set of collective variables has now become less
critical for capturing the structural transformations of the system. The method
is illustrated by studying the full-atom, explicit solvent models of alanine
dipeptide and tripeptide, as well as the system of a polyalanine-10 molecule
with 20 collective variables.
|
Capsule Network Performance on Complex Data | In recent years, convolutional neural networks (CNN) have played an important
role in the field of deep learning. Variants of CNN's have proven to be very
successful in classification tasks across different domains. However, there are
two big drawbacks to CNN's: their failure to take into account of important
spatial hierarchies between features, and their lack of rotational invariance.
As long as certain key features of an object are present in the test data,
CNN's classify the test data as the object, disregarding features' relative
spatial orientation to each other. This causes false positives. The lack of
rotational invariance in CNN's would cause the network to incorrectly assign
the object another label, causing false negatives. To address this concern,
Hinton et al. propose a novel type of neural network using the concept of
capsules in a recent paper. With the use of dynamic routing and reconstruction
regularization, the capsule network model would be both rotation invariant and
spatially aware. The capsule network has shown its potential by achieving a
state-of-the-art result of 0.25% test error on MNIST without data augmentation
such as rotation and scaling, better than the previous baseline of 0.39%. To
further test out the application of capsule networks on data with higher
dimensionality, we attempt to find the best set of configurations that yield
the optimal test error on CIFAR10 dataset.
|
A General Memory-Bounded Learning Algorithm | Designing bounded-memory algorithms is becoming increasingly important
nowadays. Previous works studying bounded-memory algorithms focused on proving
impossibility results, while the design of bounded-memory algorithms was left
relatively unexplored. To remedy this situation, in this work we design a
general bounded-memory learning algorithm, when the underlying distribution is
known. The core idea of the algorithm is not to save the exact example
received, but only a few important bits that give sufficient information. This
algorithm applies to any hypothesis class that has an "anti-mixing" property.
This paper complements previous works on unlearnability with bounded memory and
provides a step towards a full characterization of bounded-memory learning.
|
An Architecture Combining Convolutional Neural Network (CNN) and Support
Vector Machine (SVM) for Image Classification | Convolutional neural networks (CNNs) are similar to "ordinary" neural
networks in the sense that they are made up of hidden layers consisting of
neurons with "learnable" parameters. These neurons receive inputs, performs a
dot product, and then follows it with a non-linearity. The whole network
expresses the mapping between raw image pixels and their class scores.
Conventionally, the Softmax function is the classifier used at the last layer
of this network. However, there have been studies (Alalshekmubarak and Smith,
2013; Agarap, 2017; Tang, 2013) conducted to challenge this norm. The cited
studies introduce the usage of linear support vector machine (SVM) in an
artificial neural network architecture. This project is yet another take on the
subject, and is inspired by (Tang, 2013). Empirical data has shown that the
CNN-SVM model was able to achieve a test accuracy of ~99.04% using the MNIST
dataset (LeCun, Cortes, and Burges, 2010). On the other hand, the CNN-Softmax
was able to achieve a test accuracy of ~99.23% using the same dataset. Both
models were also tested on the recently-published Fashion-MNIST dataset (Xiao,
Rasul, and Vollgraf, 2017), which is suppose to be a more difficult image
classification dataset than MNIST (Zalandoresearch, 2017). This proved to be
the case as CNN-SVM reached a test accuracy of ~90.72%, while the CNN-Softmax
reached a test accuracy of ~91.86%. The said results may be improved if data
preprocessing techniques were employed on the datasets, and if the base CNN
model was a relatively more sophisticated than the one used in this study.
|
DGCNN: Disordered Graph Convolutional Neural Network Based on the
Gaussian Mixture Model | Convolutional neural networks (CNNs) can be applied to graph similarity
matching, in which case they are called graph CNNs. Graph CNNs are attracting
increasing attention due to their effectiveness and efficiency. However, the
existing convolution approaches focus only on regular data forms and require
the transfer of the graph or key node neighborhoods of the graph into the same
fixed form. During this transfer process, structural information of the graph
can be lost, and some redundant information can be incorporated. To overcome
this problem, we propose the disordered graph convolutional neural network
(DGCNN) based on the mixed Gaussian model, which extends the CNN by adding a
preprocessing layer called the disordered graph convolutional layer (DGCL). The
DGCL uses a mixed Gaussian function to realize the mapping between the
convolution kernel and the nodes in the neighborhood of the graph. The output
of the DGCL is the input of the CNN. We further implement a
backward-propagation optimization process of the convolutional layer by which
we incorporate the feature-learning model of the irregular node neighborhood
structure into the network. Thereafter, the optimization of the convolution
kernel becomes part of the neural network learning process. The DGCNN can
accept arbitrary scaled and disordered neighborhood graph structures as the
receptive fields of CNNs, which reduces information loss during graph
transformation. Finally, we perform experiments on multiple standard graph
datasets. The results show that the proposed method outperforms the
state-of-the-art methods in graph classification and retrieval.
|
Shape optimization in laminar flow with a label-guided variational
autoencoder | Computational design optimization in fluid dynamics usually requires to solve
non-linear partial differential equations numerically. In this work, we explore
a Bayesian optimization approach to minimize an object's drag coefficient in
laminar flow based on predicting drag directly from the object shape. Jointly
training an architecture combining a variational autoencoder mapping shapes to
latent representations and Gaussian process regression allows us to generate
improved shapes in the two dimensional case we consider.
|
Gradient Normalization & Depth Based Decay For Deep Learning | In this paper we introduce a novel method of gradient normalization and decay
with respect to depth. Our method leverages the simple concept of normalizing
all gradients in a deep neural network, and then decaying said gradients with
respect to their depth in the network. Our proposed normalization and decay
techniques can be used in conjunction with most current state of the art
optimizers and are a very simple addition to any network. This method, although
simple, showed improvements in convergence time on state of the art networks
such as DenseNet and ResNet on image classification tasks, as well as on an
LSTM for natural language processing tasks.
|
Robust Deep Reinforcement Learning with Adversarial Attacks | This paper proposes adversarial attacks for Reinforcement Learning (RL) and
then improves the robustness of Deep Reinforcement Learning algorithms (DRL) to
parameter uncertainties with the help of these attacks. We show that even a
naively engineered attack successfully degrades the performance of DRL
algorithm. We further improve the attack using gradient information of an
engineered loss function which leads to further degradation in performance.
These attacks are then leveraged during training to improve the robustness of
RL within robust control framework. We show that this adversarial training of
DRL algorithms like Deep Double Q learning and Deep Deterministic Policy
Gradients leads to significant increase in robustness to parameter variations
for RL benchmarks such as Cart-pole, Mountain Car, Hopper and Half Cheetah
environment.
|
DeePMD-kit: A deep learning package for many-body potential energy
representation and molecular dynamics | Recent developments in many-body potential energy representation via deep
learning have brought new hopes to addressing the accuracy-versus-efficiency
dilemma in molecular simulations. Here we describe DeePMD-kit, a package
written in Python/C++ that has been designed to minimize the effort required to
build deep learning based representation of potential energy and force field
and to perform molecular dynamics. Potential applications of DeePMD-kit span
from finite molecules to extended systems and from metallic systems to
chemically bonded systems. DeePMD-kit is interfaced with TensorFlow, one of the
most popular deep learning frameworks, making the training process highly
automatic and efficient. On the other end, DeePMD-kit is interfaced with
high-performance classical molecular dynamics and quantum (path-integral)
molecular dynamics packages, i.e., LAMMPS and the i-PI, respectively. Thus,
upon training, the potential energy and force field models can be used to
perform efficient molecular simulations for different purposes. As an example
of the many potential applications of the package, we use DeePMD-kit to learn
the interatomic potential energy and forces of a water model using data
obtained from density functional theory. We demonstrate that the resulted
molecular dynamics model reproduces accurately the structural information
contained in the original model.
|
NestedNet: Learning Nested Sparse Structures in Deep Neural Networks | Recently, there have been increasing demands to construct compact deep
architectures to remove unnecessary redundancy and to improve the inference
speed. While many recent works focus on reducing the redundancy by eliminating
unneeded weight parameters, it is not possible to apply a single deep
architecture for multiple devices with different resources. When a new device
or circumstantial condition requires a new deep architecture, it is necessary
to construct and train a new network from scratch. In this work, we propose a
novel deep learning framework, called a nested sparse network, which exploits
an n-in-1-type nested structure in a neural network. A nested sparse network
consists of multiple levels of networks with a different sparsity ratio
associated with each level, and higher level networks share parameters with
lower level networks to enable stable nested learning. The proposed framework
realizes a resource-aware versatile architecture as the same network can meet
diverse resource requirements. Moreover, the proposed nested network can learn
different forms of knowledge in its internal networks at different levels,
enabling multiple tasks using a single network, such as coarse-to-fine
hierarchical classification. In order to train the proposed nested sparse
network, we propose efficient weight connection learning and channel and layer
scheduling strategies. We evaluate our network in multiple tasks, including
adaptive deep compression, knowledge distillation, and learning class
hierarchy, and demonstrate that nested sparse networks perform competitively,
but more efficiently, compared to existing methods.
|
On Quadratic Penalties in Elastic Weight Consolidation | Elastic weight consolidation (EWC, Kirkpatrick et al, 2017) is a novel
algorithm designed to safeguard against catastrophic forgetting in neural
networks. EWC can be seen as an approximation to Laplace propagation (Eskin et
al, 2004), and this view is consistent with the motivation given by Kirkpatrick
et al (2017). In this note, I present an extended derivation that covers the
case when there are more than two tasks. I show that the quadratic penalties in
EWC are inconsistent with this derivation and might lead to double-counting
data from earlier tasks.
|
Generalized Zero-Shot Learning via Synthesized Examples | We present a generative framework for generalized zero-shot learning where
the training and test classes are not necessarily disjoint. Built upon a
variational autoencoder based architecture, consisting of a probabilistic
encoder and a probabilistic conditional decoder, our model can generate novel
exemplars from seen/unseen classes, given their respective class attributes.
These exemplars can subsequently be used to train any off-the-shelf
classification model. One of the key aspects of our encoder-decoder
architecture is a feedback-driven mechanism in which a discriminator (a
multivariate regressor) learns to map the generated exemplars to the
corresponding class attribute vectors, leading to an improved generator. Our
model's ability to generate and leverage examples from unseen classes to train
the classification model naturally helps to mitigate the bias towards
predicting seen classes in generalized zero-shot learning settings. Through a
comprehensive set of experiments, we show that our model outperforms several
state-of-the-art methods, on several benchmark datasets, for both standard as
well as generalized zero-shot learning.
|
DeepConfig: Automating Data Center Network Topologies Management with
Machine Learning | In recent years, many techniques have been developed to improve the
performance and efficiency of data center networks. While these techniques
provide high accuracy, they are often designed using heuristics that leverage
domain-specific properties of the workload or hardware.
In this vision paper, we argue that many data center networking techniques,
e.g., routing, topology augmentation, energy savings, with diverse goals
actually share design and architectural similarity. We present a design for
developing general intermediate representations of network topologies using
deep learning that is amenable to solving classes of data center problems. We
develop a framework, DeepConfig, that simplifies the processing of configuring
and training deep learning agents that use the intermediate representation to
learns different tasks. To illustrate the strength of our approach, we
configured, implemented, and evaluated a DeepConfig-Agent that tackles the data
center topology augmentation problem. Our initial results are promising ---
DeepConfig performs comparably to the optimal.
|
Learning Modality-Invariant Representations for Speech and Images | In this paper, we explore the unsupervised learning of a semantic embedding
space for co-occurring sensory inputs. Specifically, we focus on the task of
learning a semantic vector space for both spoken and handwritten digits using
the TIDIGITs and MNIST datasets. Current techniques encode image and
audio/textual inputs directly to semantic embeddings. In contrast, our
technique maps an input to the mean and log variance vectors of a diagonal
Gaussian from which sample semantic embeddings are drawn. In addition to
encouraging semantic similarity between co-occurring inputs,our loss function
includes a regularization term borrowed from variational autoencoders (VAEs)
which drives the posterior distributions over embeddings to be unit Gaussian.
We can use this regularization term to filter out modality information while
preserving semantic information. We speculate this technique may be more
broadly applicable to other areas of cross-modality/domain information
retrieval and transfer learning.
|
MINOS: Multimodal Indoor Simulator for Navigation in Complex
Environments | We present MINOS, a simulator designed to support the development of
multisensory models for goal-directed navigation in complex indoor
environments. The simulator leverages large datasets of complex 3D environments
and supports flexible configuration of multimodal sensor suites. We use MINOS
to benchmark deep-learning-based navigation methods, to analyze the influence
of environmental complexity on navigation performance, and to carry out a
controlled study of multimodality in sensorimotor learning. The experiments
show that current deep reinforcement learning approaches fail in large
realistic environments. The experiments also indicate that multimodality is
beneficial in learning to navigate cluttered scenes. MINOS is released
open-source to the research community at http://minosworld.org . A video that
shows MINOS can be found at https://youtu.be/c0mL9K64q84
|
StrassenNets: Deep Learning with a Multiplication Budget | A large fraction of the arithmetic operations required to evaluate deep
neural networks (DNNs) consists of matrix multiplications, in both convolution
and fully connected layers. We perform end-to-end learning of low-cost
approximations of matrix multiplications in DNN layers by casting matrix
multiplications as 2-layer sum-product networks (SPNs) (arithmetic circuits)
and learning their (ternary) edge weights from data. The SPNs disentangle
multiplication and addition operations and enable us to impose a budget on the
number of multiplication operations. Combining our method with knowledge
distillation and applying it to image classification DNNs (trained on ImageNet)
and language modeling DNNs (using LSTMs), we obtain a first-of-a-kind reduction
in number of multiplications (over 99.5%) while maintaining the predictive
performance of the full-precision models. Finally, we demonstrate that the
proposed framework is able to rediscover Strassen's matrix multiplication
algorithm, learning to multiply $2 \times 2$ matrices using only 7
multiplications instead of 8.
|
Saving Gradient and Negative Curvature Computations: Finding Local
Minima More Efficiently | We propose a family of nonconvex optimization algorithms that are able to
save gradient and negative curvature computations to a large extent, and are
guaranteed to find an approximate local minimum with improved runtime
complexity. At the core of our algorithms is the division of the entire domain
of the objective function into small and large gradient regions: our algorithms
only perform gradient descent based procedure in the large gradient region, and
only perform negative curvature descent in the small gradient region. Our novel
analysis shows that the proposed algorithms can escape the small gradient
region in only one negative curvature descent step whenever they enter it, and
thus they only need to perform at most $N_{\epsilon}$ negative curvature
direction computations, where $N_{\epsilon}$ is the number of times the
algorithms enter small gradient regions. For both deterministic and stochastic
settings, we show that the proposed algorithms can potentially beat the
state-of-the-art local minima finding algorithms. For the finite-sum setting,
our algorithm can also outperform the best algorithm in a certain regime.
|
Eye In-Painting with Exemplar Generative Adversarial Networks | This paper introduces a novel approach to in-painting where the identity of
the object to remove or change is preserved and accounted for at inference
time: Exemplar GANs (ExGANs). ExGANs are a type of conditional GAN that utilize
exemplar information to produce high-quality, personalized in painting results.
We propose using exemplar information in the form of a reference image of the
region to in-paint, or a perceptual code describing that object. Unlike
previous conditional GAN formulations, this extra information can be inserted
at multiple points within the adversarial network, thus increasing its
descriptive power. We show that ExGANs can produce photo-realistic personalized
in-painting results that are both perceptually and semantically plausible by
applying them to the task of closed to-open eye in-painting in natural
pictures. A new benchmark dataset is also introduced for the task of eye
in-painting for future comparisons.
|
Training Ensembles to Detect Adversarial Examples | We propose a new ensemble method for detecting and classifying adversarial
examples generated by state-of-the-art attacks, including DeepFool and C&W. Our
method works by training the members of an ensemble to have low classification
error on random benign examples while simultaneously minimizing agreement on
examples outside the training distribution. We evaluate on both MNIST and
CIFAR-10, against oblivious and both white- and black-box adversaries.
|
Cavs: A Vertex-centric Programming Interface for Dynamic Neural Networks | Recent deep learning (DL) models have moved beyond static network
architectures to dynamic ones, handling data where the network structure
changes every example, such as sequences of variable lengths, trees, and
graphs. Existing dataflow-based programming models for DL---both static and
dynamic declaration---either cannot readily express these dynamic models, or
are inefficient due to repeated dataflow graph construction and processing, and
difficulties in batched execution. We present Cavs, a vertex-centric
programming interface and optimized system implementation for dynamic DL
models. Cavs represents dynamic network structure as a static vertex function
$\mathcal{F}$ and a dynamic instance-specific graph $\mathcal{G}$, and performs
backpropagation by scheduling the execution of $\mathcal{F}$ following the
dependencies in $\mathcal{G}$. Cavs bypasses expensive graph construction and
preprocessing overhead, allows for the use of static graph optimization
techniques on pre-defined operations in $\mathcal{F}$, and naturally exposes
batched execution opportunities over different graphs. Experiments comparing
Cavs to two state-of-the-art frameworks for dynamic NNs (TensorFlow Fold and
DyNet) demonstrate the efficacy of this approach: Cavs achieves a near one
order of magnitude speedup on training of various dynamic NN architectures, and
ablations demonstrate the contribution of our proposed batching and memory
management strategies.
|
PacGAN: The power of two samples in generative adversarial networks | Generative adversarial networks (GANs) are innovative techniques for learning
generative models of complex data distributions from samples. Despite
remarkable recent improvements in generating realistic images, one of their
major shortcomings is the fact that in practice, they tend to produce samples
with little diversity, even when trained on diverse datasets. This phenomenon,
known as mode collapse, has been the main focus of several recent advances in
GANs. Yet there is little understanding of why mode collapse happens and why
existing approaches are able to mitigate mode collapse. We propose a principled
approach to handling mode collapse, which we call packing. The main idea is to
modify the discriminator to make decisions based on multiple samples from the
same class, either real or artificially generated. We borrow analysis tools
from binary hypothesis testing---in particular the seminal result of Blackwell
[Bla53]---to prove a fundamental connection between packing and mode collapse.
We show that packing naturally penalizes generators with mode collapse, thereby
favoring generator distributions with less mode collapse during the training
process. Numerical experiments on benchmark datasets suggests that packing
provides significant improvements in practice as well.
|
Deep Reinforcement Learning Boosted by External Knowledge | Recent improvements in deep reinforcement learning have allowed to solve
problems in many 2D domains such as Atari games. However, in complex 3D
environments, numerous learning episodes are required which may be too time
consuming or even impossible especially in real-world scenarios. We present a
new architecture to combine external knowledge and deep reinforcement learning
using only visual input. A key concept of our system is augmenting image input
by adding environment feature information and combining two sources of
decision. We evaluate the performances of our method in a 3D
partially-observable environment from the Microsoft Malmo platform.
Experimental evaluation exhibits higher performance and faster learning
compared to a single reinforcement learning model.
|
Convergence Rates for Deterministic and Stochastic Subgradient Methods
Without Lipschitz Continuity | We extend the classic convergence rate theory for subgradient methods to
apply to non-Lipschitz functions. For the deterministic projected subgradient
method, we present a global $O(1/\sqrt{T})$ convergence rate for any convex
function which is locally Lipschitz around its minimizers. This approach is
based on Shor's classic subgradient analysis and implies generalizations of the
standard convergence rates for gradient descent on functions with Lipschitz or
H\"older continuous gradients. Further, we show a $O(1/\sqrt{T})$ convergence
rate for the stochastic projected subgradient method on convex functions with
at most quadratic growth, which improves to $O(1/T)$ under either strong
convexity or a weaker quadratic lower bound condition.
|
Sparse Phase Retrieval via Sparse PCA Despite Model Misspecification: A
Simplified and Extended Analysis | We consider the problem of high-dimensional misspecified phase retrieval.
This is where we have an $s$-sparse signal vector $\mathbf{x}_*$ in
$\mathbb{R}^n$, which we wish to recover using sampling vectors
$\textbf{a}_1,\ldots,\textbf{a}_m$, and measurements $y_1,\ldots,y_m$, which
are related by the equation $f(\left<\textbf{a}_i,\textbf{x}_*\right>) = y_i$.
Here, $f$ is an unknown link function satisfying a positive correlation with
the quadratic function. This problem was analyzed in a recent paper by Neykov,
Wang and Liu, who provided recovery guarantees for a two-stage algorithm with
sample complexity $m = O(s^2\log n)$. In this paper, we show that the first
stage of their algorithm suffices for signal recovery with the same sample
complexity, and extend the analysis to non-Gaussian measurements. Furthermore,
we show how the algorithm can be generalized to recover a signal vector
$\textbf{x}_*$ efficiently given geometric prior information other than
sparsity.
|
A Novel Document Generation Process for Topic Detection based on
Hierarchical Latent Tree Models | We propose a novel document generation process based on hierarchical latent
tree models (HLTMs) learned from data. An HLTM has a layer of observed word
variables at the bottom and multiple layers of latent variables on top. For
each document, we first sample values for the latent variables layer by layer
via logic sampling, then draw relative frequencies for the words conditioned on
the values of the latent variables, and finally generate words for the document
using the relative word frequencies. The motivation for the work is to take
word counts into consideration with HLTMs. In comparison with LDA-based
hierarchical document generation processes, the new process achieves
drastically better model fit with much fewer parameters. It also yields more
meaningful topics and topic hierarchies. It is the new state-of-the-art for the
hierarchical topic detection.
|
Neural Component Analysis for Fault Detection | Principal component analysis (PCA) is largely adopted for chemical process
monitoring and numerous PCA-based systems have been developed to solve various
fault detection and diagnosis problems. Since PCA-based methods assume that the
monitored process is linear, nonlinear PCA models, such as autoencoder models
and kernel principal component analysis (KPCA), has been proposed and applied
to nonlinear process monitoring. However, KPCA-based methods need to perform
eigen-decomposition (ED) on the kernel Gram matrix whose dimensions depend on
the number of training data. Moreover, prefixed kernel parameters cannot be
most effective for different faults which may need different parameters to
maximize their respective detection performances. Autoencoder models lack the
consideration of orthogonal constraints which is crucial for PCA-based
algorithms. To address these problems, this paper proposes a novel nonlinear
method, called neural component analysis (NCA), which intends to train a
feedforward neural work with orthogonal constraints such as those used in PCA.
NCA can adaptively learn its parameters through backpropagation and the
dimensionality of the nonlinear features has no relationship with the number of
training samples. Extensive experimental results on the Tennessee Eastman (TE)
benchmark process show the superiority of NCA in terms of missed detection rate
(MDR) and false alarm rate (FAR). The source code of NCA can be found in
https://github.com/haitaozhao/Neural-Component-Analysis.git.
|
GibbsNet: Iterative Adversarial Inference for Deep Graphical Models | Directed latent variable models that formulate the joint distribution as
$p(x,z) = p(z) p(x \mid z)$ have the advantage of fast and exact sampling.
However, these models have the weakness of needing to specify $p(z)$, often
with a simple fixed prior that limits the expressiveness of the model.
Undirected latent variable models discard the requirement that $p(z)$ be
specified with a prior, yet sampling from them generally requires an iterative
procedure such as blocked Gibbs-sampling that may require many steps to draw
samples from the joint distribution $p(x, z)$. We propose a novel approach to
learning the joint distribution between the data and a latent code which uses
an adversarially learned iterative procedure to gradually refine the joint
distribution, $p(x, z)$, to better match with the data distribution on each
step. GibbsNet is the best of both worlds both in theory and in practice.
Achieving the speed and simplicity of a directed latent variable model, it is
guaranteed (assuming the adversarial game reaches the virtual training criteria
global minimum) to produce samples from $p(x, z)$ with only a few sampling
iterations. Achieving the expressiveness and flexibility of an undirected
latent variable model, GibbsNet does away with the need for an explicit $p(z)$
and has the ability to do attribute prediction, class-conditional generation,
and joint image-attribute modeling in a single model which is not trained for
any of these specific tasks. We show empirically that GibbsNet is able to learn
a more complex $p(z)$ and show that this leads to improved inpainting and
iterative refinement of $p(x, z)$ for dozens of steps and stable generation
without collapse for thousands of steps, despite being trained on only a few
steps.
|
Outlier Detection by Consistent Data Selection Method | Often the challenge associated with tasks like fraud and spam detection[1] is
the lack of all likely patterns needed to train suitable supervised learning
models. In order to overcome this limitation, such tasks are attempted as
outlier or anomaly detection tasks. We also hypothesize that out- liers have
behavioral patterns that change over time. Limited data and continuously
changing patterns makes learning significantly difficult. In this work we are
proposing an approach that detects outliers in large data sets by relying on
data points that are consistent. The primary contribution of this work is that
it will quickly help retrieve samples for both consistent and non-outlier data
sets and is also mindful of new outlier patterns. No prior knowledge of each
set is required to extract the samples. The method consists of two phases, in
the first phase, consistent data points (non- outliers) are retrieved by an
ensemble method of unsupervised clustering techniques and in the second phase a
one class classifier trained on the consistent data point set is ap- plied on
the remaining sample set to identify the outliers. The approach is tested on
three publicly available data sets and the performance scores are competitive.
|
Deep learning enhanced mobile-phone microscopy | Mobile-phones have facilitated the creation of field-portable, cost-effective
imaging and sensing technologies that approach laboratory-grade instrument
performance. However, the optical imaging interfaces of mobile-phones are not
designed for microscopy and produce spatial and spectral distortions in imaging
microscopic specimens. Here, we report on the use of deep learning to correct
such distortions introduced by mobile-phone-based microscopes, facilitating the
production of high-resolution, denoised and colour-corrected images, matching
the performance of benchtop microscopes with high-end objective lenses, also
extending their limited depth-of-field. After training a convolutional neural
network, we successfully imaged various samples, including blood smears,
histopathology tissue sections, and parasites, where the recorded images were
highly compressed to ease storage and transmission for telemedicine
applications. This method is applicable to other low-cost, aberrated imaging
systems, and could offer alternatives for costly and bulky microscopes, while
also providing a framework for standardization of optical images for clinical
and biomedical applications.
|
Benchmarking Single Image Dehazing and Beyond | We present a comprehensive study and evaluation of existing single image
dehazing algorithms, using a new large-scale benchmark consisting of both
synthetic and real-world hazy images, called REalistic Single Image DEhazing
(RESIDE). RESIDE highlights diverse data sources and image contents, and is
divided into five subsets, each serving different training or evaluation
purposes. We further provide a rich variety of criteria for dehazing algorithm
evaluation, ranging from full-reference metrics, to no-reference metrics, to
subjective evaluation and the novel task-driven evaluation. Experiments on
RESIDE shed light on the comparisons and limitations of state-of-the-art
dehazing algorithms, and suggest promising future directions.
|
Transportation analysis of denoising autoencoders: a novel method for
analyzing deep neural networks | The feature map obtained from the denoising autoencoder (DAE) is investigated
by determining transportation dynamics of the DAE, which is a cornerstone for
deep learning. Despite the rapid development in its application, deep neural
networks remain analytically unexplained, because the feature maps are nested
and parameters are not faithful. In this paper, we address the problem of the
formulation of nested complex of parameters by regarding the feature map as a
transport map. Even when a feature map has different dimensions between input
and output, we can regard it as a transportation map by considering that both
the input and output spaces are embedded in a common high-dimensional space. In
addition, the trajectory is a geometric object and thus, is independent of
parameterization. In this manner, transportation can be regarded as a universal
character of deep neural networks. By determining and analyzing the
transportation dynamics, we can understand the behavior of a deep neural
network. In this paper, we investigate a fundamental case of deep neural
networks: the DAE. We derive the transport map of the DAE, and reveal that the
infinitely deep DAE transports mass to decrease a certain quantity, such as
entropy, of the data distribution. These results though analytically simple,
shed light on the correspondence between deep neural networks and the
Wasserstein gradient flows.
|
Temporal Stability in Predictive Process Monitoring | Predictive process monitoring is concerned with the analysis of events
produced during the execution of a business process in order to predict as
early as possible the final outcome of an ongoing case. Traditionally,
predictive process monitoring methods are optimized with respect to accuracy.
However, in environments where users make decisions and take actions in
response to the predictions they receive, it is equally important to optimize
the stability of the successive predictions made for each case. To this end,
this paper defines a notion of temporal stability for binary classification
tasks in predictive process monitoring and evaluates existing methods with
respect to both temporal stability and accuracy. We find that methods based on
XGBoost and LSTM neural networks exhibit the highest temporal stability. We
then show that temporal stability can be enhanced by hyperparameter-optimizing
random forests and XGBoost classifiers with respect to inter-run stability.
Finally, we show that time series smoothing techniques can further enhance
temporal stability at the expense of slightly lower accuracy.
|
Concept Formation and Dynamics of Repeated Inference in Deep Generative
Models | Deep generative models are reported to be useful in broad applications
including image generation. Repeated inference between data space and latent
space in these models can denoise cluttered images and improve the quality of
inferred results. However, previous studies only qualitatively evaluated image
outputs in data space, and the mechanism behind the inference has not been
investigated. The purpose of the current study is to numerically analyze
changes in activity patterns of neurons in the latent space of a deep
generative model called a "variational auto-encoder" (VAE). What kinds of
inference dynamics the VAE demonstrates when noise is added to the input data
are identified. The VAE embeds a dataset with clear cluster structures in the
latent space and the center of each cluster of multiple correlated data points
(memories) is referred as the concept. Our study demonstrated that transient
dynamics of inference first approaches a concept, and then moves close to a
memory. Moreover, the VAE revealed that the inference dynamics approaches a
more abstract concept to the extent that the uncertainty of input data
increases due to noise. It was demonstrated that by increasing the number of
the latent variables, the trend of the inference dynamics to approach a concept
can be enhanced, and the generalization ability of the VAE can be improved.
|
Empirical Evaluation of Kernel PCA Approximation Methods in
Classification Tasks | Kernel Principal Component Analysis (KPCA) is a popular dimensionality
reduction technique with a wide range of applications. However, it suffers from
the problem of poor scalability. Various approximation methods have been
proposed in the past to overcome this problem. The Nystr\"om method, Randomized
Nonlinear Component Analysis (RNCA) and Streaming Kernel Principal Component
Analysis (SKPCA) were proposed to deal with the scalability issue of KPCA.
Despite having theoretical guarantees, their performance in real world learning
tasks have not been explored previously. In this work the evaluation of SKPCA,
RNCA and Nystr\"om method for the task of classification is done for several
real world datasets. The results obtained indicate that SKPCA based features
gave much better classification accuracy when compared to the other methods for
a very large dataset.
|
Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box
Machine Learning Models | Many machine learning algorithms are vulnerable to almost imperceptible
perturbations of their inputs. So far it was unclear how much risk adversarial
perturbations carry for the safety of real-world machine learning applications
because most methods used to generate such perturbations rely either on
detailed model information (gradient-based attacks) or on confidence scores
such as class probabilities (score-based attacks), neither of which are
available in most real-world scenarios. In many such cases one currently needs
to retreat to transfer-based attacks which rely on cumbersome substitute
models, need access to the training data and can be defended against. Here we
emphasise the importance of attacks which solely rely on the final model
decision. Such decision-based attacks are (1) applicable to real-world
black-box models such as autonomous cars, (2) need less knowledge and are
easier to apply than transfer-based attacks and (3) are more robust to simple
defences than gradient- or score-based attacks. Previous attacks in this
category were limited to simple models or simple datasets. Here we introduce
the Boundary Attack, a decision-based attack that starts from a large
adversarial perturbation and then seeks to reduce the perturbation while
staying adversarial. The attack is conceptually simple, requires close to no
hyperparameter tuning, does not rely on substitute models and is competitive
with the best gradient-based attacks in standard computer vision tasks like
ImageNet. We apply the attack on two black-box algorithms from Clarifai.com.
The Boundary Attack in particular and the class of decision-based attacks in
general open new avenues to study the robustness of machine learning models and
raise new questions regarding the safety of deployed machine learning systems.
An implementation of the attack is available as part of Foolbox at
https://github.com/bethgelab/foolbox .
|
Deep Learning for IoT Big Data and Streaming Analytics: A Survey | In the era of the Internet of Things (IoT), an enormous amount of sensing
devices collect and/or generate various sensory data over time for a wide range
of fields and applications. Based on the nature of the application, these
devices will result in big or fast/real-time data streams. Applying analytics
over such data streams to discover new information, predict future insights,
and make control decisions is a crucial process that makes IoT a worthy
paradigm for businesses and a quality-of-life improving technology. In this
paper, we provide a thorough overview on using a class of advanced machine
learning techniques, namely Deep Learning (DL), to facilitate the analytics and
learning in the IoT domain. We start by articulating IoT data characteristics
and identifying two major treatments for IoT data from a machine learning
perspective, namely IoT big data analytics and IoT streaming data analytics. We
also discuss why DL is a promising approach to achieve the desired analytics in
these types of data and applications. The potential of using emerging DL
techniques for IoT data analytics are then discussed, and its promises and
challenges are introduced. We present a comprehensive background on different
DL architectures and algorithms. We also analyze and summarize major reported
research attempts that leveraged DL in the IoT domain. The smart IoT devices
that have incorporated DL in their intelligence background are also discussed.
DL implementation approaches on the fog and cloud centers in support of IoT
applications are also surveyed. Finally, we shed light on some challenges and
potential directions for future research. At the end of each section, we
highlight the lessons learned based on our experiments and review of the recent
literature.
|
Generating and Estimating Nonverbal Alphabets for Situated and
Multimodal Communications | In this paper, we discuss the formalized approach for generating and
estimating symbols (and alphabets), which can be communicated by the wide range
of non-verbal means based on specific user requirements (medium, priorities,
type of information that needs to be conveyed). The short characterization of
basic terms and parameters of such symbols (and alphabets) with approaches to
generate them are given. Then the framework, experimental setup, and some
machine learning methods to estimate usefulness and effectiveness of the
nonverbal alphabets and systems are presented. The previous results demonstrate
that usage of multimodal data sources (like wearable accelerometer, heart
monitor, muscle movements sensors, braincomputer interface) along with machine
learning approaches can provide the deeper understanding of the usefulness and
effectiveness of such alphabets and systems for nonverbal and situated
communication. The symbols (and alphabets) generated and estimated by such
methods may be useful in various applications: from synthetic languages and
constructed scripts to multimodal nonverbal and situated interaction between
people and artificial intelligence systems through Human-Computer Interfaces,
such as mouse gestures, touchpads, body gestures, eyetracking cameras,
wearables, and brain-computing interfaces, especially in applications for
elderly care and people with disabilities.
|
Deep Echo State Network (DeepESN): A Brief Survey | The study of deep recurrent neural networks (RNNs) and, in particular, of
deep Reservoir Computing (RC) is gaining an increasing research attention in
the neural networks community. The recently introduced Deep Echo State Network
(DeepESN) model opened the way to an extremely efficient approach for designing
deep neural networks for temporal data. At the same time, the study of DeepESNs
allowed to shed light on the intrinsic properties of state dynamics developed
by hierarchical compositions of recurrent layers, i.e. on the bias of depth in
RNNs architectural design. In this paper, we summarize the advancements in the
development, analysis and applications of DeepESNs.
|
Scaling Limit: Exact and Tractable Analysis of Online Learning
Algorithms with Applications to Regularized Regression and PCA | We present a framework for analyzing the exact dynamics of a class of online
learning algorithms in the high-dimensional scaling limit. Our results are
applied to two concrete examples: online regularized linear regression and
principal component analysis. As the ambient dimension tends to infinity, and
with proper time scaling, we show that the time-varying joint empirical
measures of the target feature vector and its estimates provided by the
algorithms will converge weakly to a deterministic measured-valued process that
can be characterized as the unique solution of a nonlinear PDE. Numerical
solutions of this PDE can be efficiently obtained. These solutions lead to
precise predictions of the performance of the algorithms, as many practical
performance metrics are linear functionals of the joint empirical measures. In
addition to characterizing the dynamic performance of online learning
algorithms, our asymptotic analysis also provides useful insights. In
particular, in the high-dimensional limit, and due to exchangeability, the
original coupled dynamics associated with the algorithms will be asymptotically
"decoupled", with each coordinate independently solving a 1-D effective
minimization problem via stochastic gradient descent. Exploiting this insight
for nonconvex optimization problems may prove an interesting line of future
research.
|
A Streaming Algorithm for Graph Clustering | We introduce a novel algorithm to perform graph clustering in the edge
streaming setting. In this model, the graph is presented as a sequence of edges
that can be processed strictly once. Our streaming algorithm has an extremely
low memory footprint as it stores only three integers per node and does not
keep any edge in memory. We provide a theoretical justification of the design
of the algorithm based on the modularity function, which is a usual metric to
evaluate the quality of a graph partition. We perform experiments on massive
real-life graphs ranging from one million to more than one billion edges and we
show that this new algorithm runs more than ten times faster than existing
algorithms and leads to similar or better detection scores on the largest
graphs.
|
Predicting Yelp Star Reviews Based on Network Structure with Deep
Learning | In this paper, we tackle the real-world problem of predicting Yelp
star-review rating based on business features (such as images, descriptions),
user features (average previous ratings), and, of particular interest, network
properties (which businesses has a user rated before). We compare multiple
models on different sets of features -- from simple linear regression on
network features only to deep learning models on network and item features.
In recent years, breakthroughs in deep learning have led to increased
accuracy in common supervised learning tasks, such as image classification,
captioning, and language understanding. However, the idea of combining deep
learning with network feature and structure appears to be novel. While the
problem of predicting future interactions in a network has been studied at
length, these approaches have often ignored either node-specific data or global
structure.
We demonstrate that taking a mixed approach combining both node-level
features and network information can effectively be used to predict Yelp-review
star ratings. We evaluate on the Yelp dataset by splitting our data along the
time dimension (as would naturally occur in the real-world) and comparing our
model against others which do no take advantage of the network structure and/or
deep learning.
|
CUSBoost: Cluster-based Under-sampling with Boosting for Imbalanced
Classification | Class imbalance classification is a challenging research problem in data
mining and machine learning, as most of the real-life datasets are often
imbalanced in nature. Existing learning algorithms maximise the classification
accuracy by correctly classifying the majority class, but misclassify the
minority class. However, the minority class instances are representing the
concept with greater interest than the majority class instances in real-life
applications. Recently, several techniques based on sampling methods
(under-sampling of the majority class and over-sampling the minority class),
cost-sensitive learning methods, and ensemble learning have been used in the
literature for classifying imbalanced datasets. In this paper, we introduce a
new clustering-based under-sampling approach with boosting (AdaBoost)
algorithm, called CUSBoost, for effective imbalanced classification. The
proposed algorithm provides an alternative to RUSBoost (random under-sampling
with AdaBoost) and SMOTEBoost (synthetic minority over-sampling with AdaBoost)
algorithms. We evaluated the performance of CUSBoost algorithm with the
state-of-the-art methods based on ensemble learning like AdaBoost, RUSBoost,
SMOTEBoost on 13 imbalance binary and multi-class datasets with various
imbalance ratios. The experimental results show that the CUSBoost is a
promising and effective approach for dealing with highly imbalanced datasets.
|
Music Generation by Deep Learning - Challenges and Directions | In addition to traditional tasks such as prediction, classification and
translation, deep learning is receiving growing attention as an approach for
music generation, as witnessed by recent research groups such as Magenta at
Google and CTRL (Creator Technology Research Lab) at Spotify. The motivation is
in using the capacity of deep learning architectures and training techniques to
automatically learn musical styles from arbitrary musical corpora and then to
generate samples from the estimated distribution. However, a direct application
of deep learning to generate content rapidly reaches limits as the generated
content tends to mimic the training set without exhibiting true creativity.
Moreover, deep learning architectures do not offer direct ways for controlling
generation (e.g., imposing some tonality or other arbitrary constraints).
Furthermore, deep learning architectures alone are autistic automata which
generate music autonomously without human user interaction, far from the
objective of interactively assisting musicians to compose and refine music.
Issues such as: control, structure, creativity and interactivity are the focus
of our analysis. In this paper, we select some limitations of a direct
application of deep learning to music generation, analyze why the issues are
not fulfilled and how to address them by possible approaches. Various examples
of recent systems are cited as examples of promising directions.
|
Logo Synthesis and Manipulation with Clustered Generative Adversarial
Networks | Designing a logo for a new brand is a lengthy and tedious back-and-forth
process between a designer and a client. In this paper we explore to what
extent machine learning can solve the creative task of the designer. For this,
we build a dataset -- LLD -- of 600k+ logos crawled from the world wide web.
Training Generative Adversarial Networks (GANs) for logo synthesis on such
multi-modal data is not straightforward and results in mode collapse for some
state-of-the-art methods. We propose the use of synthetic labels obtained
through clustering to disentangle and stabilize GAN training. We are able to
generate a high diversity of plausible logos and we demonstrate latent space
exploration techniques to ease the logo design task in an interactive manner.
Moreover, we validate the proposed clustered GAN training on CIFAR 10,
achieving state-of-the-art Inception scores when using synthetic labels
obtained via clustering the features of an ImageNet classifier. GANs can cope
with multi-modal data by means of synthetic labels achieved through clustering,
and our results show the creative potential of such techniques for logo
synthesis and manipulation. Our dataset and models will be made publicly
available at https://data.vision.ee.ethz.ch/cvl/lld/.
|
Integrated Model, Batch and Domain Parallelism in Training Neural
Networks | We propose a new integrated method of exploiting model, batch and domain
parallelism for the training of deep neural networks (DNNs) on large
distributed-memory computers using minibatch stochastic gradient descent (SGD).
Our goal is to find an efficient parallelization strategy for a fixed batch
size using $P$ processes. Our method is inspired by the communication-avoiding
algorithms in numerical linear algebra. We see $P$ processes as logically
divided into a $P_r \times P_c$ grid where the $P_r$ dimension is implicitly
responsible for model/domain parallelism and the $P_c$ dimension is implicitly
responsible for batch parallelism. In practice, the integrated matrix-based
parallel algorithm encapsulates these types of parallelism automatically. We
analyze the communication complexity and analytically demonstrate that the
lowest communication costs are often achieved neither with pure model nor with
pure data parallelism. We also show how the domain parallel approach can help
in extending the theoretical scaling limit of the typical batch parallel
method.
|
Sequential Prediction of Social Media Popularity with Deep Temporal
Context Networks | Prediction of popularity has profound impact for social media, since it
offers opportunities to reveal individual preference and public attention from
evolutionary social systems. Previous research, although achieves promising
results, neglects one distinctive characteristic of social data, i.e.,
sequentiality. For example, the popularity of online content is generated over
time with sequential post streams of social media. To investigate the
sequential prediction of popularity, we propose a novel prediction framework
called Deep Temporal Context Networks (DTCN) by incorporating both temporal
context and temporal attention into account. Our DTCN contains three main
components, from embedding, learning to predicting. With a joint embedding
network, we obtain a unified deep representation of multi-modal user-post data
in a common embedding space. Then, based on the embedded data sequence over
time, temporal context learning attempts to recurrently learn two adaptive
temporal contexts for sequential popularity. Finally, a novel temporal
attention is designed to predict new popularity (the popularity of a new
user-post pair) with temporal coherence across multiple time-scales.
Experiments on our released image dataset with about 600K Flickr photos
demonstrate that DTCN outperforms state-of-the-art deep prediction algorithms,
with an average of 21.51% relative performance improvement in the popularity
prediction (Spearman Ranking Correlation).
|
A Data-Driven Sparse-Learning Approach to Model Reduction in Chemical
Reaction Networks | In this paper, we propose an optimization-based sparse learning approach to
identify the set of most influential reactions in a chemical reaction network.
This reduced set of reactions is then employed to construct a reduced chemical
reaction mechanism, which is relevant to chemical interaction network modeling.
The problem of identifying influential reactions is first formulated as a
mixed-integer quadratic program, and then a relaxation method is leveraged to
reduce the computational complexity of our approach. Qualitative and
quantitative validation of the sparse encoding approach demonstrates that the
model captures important network structural properties with moderate
computational load.
|
Practical Bayesian optimization in the presence of outliers | Inference in the presence of outliers is an important field of research as
outliers are ubiquitous and may arise across a variety of problems and domains.
Bayesian optimization is method that heavily relies on probabilistic inference.
This allows outstanding sample efficiency because the probabilistic machinery
provides a memory of the whole optimization process. However, that virtue
becomes a disadvantage when the memory is populated with outliers, inducing
bias in the estimation. In this paper, we present an empirical evaluation of
Bayesian optimization methods in the presence of outliers. The empirical
evidence shows that Bayesian optimization with robust regression often produces
suboptimal results. We then propose a new algorithm which combines robust
regression (a Gaussian process with Student-t likelihood) with outlier
diagnostics to classify data points as outliers or inliers. By using an
scheduler for the classification of outliers, our method is more efficient and
has better convergence over the standard robust regression. Furthermore, we
show that even in controlled situations with no expected outliers, our method
is able to produce better results.
|
Learning From Noisy Singly-labeled Data | Supervised learning depends on annotated examples, which are taken to be the
\emph{ground truth}. But these labels often come from noisy crowdsourcing
platforms, like Amazon Mechanical Turk. Practitioners typically collect
multiple labels per example and aggregate the results to mitigate noise (the
classic crowdsourcing problem). Given a fixed annotation budget and unlimited
unlabeled data, redundant annotation comes at the expense of fewer labeled
examples. This raises two fundamental questions: (1) How can we best learn from
noisy workers? (2) How should we allocate our labeling budget to maximize the
performance of a classifier? We propose a new algorithm for jointly modeling
labels and worker quality from noisy crowd-sourced data. The alternating
minimization proceeds in rounds, estimating worker quality from disagreement
with the current model and then updating the model by optimizing a loss
function that accounts for the current estimate of worker quality. Unlike
previous approaches, even with only one annotation per example, our algorithm
can estimate worker quality. We establish a generalization error bound for
models learned with our algorithm and establish theoretically that it's better
to label many examples once (vs less multiply) when worker quality is above a
threshold. Experiments conducted on both ImageNet (with simulated noisy
workers) and MS-COCO (using the real crowdsourced labels) confirm our
algorithm's benefits.
|
Over the Air Deep Learning Based Radio Signal Classification | We conduct an in depth study on the performance of deep learning based radio
signal classification for radio communications signals. We consider a rigorous
baseline method using higher order moments and strong boosted gradient tree
classification and compare performance between the two approaches across a
range of configurations and channel impairments. We consider the effects of
carrier frequency offset, symbol rate, and multi-path fading in simulation and
conduct over-the-air measurement of radio classification performance in the lab
using software radios and compare performance and training strategies for both.
Finally we conclude with a discussion of remaining problems, and design
considerations for using such techniques.
|
Potential-Function Proofs for First-Order Methods | This note discusses proofs for convergence of first-order methods based on
simple potential-function arguments. We cover methods like gradient descent
(for both smooth and non-smooth settings), mirror descent, and some accelerated
variants.
|
On the organization of grid and place cells: Neural de-noising via
subspace learning | Place cells in the hippocampus are active when an animal visits a certain
location (referred to as a place field) within an environment. Grid cells in
the medial entorhinal cortex (MEC) respond at multiple locations, with firing
fields that form a periodic and hexagonal tiling of the environment. The joint
activity of grid and place cell populations, as a function of location, forms a
neural code for space. An ensemble of codes is generated by varying grid and
place cell population parameters. For each code in this ensemble, codewords are
generated by stimulating a network with a discrete set of locations. In this
manuscript, we develop an understanding of the relationships between coding
theoretic properties of these combined populations and code construction
parameters. These relationships are revisited by measuring the performances of
biologically realizable algorithms implemented by networks of place and grid
cell populations, as well as constraint neurons, which perform de-noising
operations. Objectives of this work include the investigation of coding
theoretic limitations of the mammalian neural code for location and how
communication between grid and place cell networks may improve the accuracy of
each population's representation. Simulations demonstrate that de-noising
mechanisms analyzed here can significantly improve fidelity of this neural
representation of space. Further, patterns observed in connectivity of each
population of simulated cells suggest that
inter-hippocampal-medial-entorhinal-cortical connectivity decreases downward
along the dorsoventral axis.
|
Multi-focus Attention Network for Efficient Deep Reinforcement Learning | Deep reinforcement learning (DRL) has shown incredible performance in
learning various tasks to the human level. However, unlike human perception,
current DRL models connect the entire low-level sensory input to the
state-action values rather than exploiting the relationship between and among
entities that constitute the sensory input. Because of this difference, DRL
needs vast amount of experience samples to learn. In this paper, we propose a
Multi-focus Attention Network (MANet) which mimics human ability to spatially
abstract the low-level sensory input into multiple entities and attend to them
simultaneously. The proposed method first divides the low-level input into
several segments which we refer to as partial states. After this segmentation,
parallel attention layers attend to the partial states relevant to solving the
task. Our model estimates state-action values using these attended partial
states. In our experiments, MANet attains highest scores with significantly
less experience samples. Additionally, the model shows higher performance
compared to the Deep Q-network and the single attention model as benchmarks.
Furthermore, we extend our model to attentive communication model for
performing multi-agent cooperative tasks. In multi-agent cooperative task
experiments, our model shows 20% faster learning than existing state-of-the-art
model.
|
QLBS: Q-Learner in the Black-Scholes(-Merton) Worlds | This paper presents a discrete-time option pricing model that is rooted in
Reinforcement Learning (RL), and more specifically in the famous Q-Learning
method of RL. We construct a risk-adjusted Markov Decision Process for a
discrete-time version of the classical Black-Scholes-Merton (BSM) model, where
the option price is an optimal Q-function, while the optimal hedge is a second
argument of this optimal Q-function, so that both the price and hedge are parts
of the same formula. Pricing is done by learning to dynamically optimize
risk-adjusted returns for an option replicating portfolio, as in the Markowitz
portfolio theory. Using Q-Learning and related methods, once created in a
parametric setting, the model is able to go model-free and learn to price and
hedge an option directly from data, and without an explicit model of the world.
This suggests that RL may provide efficient data-driven and model-free methods
for optimal pricing and hedging of options, once we depart from the academic
continuous-time limit, and vice versa, option pricing methods developed in
Mathematical Finance may be viewed as special cases of model-based
Reinforcement Learning. Further, due to simplicity and tractability of our
model which only needs basic linear algebra (plus Monte Carlo simulation, if we
work with synthetic data), and its close relation to the original BSM model, we
suggest that our model could be used for benchmarking of different RL
algorithms for financial trading applications
|
Inverse Reinforcement Learning for Marketing | Learning customer preferences from an observed behaviour is an important
topic in the marketing literature. Structural models typically model
forward-looking customers or firms as utility-maximizing agents whose utility
is estimated using methods of Stochastic Optimal Control. We suggest an
alternative approach to study dynamic consumer demand, based on Inverse
Reinforcement Learning (IRL). We develop a version of the Maximum Entropy IRL
that leads to a highly tractable model formulation that amounts to
low-dimensional convex optimization in the search for optimal model parameters.
Using simulations of consumer demand, we show that observational noise for
identical customers can be easily confused with an apparent consumer
heterogeneity.
|
Stochastic Low-Rank Bandits | Many problems in computer vision and recommender systems involve low-rank
matrices. In this work, we study the problem of finding the maximum entry of a
stochastic low-rank matrix from sequential observations. At each step, a
learning agent chooses pairs of row and column arms, and receives the noisy
product of their latent values as a reward. The main challenge is that the
latent values are unobserved. We identify a class of non-negative matrices
whose maximum entry can be found statistically efficiently and propose an
algorithm for finding them, which we call LowRankElim. We derive a
$\DeclareMathOperator{\poly}{poly} O((K + L) \poly(d) \Delta^{-1} \log n)$
upper bound on its $n$-step regret, where $K$ is the number of rows, $L$ is the
number of columns, $d$ is the rank of the matrix, and $\Delta$ is the minimum
gap. The bound depends on other problem-specific constants that clearly do not
depend $K L$. To the best of our knowledge, this is the first such result in
the literature.
|
Stability Selection for Structured Variable Selection | In variable or graph selection problems, finding a right-sized model or
controlling the number of false positives is notoriously difficult. Recently, a
meta-algorithm called Stability Selection was proposed that can provide
reliable finite-sample control of the number of false positives. Its benefits
were demonstrated when used in conjunction with the lasso and orthogonal
matching pursuit algorithms.
In this paper, we investigate the applicability of stability selection to
structured selection algorithms: the group lasso and the structured
input-output lasso. We find that using stability selection often increases the
power of both algorithms, but that the presence of complex structure reduces
the reliability of error control under stability selection. We give strategies
for setting tuning parameters to obtain a good model size under stability
selection, and highlight its strengths and weaknesses compared to competing
methods screen and clean and cross-validation. We give guidelines about when to
use which error control method.
|
Differentiable lower bound for expected BLEU score | In natural language processing tasks performance of the models is often
measured with some non-differentiable metric, such as BLEU score. To use
efficient gradient-based methods for optimization, it is a common workaround to
optimize some surrogate loss function. This approach is effective if
optimization of such loss also results in improving target metric. The
corresponding problem is referred to as loss-evaluation mismatch. In the
present work we propose a method for calculation of differentiable lower bound
of expected BLEU score that does not involve computationally expensive sampling
procedure such as the one required when using REINFORCE rule from reinforcement
learning (RL) framework.
|
Mathematics of Deep Learning | Recently there has been a dramatic increase in the performance of recognition
systems due to the introduction of deep architectures for representation
learning and classification. However, the mathematical reasons for this success
remain elusive. This tutorial will review recent work that aims to provide a
mathematical justification for several properties of deep networks, such as
global optimality, geometric stability, and invariance of the learned
representations.
|
Exponential convergence of testing error for stochastic gradient methods | We consider binary classification problems with positive definite kernels and
square loss, and study the convergence rates of stochastic gradient methods. We
show that while the excess testing loss (squared loss) converges slowly to zero
as the number of observations (and thus iterations) goes to infinity, the
testing error (classification error) converges exponentially fast if low-noise
conditions are assumed.
|
Ballpark Crowdsourcing: The Wisdom of Rough Group Comparisons | Crowdsourcing has become a popular method for collecting labeled training
data. However, in many practical scenarios traditional labeling can be
difficult for crowdworkers (for example, if the data is high-dimensional or
unintuitive, or the labels are continuous).
In this work, we develop a novel model for crowdsourcing that can complement
standard practices by exploiting people's intuitions about groups and relations
between them. We employ a recent machine learning setting, called Ballpark
Learning, that can estimate individual labels given only coarse, aggregated
signal over groups of data points. To address the important case of continuous
labels, we extend the Ballpark setting (which focused on classification) to
regression problems. We formulate the problem as a convex optimization problem
and propose fast, simple methods with an innate robustness to outliers.
We evaluate our methods on real-world datasets, demonstrating how useful
constraints about groups can be harnessed from a crowd of non-experts. Our
methods can rival supervised models trained on many true labels, and can obtain
considerably better results from the crowd than a standard label-collection
process (for a lower price). By collecting rough guesses on groups of instances
and using machine learning to infer the individual labels, our lightweight
framework is able to address core crowdsourcing challenges and train machine
learning models in a cost-effective way.
|
FFT-Based Deep Learning Deployment in Embedded Systems | Deep learning has delivered its powerfulness in many application domains,
especially in image and speech recognition. As the backbone of deep learning,
deep neural networks (DNNs) consist of multiple layers of various types with
hundreds to thousands of neurons. Embedded platforms are now becoming essential
for deep learning deployment due to their portability, versatility, and energy
efficiency. The large model size of DNNs, while providing excellent accuracy,
also burdens the embedded platforms with intensive computation and storage.
Researchers have investigated on reducing DNN model size with negligible
accuracy loss. This work proposes a Fast Fourier Transform (FFT)-based DNN
training and inference model suitable for embedded platforms with reduced
asymptotic complexity of both computation and storage, making our approach
distinguished from existing approaches. We develop the training and inference
algorithms based on FFT as the computing kernel and deploy the FFT-based
inference model on embedded platforms achieving extraordinary processing speed.
|
Predicting Station-level Hourly Demands in a Large-scale Bike-sharing
Network: A Graph Convolutional Neural Network Approach | This study proposes a novel Graph Convolutional Neural Network with
Data-driven Graph Filter (GCNN-DDGF) model that can learn hidden heterogeneous
pairwise correlations between stations to predict station-level hourly demand
in a large-scale bike-sharing network. Two architectures of the GCNN-DDGF model
are explored; GCNNreg-DDGF is a regular GCNN-DDGF model which contains the
convolution and feedforward blocks, and GCNNrec-DDGF additionally contains a
recurrent block from the Long Short-term Memory neural network architecture to
capture temporal dependencies in the bike-sharing demand series. Furthermore,
four types of GCNN models are proposed whose adjacency matrices are based on
various bike-sharing system data, including Spatial Distance matrix (SD),
Demand matrix (DE), Average Trip Duration matrix (ATD), and Demand Correlation
matrix (DC). These six types of GCNN models and seven other benchmark models
are built and compared on a Citi Bike dataset from New York City which includes
272 stations and over 28 million transactions from 2013 to 2016. Results show
that the GCNNrec-DDGF performs the best in terms of the Root Mean Square Error,
the Mean Absolute Error and the coefficient of determination (R2), followed by
the GCNNreg-DDGF. They outperform the other models. Through a more detailed
graph network analysis based on the learned DDGF, insights are obtained on the
black box of the GCNN-DDGF model. It is found to capture some information
similar to details embedded in the SD, DE and DC matrices. More importantly, it
also uncovers hidden heterogeneous pairwise correlations between stations that
are not revealed by any of those matrices.
|
Learning Low-shot facial representations via 2D warping | In this work, we mainly study the influence of the 2D warping module for
one-shot face recognition.
|
Deep Prior | The recent literature on deep learning offers new tools to learn a rich
probability distribution over high dimensional data such as images or sounds.
In this work we investigate the possibility of learning the prior distribution
over neural network parameters using such tools. Our resulting variational
Bayes algorithm generalizes well to new tasks, even when very few training
examples are provided. Furthermore, this learned prior allows the model to
extrapolate correctly far from a given task's training data on a meta-dataset
of periodic signals.
|
Learning Compact Recurrent Neural Networks with Block-Term Tensor
Decomposition | Recurrent Neural Networks (RNNs) are powerful sequence modeling tools.
However, when dealing with high dimensional inputs, the training of RNNs
becomes computational expensive due to the large number of model parameters.
This hinders RNNs from solving many important computer vision tasks, such as
Action Recognition in Videos and Image Captioning. To overcome this problem, we
propose a compact and flexible structure, namely Block-Term tensor
decomposition, which greatly reduces the parameters of RNNs and improves their
training efficiency. Compared with alternative low-rank approximations, such as
tensor-train RNN (TT-RNN), our method, Block-Term RNN (BT-RNN), is not only
more concise (when using the same rank), but also able to attain a better
approximation to the original RNNs with much fewer parameters. On three
challenging tasks, including Action Recognition in Videos, Image Captioning and
Image Generation, BT-RNN outperforms TT-RNN and the standard RNN in terms of
both prediction accuracy and convergence rate. Specifically, BT-LSTM utilizes
17,388 times fewer parameters than the standard LSTM to achieve an accuracy
improvement over 15.6\% in the Action Recognition task on the UCF11 dataset.
|
Rasa: Open Source Language Understanding and Dialogue Management | We introduce a pair of tools, Rasa NLU and Rasa Core, which are open source
python libraries for building conversational software. Their purpose is to make
machine-learning based dialogue management and language understanding
accessible to non-specialist software developers. In terms of design
philosophy, we aim for ease of use, and bootstrapping from minimal (or no)
initial training data. Both packages are extensively documented and ship with a
comprehensive suite of tests. The code is available at
https://github.com/RasaHQ/
|
Rate of Change Analysis for Interestingness Measures | The use of Association Rule Mining techniques in diverse contexts and domains
has resulted in the creation of numerous interestingness measures. This, in
turn, has motivated researchers to come up with various classification schemes
for these measures. One popular approach to classify the objective measures is
to assess the set of mathematical properties they satisfy in order to help
practitioners select the right measure for a given problem. In this research,
we discuss the insufficiency of the existing properties in literature to
capture certain behaviors of interestingness measures. This motivates us to
present a novel approach to analyze and classify measures. We refer to this as
a rate of change analysis (RCA). In this analysis a measure is described by how
it varies if there is a unit change in the frequency count
$(f_{11},f_{10},f_{01},f_{00})$, for different pre-existing states of the
frequency counts. More formally, we look at the first partial derivative of the
measure with respect to the various frequency count variables. We then use this
analysis to define two new properties, Unit-Null Asymptotic Invariance (UNAI)
and Unit-Null Zero Rate (UNZR). UNAI looks at the asymptotic effect of adding
frequency patterns, while UNZR looks at the initial effect of adding frequency
patterns when they do not pre-exist in the dataset. We present a comprehensive
analysis of 50 interestingness measures and classify them in accordance with
the two properties. We also present empirical studies, involving both synthetic
and real-world datasets, which are used to cluster various measures according
to the rule ranking patterns of the measures. The study concludes with the
observation that classification of measures using the empirical clusters share
significant similarities to the classification of measures done through the
properties presented in this research.
|
Towards Deep Modeling of Music Semantics using EEG Regularizers | Modeling of music audio semantics has been previously tackled through
learning of mappings from audio data to high-level tags or latent unsupervised
spaces. The resulting semantic spaces are theoretically limited, either because
the chosen high-level tags do not cover all of music semantics or because audio
data itself is not enough to determine music semantics. In this paper, we
propose a generic framework for semantics modeling that focuses on the
perception of the listener, through EEG data, in addition to audio data. We
implement this framework using a novel end-to-end 2-view Neural Network (NN)
architecture and a Deep Canonical Correlation Analysis (DCCA) loss function
that forces the semantic embedding spaces of both views to be maximally
correlated. We also detail how the EEG dataset was collected and use it to
train our proposed model. We evaluate the learned semantic space in a transfer
learning context, by using it as an audio feature extractor in an independent
dataset and proxy task: music audio-lyrics cross-modal retrieval. We show that
our embedding model outperforms Spotify features and performs comparably to a
state-of-the-art embedding model that was trained on 700 times more data. We
further discuss improvements to the model that are likely to improve its
performance.
|
Pointwise Convolutional Neural Networks | Deep learning with 3D data such as reconstructed point clouds and CAD models
has received great research interests recently. However, the capability of
using point clouds with convolutional neural network has been so far not fully
explored. In this paper, we present a convolutional neural network for semantic
segmentation and object recognition with 3D point clouds. At the core of our
network is pointwise convolution, a new convolution operator that can be
applied at each point of a point cloud. Our fully convolutional network design,
while being surprisingly simple to implement, can yield competitive accuracy in
both semantic segmentation and object recognition task.
|
Proximodistal Exploration in Motor Learning as an Emergent Property of
Optimization | To harness the complexity of their high-dimensional bodies during
sensorimotor development, infants are guided by patterns of freezing and
freeing of degrees of freedom. For instance, when learning to reach, infants
free the degrees of freedom in their arm proximodistally, i.e. from joints that
are closer to the body to those that are more distant. Here, we formulate and
study computationally the hypothesis that such patterns can emerge
spontaneously as the result of a family of stochastic optimization processes
(evolution strategies with covariance-matrix adaptation), without an innate
encoding of a maturational schedule. In particular, we present simulated
experiments with an arm where a computational learner progressively acquires
reaching skills through adaptive exploration, and we show that a proximodistal
organization appears spontaneously, which we denote PDFF (ProximoDistal
Freezing and Freeing of degrees of freedom). We also compare this emergent
organization between different arm morphologies -- from human-like to quite
unnatural ones -- to study the effect of different kinematic structures on the
emergence of PDFF. Keywords: human motor learning; proximo-distal exploration;
stochastic optimization; modelling; evolution strategies; cross-entropy
methods; policy search; morphology.}
|
Spatial-temporal wind field prediction by Artificial Neural Networks | The prediction of near surface wind speed is becoming increasingly vital for
the operation of electrical energy grids as the capacity of installed wind
power grows. The majority of predictive wind speed modeling has focused on
point-based time-series forecasting. Effectively balancing demand and supply in
the presence of distributed wind turbine electricity generation, however,
requires the prediction of wind fields in space and time. Additionally,
predictions of full wind fields are particularly useful for future power
planning such as the optimization of electricity power supply systems. In this
paper, we propose a composite artificial neural network (ANN) model to predict
the 6-hour and 24-hour ahead average wind speed over a large area (~3.15*106
km2). The ANN model consists of a convolutional input layer, a Long Short-Term
Memory (LSTM) hidden layer, and a transposed convolutional layer as the output
layer. We compare the ANN model with two non-parametric models, a null
persistence model and a mean value model, and find that the ANN model has
substantially smaller error than each of these models. Additionally, the ANN
model also generally performs better than integrated autoregressive moving
average models, which are trained for optimal performance in specific
locations.
|
DANCin SEQ2SEQ: Fooling Text Classifiers with Adversarial Text Example
Generation | Machine learning models are powerful but fallible. Generating adversarial
examples - inputs deliberately crafted to cause model misclassification or
other errors - can yield important insight into model assumptions and
vulnerabilities. Despite significant recent work on adversarial example
generation targeting image classifiers, relatively little work exists exploring
adversarial example generation for text classifiers; additionally, many
existing adversarial example generation algorithms require full access to
target model parameters, rendering them impractical for many real-world
attacks. In this work, we introduce DANCin SEQ2SEQ, a GAN-inspired algorithm
for adversarial text example generation targeting largely black-box text
classifiers. We recast adversarial text example generation as a reinforcement
learning problem, and demonstrate that our algorithm offers preliminary but
promising steps towards generating semantically meaningful adversarial text
examples in a real-world attack scenario.
|
Stochastic Particle Gradient Descent for Infinite Ensembles | The superior performance of ensemble methods with infinite models are well
known. Most of these methods are based on optimization problems in
infinite-dimensional spaces with some regularization, for instance, boosting
methods and convex neural networks use $L^1$-regularization with the
non-negative constraint. However, due to the difficulty of handling
$L^1$-regularization, these problems require early stopping or a rough
approximation to solve it inexactly. In this paper, we propose a new ensemble
learning method that performs in a space of probability measures, that is, our
method can handle the $L^1$-constraint and the non-negative constraint in a
rigorous way. Such an optimization is realized by proposing a general purpose
stochastic optimization method for learning probability measures via
parameterization using transport maps on base models. As a result of running
the method, a transport map to output an infinite ensemble is obtained, which
forms a residual-type network. From the perspective of functional gradient
methods, we give a convergence rate as fast as that of a stochastic
optimization method for finite dimensional nonconvex problems. Moreover, we
show an interior optimality property of a local optimality condition used in
our analysis.
|
Nonparametric Neural Networks | Automatically determining the optimal size of a neural network for a given
task without prior information currently requires an expensive global search
and training many networks from scratch. In this paper, we address the problem
of automatically finding a good network size during a single training cycle. We
introduce *nonparametric neural networks*, a non-probabilistic framework for
conducting optimization over all possible network sizes and prove its soundness
when network growth is limited via an L_p penalty. We train networks under this
framework by continuously adding new units while eliminating redundant units
via an L_2 penalty. We employ a novel optimization algorithm, which we term
*adaptive radial-angular gradient descent* or *AdaRad*, and obtain promising
results.
|
AI2-THOR: An Interactive 3D Environment for Visual AI | We introduce The House Of inteRactions (THOR), a framework for visual AI
research, available at http://ai2thor.allenai.org. AI2-THOR consists of near
photo-realistic 3D indoor scenes, where AI agents can navigate in the scenes
and interact with objects to perform tasks. AI2-THOR enables research in many
different domains including but not limited to deep reinforcement learning,
imitation learning, learning by interaction, planning, visual question
answering, unsupervised representation learning, object detection and
segmentation, and learning models of cognition. The goal of AI2-THOR is to
facilitate building visually intelligent models and push the research forward
in this domain.
|
Graph-Sparse Logistic Regression | We introduce Graph-Sparse Logistic Regression, a new algorithm for
classification for the case in which the support should be sparse but connected
on a graph. We val- idate this algorithm against synthetic data and benchmark
it against L1-regularized Logistic Regression. We then explore our technique in
the bioinformatics context of proteomics data on the interactome graph. We make
all our experimental code public and provide GSLR as an open source package.
|
Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning | Deep learning models have achieved high performance on many tasks, and thus
have been applied to many security-critical scenarios. For example, deep
learning-based face recognition systems have been used to authenticate users to
access many security-sensitive applications like payment apps. Such usages of
deep learning systems provide the adversaries with sufficient incentives to
perform attacks against these systems for their adversarial purposes. In this
work, we consider a new type of attacks, called backdoor attacks, where the
attacker's goal is to create a backdoor into a learning-based authentication
system, so that he can easily circumvent the system by leveraging the backdoor.
Specifically, the adversary aims at creating backdoor instances, so that the
victim learning system will be misled to classify the backdoor instances as a
target label specified by the adversary. In particular, we study backdoor
poisoning attacks, which achieve backdoor attacks using poisoning strategies.
Different from all existing work, our studied poisoning strategies can apply
under a very weak threat model: (1) the adversary has no knowledge of the model
and the training set used by the victim system; (2) the attacker is allowed to
inject only a small amount of poisoning samples; (3) the backdoor key is hard
to notice even by human beings to achieve stealthiness. We conduct evaluation
to demonstrate that a backdoor adversary can inject only around 50 poisoning
samples, while achieving an attack success rate of above 90%. We are also the
first work to show that a data poisoning attack can create physically
implementable backdoors without touching the training process. Our work
demonstrates that backdoor poisoning attacks pose real threats to a learning
system, and thus highlights the importance of further investigation and
proposing defense strategies against them.
|
Safe Policy Search with Gaussian Process Models | We propose a method to optimise the parameters of a policy which will be used
to safely perform a given task in a data-efficient manner. We train a Gaussian
process model to capture the system dynamics, based on the PILCO framework. Our
model has useful analytic properties, which allow closed form computation of
error gradients and estimating the probability of violating given state space
constraints. During training, as well as operation, only policies that are
deemed safe are implemented on the real system, minimising the risk of failure.
|
CoDraw: Collaborative Drawing as a Testbed for Grounded Goal-driven
Communication | In this work, we propose a goal-driven collaborative task that combines
language, perception, and action. Specifically, we develop a Collaborative
image-Drawing game between two agents, called CoDraw. Our game is grounded in a
virtual world that contains movable clip art objects. The game involves two
players: a Teller and a Drawer. The Teller sees an abstract scene containing
multiple clip art pieces in a semantically meaningful configuration, while the
Drawer tries to reconstruct the scene on an empty canvas using available clip
art pieces. The two players communicate with each other using natural language.
We collect the CoDraw dataset of ~10K dialogs consisting of ~138K messages
exchanged between human players. We define protocols and metrics to evaluate
learned agents in this testbed, highlighting the need for a novel "crosstalk"
evaluation condition which pairs agents trained independently on disjoint
subsets of the training data. We present models for our task and benchmark them
using both fully automated evaluation and by having them play the game live
with humans.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.