title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|
Learning to Reweight Examples for Robust Deep Learning | Deep neural networks have been shown to be very powerful modeling tools for
many supervised learning tasks involving complex input patterns. However, they
can also easily overfit to training set biases and label noises. In addition to
various regularizers, example reweighting algorithms are popular solutions to
these problems, but they require careful tuning of additional hyperparameters,
such as example mining schedules and regularization hyperparameters. In
contrast to past reweighting methods, which typically consist of functions of
the cost value of each example, in this work we propose a novel meta-learning
algorithm that learns to assign weights to training examples based on their
gradient directions. To determine the example weights, our method performs a
meta gradient descent step on the current mini-batch example weights (which are
initialized from zero) to minimize the loss on a clean unbiased validation set.
Our proposed method can be easily implemented on any type of deep network, does
not require any additional hyperparameter tuning, and achieves impressive
performance on class imbalance and corrupted label problems where only a small
amount of clean validation data is available.
|
AAANE: Attention-based Adversarial Autoencoder for Multi-scale Network
Embedding | Network embedding represents nodes in a continuous vector space and preserves
structure information from the Network. Existing methods usually adopt a
"one-size-fits-all" approach when concerning multi-scale structure information,
such as first- and second-order proximity of nodes, ignoring the fact that
different scales play different roles in the embedding learning. In this paper,
we propose an Attention-based Adversarial Autoencoder Network Embedding(AAANE)
framework, which promotes the collaboration of different scales and lets them
vote for robust representations. The proposed AAANE consists of two components:
1) Attention-based autoencoder effectively capture the highly non-linear
network structure, which can de-emphasize irrelevant scales during training. 2)
An adversarial regularization guides the autoencoder learn robust
representations by matching the posterior distribution of the latent embeddings
to given prior distribution. This is the first attempt to introduce attention
mechanisms to multi-scale network embedding. Experimental results on real-world
networks show that our learned attention parameters are different for every
network and the proposed approach outperforms existing state-of-the-art
approaches for network embedding.
|
A Proximal Block Coordinate Descent Algorithm for Deep Neural Network
Training | Training deep neural networks (DNNs) efficiently is a challenge due to the
associated highly nonconvex optimization. The backpropagation (backprop)
algorithm has long been the most widely used algorithm for gradient computation
of parameters of DNNs and is used along with gradient descent-type algorithms
for this optimization task. Recent work have shown the efficiency of block
coordinate descent (BCD) type methods empirically for training DNNs. In view of
this, we propose a novel algorithm based on the BCD method for training DNNs
and provide its global convergence results built upon the powerful framework of
the Kurdyka-Lojasiewicz (KL) property. Numerical experiments on standard
datasets demonstrate its competitive efficiency against standard optimizers
with backprop.
|
Comparing Generative Adversarial Network Techniques for Image Creation
and Modification | Generative adversarial networks (GANs) have demonstrated to be successful at
generating realistic real-world images. In this paper we compare various GAN
techniques, both supervised and unsupervised. The effects on training stability
of different objective functions are compared. We add an encoder to the
network, making it possible to encode images to the latent space of the GAN.
The generator, discriminator and encoder are parameterized by deep
convolutional neural networks. For the discriminator network we experimented
with using the novel Capsule Network, a state-of-the-art technique for
detecting global features in images. Experiments are performed using a digit
and face dataset, with various visualizations illustrating the results. The
results show that using the encoder network it is possible to reconstruct
images. With the conditional GAN we can alter visual attributes of generated or
encoded images. The experiments with the Capsule Network as discriminator
result in generated images of a lower quality, compared to a standard
convolutional neural network.
|
Entanglement-guided architectures of machine learning by quantum tensor
network | It is a fundamental, but still elusive question whether the schemes based on
quantum mechanics, in particular on quantum entanglement, can be used for
classical information processing and machine learning. Even partial answer to
this question would bring important insights to both fields of machine learning
and quantum mechanics. In this work, we implement simple numerical experiments,
related to pattern/images classification, in which we represent the classifiers
by many-qubit quantum states written in the matrix product states (MPS).
Classical machine learning algorithm is applied to these quantum states to
learn the classical data. We explicitly show how quantum entanglement (i.e.,
single-site and bipartite entanglement) can emerge in such represented images.
Entanglement characterizes here the importance of data, and such information
are practically used to guide the architecture of MPS, and improve the
efficiency. The number of needed qubits can be reduced to less than 1/10 of the
original number, which is within the access of the state-of-the-art quantum
computers. We expect such numerical experiments could open new paths in
charactering classical machine learning algorithms, and at the same time shed
lights on the generic quantum simulations/computations of machine learning
tasks.
|
Gradient descent in Gaussian random fields as a toy model for
high-dimensional optimisation in deep learning | In this paper we model the loss function of high-dimensional optimization
problems by a Gaussian random field, or equivalently a Gaussian process. Our
aim is to study gradient descent in such loss functions or energy landscapes
and compare it to results obtained from real high-dimensional optimization
problems such as encountered in deep learning. In particular, we analyze the
distribution of the improved loss function after a step of gradient descent,
provide analytic expressions for the moments as well as prove asymptotic
normality as the dimension of the parameter space becomes large. Moreover, we
compare this with the expectation of the global minimum of the landscape
obtained by means of the Euler characteristic of excursion sets. Besides
complementing our analytical findings with numerical results from simulated
Gaussian random fields, we also compare it to loss functions obtained from
optimisation problems on synthetic and real data sets by proposing a "black
box" random field toy-model for a deep neural network loss function.
|
Equation Embeddings | We present an unsupervised approach for discovering semantic representations
of mathematical equations. Equations are challenging to analyze because each is
unique, or nearly unique. Our method, which we call equation embeddings, finds
good representations of equations by using the representations of their
surrounding words. We used equation embeddings to analyze four collections of
scientific articles from the arXiv, covering four computer science domains
(NLP, IR, AI, and ML) and $\sim$98.5k equations. Quantitatively, we found that
equation embeddings provide better models when compared to existing word
embedding approaches. Qualitatively, we found that equation embeddings provide
coherent semantic representations of equations and can capture semantic
similarity to other equations and to words.
|
Posterior Concentration for Sparse Deep Learning | Spike-and-Slab Deep Learning (SS-DL) is a fully Bayesian alternative to
Dropout for improving generalizability of deep ReLU networks. This new type of
regularization enables provable recovery of smooth input-output maps with
unknown levels of smoothness. Indeed, we show that the posterior distribution
concentrates at the near minimax rate for $\alpha$-H\"older smooth maps,
performing as well as if we knew the smoothness level $\alpha$ ahead of time.
Our result sheds light on architecture design for deep neural networks, namely
the choice of depth, width and sparsity level. These network attributes
typically depend on unknown smoothness in order to be optimal. We obviate this
constraint with the fully Bayes construction. As an aside, we show that SS-DL
does not overfit in the sense that the posterior concentrates on smaller
networks with fewer (up to the optimal number of) nodes and links. Our results
provide new theoretical justifications for deep ReLU networks from a Bayesian
point of view.
|
Natural Gradients in Practice: Non-Conjugate Variational Inference in
Gaussian Process Models | The natural gradient method has been used effectively in conjugate Gaussian
process models, but the non-conjugate case has been largely unexplored. We
examine how natural gradients can be used in non-conjugate stochastic settings,
together with hyperparameter learning. We conclude that the natural gradient
can significantly improve performance in terms of wall-clock time. For
ill-conditioned posteriors the benefit of the natural gradient method is
especially pronounced, and we demonstrate a practical setting where ordinary
gradients are unusable. We show how natural gradients can be computed
efficiently and automatically in any parameterization, using automatic
differentiation. Our code is integrated into the GPflow package.
|
Fast variational Bayes for heavy-tailed PLDA applied to i-vectors and
x-vectors | The standard state-of-the-art backend for text-independent speaker
recognizers that use i-vectors or x-vectors, is Gaussian PLDA (G-PLDA),
assisted by a Gaussianization step involving length normalization. G-PLDA can
be trained with both generative or discriminative methods. It has long been
known that heavy-tailed PLDA (HT-PLDA), applied without length normalization,
gives similar accuracy, but at considerable extra computational cost. We have
recently introduced a fast scoring algorithm for a discriminatively trained
HT-PLDA backend. This paper extends that work by introducing a fast,
variational Bayes, generative training algorithm. We compare old and new
backends, with and without length-normalization, with i-vectors and x-vectors,
on SRE'10, SRE'16 and SITW.
|
Handling Adversarial Concept Drift in Streaming Data | Classifiers operating in a dynamic, real world environment, are vulnerable to
adversarial activity, which causes the data distribution to change over time.
These changes are traditionally referred to as concept drift, and several
approaches have been developed in literature to deal with the problem of drift
handling and detection. However, most concept drift handling techniques,
approach it as a domain independent task, to make them applicable to a wide
gamut of reactive systems. These techniques were developed from an adversarial
agnostic perspective, where they are naive and assume that drift is a benign
change, which can be fixed by updating the model. However, this is not the case
when an active adversary is trying to evade the deployed classification system.
In such an environment, the properties of concept drift are unique, as the
drift is intended to degrade the system and at the same time designed to avoid
detection by traditional concept drift detection techniques. This special
category of drift is termed as adversarial drift, and this paper analyzes its
characteristics and impact, in a streaming environment. A novel framework for
dealing with adversarial concept drift is proposed, called the Predict-Detect
streaming framework. Experimental evaluation of the framework, on generated
adversarial drifting data streams, demonstrates that this framework is able to
provide reliable unsupervised indication of drift, and is able to recover from
drifts swiftly. While traditional partially labeled concept drift detection
methodologies fail to detect adversarial drifts, the proposed framework is able
to detect such drifts and operates with <6% labeled data, on average. Also, the
framework provides benefits for active learning over imbalanced data streams,
by innately providing for feature space honeypots, where minority class
adversarial samples may be captured.
|
A Dynamic-Adversarial Mining Approach to the Security of Machine
Learning | Operating in a dynamic real world environment requires a forward thinking and
adversarial aware design for classifiers, beyond fitting the model to the
training data. In such scenarios, it is necessary to make classifiers - a)
harder to evade, b) easier to detect changes in the data distribution over
time, and c) be able to retrain and recover from model degradation. While most
works in the security of machine learning has concentrated on the evasion
resistance (a) problem, there is little work in the areas of reacting to
attacks (b and c). Additionally, while streaming data research concentrates on
the ability to react to changes to the data distribution, they often take an
adversarial agnostic view of the security problem. This makes them vulnerable
to adversarial activity, which is aimed towards evading the concept drift
detection mechanism itself. In this paper, we analyze the security of machine
learning, from a dynamic and adversarial aware perspective. The existing
techniques of Restrictive one class classifier models, Complex learning models
and Randomization based ensembles, are shown to be myopic as they approach
security as a static task. These methodologies are ill suited for a dynamic
environment, as they leak excessive information to an adversary, who can
subsequently launch attacks which are indistinguishable from the benign data.
Based on empirical vulnerability analysis against a sophisticated adversary, a
novel feature importance hiding approach for classifier design, is proposed.
The proposed design ensures that future attacks on classifiers can be detected
and recovered from. The proposed work presents motivation, by serving as a
blueprint, for future work in the area of Dynamic-Adversarial mining, which
combines lessons learned from Streaming data mining, Adversarial learning and
Cybersecurity.
|
Security Theater: On the Vulnerability of Classifiers to Exploratory
Attacks | The increasing scale and sophistication of cyberattacks has led to the
adoption of machine learning based classification techniques, at the core of
cybersecurity systems. These techniques promise scale and accuracy, which
traditional rule or signature based methods cannot. However, classifiers
operating in adversarial domains are vulnerable to evasion attacks by an
adversary, who is capable of learning the behavior of the system by employing
intelligently crafted probes. Classification accuracy in such domains provides
a false sense of security, as detection can easily be evaded by carefully
perturbing the input samples. In this paper, a generic data driven framework is
presented, to analyze the vulnerability of classification systems to black box
probing based attacks. The framework uses an exploration exploitation based
strategy, to understand an adversary's point of view of the attack defense
cycle. The adversary assumes a black box model of the defender's classifier and
can launch indiscriminate attacks on it, without information of the defender's
model type, training data or the domain of application. Experimental evaluation
on 10 real world datasets demonstrates that even models having high perceived
accuracy (>90%), by a defender, can be effectively circumvented with a high
evasion rate (>95%, on average). The detailed attack algorithms, adversarial
model and empirical evaluation, serve.
|
Balanced Random Survival Forests for Extremely Unbalanced, Right
Censored Data | Accuracies of survival models for life expectancy prediction as well as
critical-care applications are significantly compromised due to the sparsity of
samples and extreme imbalance between the survival (usually, the majority) and
mortality class sizes. While a recent random survival forest (RSF) model
overcomes the limitations of the proportional hazard assumption, an imbalance
in the data results in an underestimation (overestimation) of the hazard of the
mortality (survival) classes. A balanced random survival forests (BRSF) model,
based on training the RSF model with data generated from a synthetic minority
sampling scheme is presented to address this gap. Theoretical results on the
effect of balancing on prediction accuracies in BRSF are reported. Benchmarking
studies were conducted using five datasets with different levels of class
imbalance from public repositories and an imbalanced dataset of 267 acute
cardiac patients, collected at the Heart, Artery, and Vein Center of Fresno,
CA. Investigations suggest that BRSF provides an improved discriminatory
strength between the survival and the mortality classes. It outperformed both
optimized Cox (without and with balancing) and RSF with an average reduction of
55\% in the prediction error over the next best alternative.
|
Unsupervised Domain Adaptation: from Simulation Engine to the RealWorld | Large-scale labeled training datasets have enabled deep neural networks to
excel on a wide range of benchmark vision tasks. However, in many applications
it is prohibitively expensive or time-consuming to obtain large quantities of
labeled data. To cope with limited labeled training data, many have attempted
to directly apply models trained on a large-scale labeled source domain to
another sparsely labeled target domain. Unfortunately, direct transfer across
domains often performs poorly due to domain shift and dataset bias. Domain
adaptation is the machine learning paradigm that aims to learn a model from a
source domain that can perform well on a different (but related) target domain.
In this paper, we summarize and compare the latest unsupervised domain
adaptation methods in computer vision applications. We classify the non-deep
approaches into sample re-weighting and intermediate subspace transformation
categories, while the deep strategy includes discrepancy-based methods,
adversarial generative models, adversarial discriminative models and
reconstruction-based methods. We also discuss some potential directions.
|
Finite-Data Performance Guarantees for the Output-Feedback Control of an
Unknown System | As the systems we control become more complex, first-principle modeling
becomes either impossible or intractable, motivating the use of machine
learning techniques for the control of systems with continuous action spaces.
As impressive as the empirical success of these methods have been, strong
theoretical guarantees of performance, safety, or robustness are few and far
between. This paper takes a step towards such providing such guarantees by
establishing finite-data performance guarantees for the robust output-feedback
control of an unknown FIR SISO system. In particular, we introduce the
"Coarse-ID control" pipeline, which is composed of a system identification step
followed by a robust controller synthesis procedure, and analyze its end-to-end
performance, providing quantitative bounds on the performance degradation
suffered due to model uncertainty as a function of the number of experiments
run to identify the system. We conclude with numerical examples demonstrating
the effectiveness of our method.
|
Bernoulli Embeddings for Graphs | Just as semantic hashing can accelerate information retrieval, binary valued
embeddings can significantly reduce latency in the retrieval of graphical data.
We introduce a simple but effective model for learning such binary vectors for
nodes in a graph. By imagining the embeddings as independent coin flips of
varying bias, continuous optimization techniques can be applied to the
approximate expected loss. Embeddings optimized in this fashion consistently
outperform the quantization of both spectral graph embeddings and various
learned real-valued embeddings, on both ranking and pre-ranking tasks for a
variety of datasets.
|
Goldbach's Function Approximation Using Deep Learning | Goldbach conjecture is one of the most famous open mathematical problems. It
states that every even number, bigger than two, can be presented as a sum of 2
prime numbers. % In this work we present a deep learning based model that
predicts the number of Goldbach partitions for a given even number.
Surprisingly, our model outperforms all state-of-the-art analytically derived
estimations for the number of couples, while not requiring prime factorization
of the given number. We believe that building a model that can accurately
predict the number of couples brings us one step closer to solving one of the
world most famous open problems. To the best of our knowledge, this is the
first attempt to consider machine learning based data-driven methods to
approximate open mathematical problems in the field of number theory, and hope
that this work will encourage such attempts.
|
Synthesizing Skeletons for Reactive Systems | We present an analysis technique for temporal specifications of reactive
systems that identifies, on the level of individual system outputs over time,
which parts of the implementation are determined by the specification, and
which parts are still open. This information is represented in the form of a
labeled transition system, which we call skeleton. Each state of the skeleton
is labeled with a three-valued assignment to the output variables: each output
can be true, false, or open, where true or false means that the value must be
true or false, respectively, and open means that either value is still
possible. We present algorithms for the verification of skeletons and for the
learning-based synthesis of skeletons from specifications in linear-time
temporal logic (LTL). The algorithm returns a skeleton that satisfies the given
LTL specification in time polynomial in the size of the minimal skeleton. Our
new analysis technique can be used to recognize and repair specifications that
underspecify critical situations. The technique thus complements existing
methods for the recognition and repair of overspecifications via the
identification of unrealizable cores.
|
SUNLayer: Stable denoising with generative networks | It has been experimentally established that deep neural networks can be used
to produce good generative models for real world data. It has also been
established that such generative models can be exploited to solve classical
inverse problems like compressed sensing and super resolution. In this work we
focus on the classical signal processing problem of image denoising. We propose
a theoretical setting that uses spherical harmonics to identify what
mathematical properties of the activation functions will allow signal denoising
with local methods.
|
Stabilizing Gradients for Deep Neural Networks via Efficient SVD
Parameterization | Vanishing and exploding gradients are two of the main obstacles in training
deep neural networks, especially in capturing long range dependencies in
recurrent neural networks~(RNNs). In this paper, we present an efficient
parametrization of the transition matrix of an RNN that allows us to stabilize
the gradients that arise in its training. Specifically, we parameterize the
transition matrix by its singular value decomposition(SVD), which allows us to
explicitly track and control its singular values. We attain efficiency by using
tools that are common in numerical linear algebra, namely Householder
reflectors for representing the orthogonal matrices that arise in the SVD. By
explicitly controlling the singular values, our proposed Spectral-RNN method
allows us to easily solve the exploding gradient problem and we observe that it
empirically solves the vanishing gradient issue to a large extent. We note that
the SVD parameterization can be used for any rectangular weight matrix, hence
it can be easily extended to any deep neural network, such as a multi-layer
perceptron. Theoretically, we demonstrate that our parameterization does not
lose any expressive power, and show how it controls generalization of RNN for
the classification task. %, and show how it potentially makes the optimization
process easier. Our extensive experimental results also demonstrate that the
proposed framework converges faster, and has good generalization, especially in
capturing long range dependencies, as shown on the synthetic addition and copy
tasks, as well as on MNIST and Penn Tree Bank data sets.
|
Logistic Regression: The Importance of Being Improper | Learning linear predictors with the logistic loss---both in stochastic and
online settings---is a fundamental task in machine learning and statistics,
with direct connections to classification and boosting. Existing "fast rates"
for this setting exhibit exponential dependence on the predictor norm, and
Hazan et al. (2014) showed that this is unfortunately unimprovable. Starting
with the simple observation that the logistic loss is $1$-mixable, we design a
new efficient improper learning algorithm for online logistic regression that
circumvents the aforementioned lower bound with a regret bound exhibiting a
doubly-exponential improvement in dependence on the predictor norm. This
provides a positive resolution to a variant of the COLT 2012 open problem of
McMahan and Streeter (2012) when improper learning is allowed. This improvement
is obtained both in the online setting and, with some extra work, in the batch
statistical setting with high probability. We also show that the improved
dependence on predictor norm is near-optimal.
Leveraging this improved dependency on the predictor norm yields the
following applications: (a) we give algorithms for online bandit multiclass
learning with the logistic loss with an $\tilde{O}(\sqrt{n})$ relative mistake
bound across essentially all parameter ranges, thus providing a solution to the
COLT 2009 open problem of Abernethy and Rakhlin (2009), and (b) we give an
adaptive algorithm for online multiclass boosting with optimal sample
complexity, thus partially resolving an open problem of Beygelzimer et al.
(2015) and Jung et al. (2017). Finally, we give information-theoretic bounds on
the optimal rates for improper logistic regression with general function
classes, thereby characterizing the extent to which our improvement for linear
classes extends to other parametric and even nonparametric settings.
|
Stochastic bandits robust to adversarial corruptions | We introduce a new model of stochastic bandits with adversarial corruptions
which aims to capture settings where most of the input follows a stochastic
pattern but some fraction of it can be adversarially changed to trick the
algorithm, e.g., click fraud, fake reviews and email spam. The goal of this
model is to encourage the design of bandit algorithms that (i) work well in
mixed adversarial and stochastic models, and (ii) whose performance
deteriorates gracefully as we move from fully stochastic to fully adversarial
models.
In our model, the rewards for all arms are initially drawn from a
distribution and are then altered by an adaptive adversary. We provide a simple
algorithm whose performance gracefully degrades with the total corruption the
adversary injected in the data, measured by the sum across rounds of the
biggest alteration the adversary made in the data in that round; this total
corruption is denoted by $C$. Our algorithm provides a guarantee that retains
the optimal guarantee (up to a logarithmic term) if the input is stochastic and
whose performance degrades linearly to the amount of corruption $C$, while
crucially being agnostic to it. We also provide a lower bound showing that this
linear degradation is necessary if the algorithm achieves optimal performance
in the stochastic setting (the lower bound works even for a known amount of
corruption, a special case in which our algorithm achieves optimal performance
without the extra logarithm).
|
Neural Nets via Forward State Transformation and Backward Loss
Transformation | This article studies (multilayer perceptron) neural networks with an emphasis
on the transformations involved --- both forward and backward --- in order to
develop a semantical/logical perspective that is in line with standard program
semantics. The common two-pass neural network training algorithms make this
viewpoint particularly fitting. In the forward direction, neural networks act
as state transformers. In the reverse direction, however, neural networks
change losses of outputs to losses of inputs, thereby acting like a
(real-valued) predicate transformer. In this way, backpropagation is functorial
by construction, as shown earlier in recent other work. We illustrate this
perspective by training a simple instance of a neural network.
|
On the Local Minima of the Empirical Risk | Population risk is always of primary interest in machine learning; however,
learning algorithms only have access to the empirical risk. Even for
applications with nonconvex nonsmooth losses (such as modern deep networks),
the population risk is generally significantly more well-behaved from an
optimization point of view than the empirical risk. In particular, sampling can
create many spurious local minima. We consider a general framework which aims
to optimize a smooth nonconvex function $F$ (population risk) given only access
to an approximation $f$ (empirical risk) that is pointwise close to $F$ (i.e.,
$\|F-f\|_{\infty} \le \nu$). Our objective is to find the
$\epsilon$-approximate local minima of the underlying function $F$ while
avoiding the shallow local minima---arising because of the tolerance
$\nu$---which exist only in $f$. We propose a simple algorithm based on
stochastic gradient descent (SGD) on a smoothed version of $f$ that is
guaranteed to achieve our goal as long as $\nu \le O(\epsilon^{1.5}/d)$. We
also provide an almost matching lower bound showing that our algorithm achieves
optimal error tolerance $\nu$ among all algorithms making a polynomial number
of queries of $f$. As a concrete example, we show that our results can be
directly used to give sample complexities for learning a ReLU unit.
|
Generalized Hadamard-Product Fusion Operators for Visual Question
Answering | We propose a generalized class of multimodal fusion operators for the task of
visual question answering (VQA). We identify generalizations of existing
multimodal fusion operators based on the Hadamard product, and show that
specific non-trivial instantiations of this generalized fusion operator exhibit
superior performance in terms of OpenEnded accuracy on the VQA task. In
particular, we introduce Nonlinearity Ensembling, Feature Gating, and
post-fusion neural network layers as fusion operator components, culminating in
an absolute percentage point improvement of $1.1\%$ on the VQA 2.0 test-dev set
over baseline fusion operators, which use the same features as input. We use
our findings as evidence that our generalized class of fusion operators could
lead to the discovery of even superior task-specific operators when used as a
search space in an architecture search over fusion operators.
|
Online Second Order Methods for Non-Convex Stochastic Optimizations | This paper proposes a family of online second order methods for possibly
non-convex stochastic optimizations based on the theory of preconditioned
stochastic gradient descent (PSGD), which can be regarded as an enhance
stochastic Newton method with the ability to handle gradient noise and
non-convexity simultaneously. We have improved the implementations of the
original PSGD in several ways, e.g., new forms of preconditioners, more
accurate Hessian vector product calculations, and better numerical stability
with vanishing or ill-conditioned Hessian, etc.. We also have unrevealed the
relationship between feature normalization and PSGD with Kronecker product
preconditioners, which explains the excellent performance of Kronecker product
preconditioners in deep neural network learning. A software package
(https://github.com/lixilinx/psgd_tf) implemented in Tensorflow is provided to
compare variations of stochastic gradient descent (SGD) and PSGD with five
different preconditioners on a wide range of benchmark problems with commonly
used neural network architectures, e.g., convolutional and recurrent neural
networks. Experimental results clearly demonstrate the advantages of PSGD in
terms of generalization performance and convergence speed.
|
A Systematic Comparison of Deep Learning Architectures in an Autonomous
Vehicle | Self-driving technology is advancing rapidly --- albeit with significant
challenges and limitations. This progress is largely due to recent developments
in deep learning algorithms. To date, however, there has been no systematic
comparison of how different deep learning architectures perform at such tasks,
or an attempt to determine a correlation between classification performance and
performance in an actual vehicle, a potentially critical factor in developing
self-driving systems. Here, we introduce the first controlled comparison of
multiple deep-learning architectures in an end-to-end autonomous driving task
across multiple testing conditions. We compared performance, under identical
driving conditions, across seven architectures including a fully-connected
network, a simple 2 layer CNN, AlexNet, VGG-16, Inception-V3, ResNet, and an
LSTM by assessing the number of laps each model was able to successfully
complete without crashing while traversing an indoor racetrack. We compared
performance across models when the conditions exactly matched those in training
as well as when the local environment and track were configured differently and
objects that were not included in the training dataset were placed on the track
in various positions. In addition, we considered performance using several
different data types for training and testing including single grayscale and
color frames, and multiple grayscale frames stacked together in sequence. With
the exception of a fully-connected network, all models performed reasonably
well (around or above 80\%) and most very well (~95\%) on at least one input
type but with considerable variation across models and inputs. Overall,
AlexNet, operating on single color frames as input, achieved the best level of
performance (100\% success rate in phase one and 55\% in phase two) while
VGG-16 performed well most consistently across image types.
|
Clipping free attacks against artificial neural networks | During the last years, a remarkable breakthrough has been made in AI domain
thanks to artificial deep neural networks that achieved a great success in many
machine learning tasks in computer vision, natural language processing, speech
recognition, malware detection and so on. However, they are highly vulnerable
to easily crafted adversarial examples. Many investigations have pointed out
this fact and different approaches have been proposed to generate attacks while
adding a limited perturbation to the original data. The most robust known
method so far is the so called C&W attack [1]. Nonetheless, a countermeasure
known as feature squeezing coupled with ensemble defense showed that most of
these attacks can be destroyed [6]. In this paper, we present a new method we
call Centered Initial Attack (CIA) whose advantage is twofold : first, it
insures by construction the maximum perturbation to be smaller than a threshold
fixed beforehand, without the clipping process that degrades the quality of
attacks. Second, it is robust against recently introduced defenses such as
feature squeezing, JPEG encoding and even against a voting ensemble of
defenses. While its application is not limited to images, we illustrate this
using five of the current best classifiers on ImageNet dataset among which two
are adversarialy retrained on purpose to be robust against attacks. With a
fixed maximum perturbation of only 1.5% on any pixel, around 80% of attacks
(targeted) fool the voting ensemble defense and nearly 100% when the
perturbation is only 6%. While this shows how it is difficult to defend against
CIA attacks, the last section of the paper gives some guidelines to limit their
impact.
|
code2vec: Learning Distributed Representations of Code | We present a neural model for representing snippets of code as continuous
distributed vectors ("code embeddings"). The main idea is to represent a code
snippet as a single fixed-length $\textit{code vector}$, which can be used to
predict semantic properties of the snippet. This is performed by decomposing
code to a collection of paths in its abstract syntax tree, and learning the
atomic representation of each path $\textit{simultaneously}$ with learning how
to aggregate a set of them. We demonstrate the effectiveness of our approach by
using it to predict a method's name from the vector representation of its body.
We evaluate our approach by training a model on a dataset of 14M methods. We
show that code vectors trained on this dataset can predict method names from
files that were completely unobserved during training. Furthermore, we show
that our model learns useful method name vectors that capture semantic
similarities, combinations, and analogies. Comparing previous techniques over
the same data set, our approach obtains a relative improvement of over 75%,
being the first to successfully predict method names based on a large,
cross-project, corpus. Our trained model, visualizations and vector
similarities are available as an interactive online demo at
http://code2vec.org. The code, data, and trained models are available at
https://github.com/tech-srl/code2vec.
|
Fr\'echet ChemNet Distance: A metric for generative models for molecules
in drug discovery | The new wave of successful generative models in machine learning has
increased the interest in deep learning driven de novo drug design. However,
assessing the performance of such generative models is notoriously difficult.
Metrics that are typically used to assess the performance of such generative
models are the percentage of chemically valid molecules or the similarity to
real molecules in terms of particular descriptors, such as the partition
coefficient (logP) or druglikeness. However, method comparison is difficult
because of the inconsistent use of evaluation metrics, the necessity for
multiple metrics, and the fact that some of these measures can easily be
tricked by simple rule-based systems. We propose a novel distance measure
between two sets of molecules, called Fr\'echet ChemNet distance (FCD), that
can be used as an evaluation metric for generative models. The FCD is similar
to a recently established performance metric for comparing image generation
methods, the Fr\'echet Inception Distance (FID). Whereas the FID uses one of
the hidden layers of InceptionNet, the FCD utilizes the penultimate layer of a
deep neural network called ChemNet, which was trained to predict drug
activities. Thus, the FCD metric takes into account chemically and biologically
relevant information about molecules, and also measures the diversity of the
set via the distribution of generated molecules. The FCD's advantage over
previous metrics is that it can detect if generated molecules are a) diverse
and have similar b) chemical and c) biological properties as real molecules. We
further provide an easy-to-use implementation that only requires the SMILES
representation of the generated molecules as input to calculate the FCD.
Implementations are available at: https://www.github.com/bioinf-jku/FCD
|
A Provably Correct Algorithm for Deep Learning that Actually Works | We describe a layer-by-layer algorithm for training deep convolutional
networks, where each step involves gradient updates for a two layer network
followed by a simple clustering algorithm. Our algorithm stems from a deep
generative model that generates mages level by level, where lower resolution
images correspond to latent semantic classes. We analyze the convergence rate
of our algorithm assuming that the data is indeed generated according to this
model (as well as additional assumptions). While we do not pretend to claim
that the assumptions are realistic for natural images, we do believe that they
capture some true properties of real data. Furthermore, we show that our
algorithm actually works in practice (on the CIFAR dataset), achieving results
in the same ballpark as that of vanilla convolutional neural networks that are
being trained by stochastic gradient descent. Finally, our proof techniques may
be of independent interest.
|
Deep Representation for Patient Visits from Electronic Health Records | We show how to learn low-dimensional representations (embeddings) of patient
visits from the corresponding electronic health record (EHR) where
International Classification of Diseases (ICD) diagnosis codes are removed. We
expect that these embeddings will be useful for the construction of predictive
statistical models anticipated to drive personalized medicine and improve
healthcare quality. These embeddings are learned using a deep neural network
trained to predict ICD diagnosis categories. We show that our embeddings
capture relevant clinical informations and can be used directly as input to
standard machine learning algorithms like multi-output classifiers for ICD code
prediction. We also show that important medical informations correspond to
particular directions in our embedding space.
|
On Matching Pursuit and Coordinate Descent | Two popular examples of first-order optimization methods over linear spaces
are coordinate descent and matching pursuit algorithms, with their randomized
variants. While the former targets the optimization by moving along
coordinates, the latter considers a generalized notion of directions.
Exploiting the connection between the two algorithms, we present a unified
analysis of both, providing affine invariant sublinear $\mathcal{O}(1/t)$ rates
on smooth objectives and linear convergence on strongly convex objectives. As a
byproduct of our affine invariant analysis of matching pursuit, our rates for
steepest coordinate descent are the tightest known. Furthermore, we show the
first accelerated convergence rate $\mathcal{O}(1/t^2)$ for matching pursuit
and steepest coordinate descent on convex objectives.
|
A General Path-Based Representation for Predicting Program Properties | Predicting program properties such as names or expression types has a wide
range of applications. It can ease the task of programming and increase
programmer productivity. A major challenge when learning from programs is
$\textit{how to represent programs in a way that facilitates effective
learning}$.
We present a $\textit{general path-based representation}$ for learning from
programs. Our representation is purely syntactic and extracted automatically.
The main idea is to represent a program using paths in its abstract syntax tree
(AST). This allows a learning model to leverage the structured nature of code
rather than treating it as a flat sequence of tokens.
We show that this representation is general and can: (i) cover different
prediction tasks, (ii) drive different learning algorithms (for both generative
and discriminative models), and (iii) work across different programming
languages.
We evaluate our approach on the tasks of predicting variable names, method
names, and full types. We use our representation to drive both CRF-based and
word2vec-based learning, for programs of four languages: JavaScript, Java,
Python and C\#. Our evaluation shows that our approach obtains better results
than task-specific handcrafted representations across different tasks and
programming languages.
|
Calibrated Prediction Intervals for Neural Network Regressors | Ongoing developments in neural network models are continually advancing the
state of the art in terms of system accuracy. However, the predicted labels
should not be regarded as the only core output; also important is a
well-calibrated estimate of the prediction uncertainty. Such estimates and
their calibration are critical in many practical applications. Despite their
obvious aforementioned advantage in relation to accuracy, contemporary neural
networks can, generally, be regarded as poorly calibrated and as such do not
produce reliable output probability estimates. Further, while post-processing
calibration solutions can be found in the relevant literature, these tend to be
for systems performing classification. In this regard, we herein present two
novel methods for acquiring calibrated predictions intervals for neural network
regressors: empirical calibration and temperature scaling. In experiments using
different regression tasks from the audio and computer vision domains, we find
that both our proposed methods are indeed capable of producing calibrated
prediction intervals for neural network regressors with any desired confidence
level, a finding that is consistent across all datasets and neural network
architectures we experimented with. In addition, we derive an additional
practical recommendation for producing more accurate calibrated prediction
intervals. We release the source code implementing our proposed methods for
computing calibrated predicted intervals. The code for computing calibrated
predicted intervals is publicly available.
|
Collaborative Filtering with Topic and Social Latent Factors
Incorporating Implicit Feedback | Recommender systems (RSs) provide an effective way of alleviating the
information overload problem by selecting personalized items for different
users. Latent factors based collaborative filtering (CF) has become the popular
approaches for RSs due to its accuracy and scalability. Recently, online social
networks and user-generated content provide diverse sources for recommendation
beyond ratings. Although {\em social matrix factorization} (Social MF) and {\em
topic matrix factorization} (Topic MF) successfully exploit social relations
and item reviews, respectively, both of them ignore some useful information. In
this paper, we investigate the effective data fusion by combining the
aforementioned approaches. First, we propose a novel model {\em \mbox{MR3}} to
jointly model three sources of information (i.e., ratings, item reviews, and
social relations) effectively for rating prediction by aligning the latent
factors and hidden topics. Second, we incorporate the implicit feedback from
ratings into the proposed model to enhance its capability and to demonstrate
its flexibility. We achieve more accurate rating prediction on real-life
datasets over various state-of-the-art methods. Furthermore, we measure the
contribution from each of the three data sources and the impact of implicit
feedback from ratings, followed by the sensitivity analysis of hyperparameters.
Empirical studies demonstrate the effectiveness and efficacy of our proposed
model and its extension.
|
Why Comparing Single Performance Scores Does Not Allow to Draw
Conclusions About Machine Learning Approaches | Developing state-of-the-art approaches for specific tasks is a major driving
force in our research community. Depending on the prestige of the task,
publishing it can come along with a lot of visibility. The question arises how
reliable are our evaluation methodologies to compare approaches?
One common methodology to identify the state-of-the-art is to partition data
into a train, a development and a test set. Researchers can train and tune
their approach on some part of the dataset and then select the model that
worked best on the development set for a final evaluation on unseen test data.
Test scores from different approaches are compared, and performance differences
are tested for statistical significance.
In this publication, we show that there is a high risk that a statistical
significance in this type of evaluation is not due to a superior learning
approach. Instead, there is a high risk that the difference is due to chance.
For example for the CoNLL 2003 NER dataset we observed in up to 26% of the
cases type I errors (false positives) with a threshold of p < 0.05, i.e.,
falsely concluding a statistically significant difference between two identical
approaches.
We prove that this evaluation setup is unsuitable to compare learning
approaches. We formalize alternative evaluation setups based on score
distributions.
|
Evaluation of Session-based Recommendation Algorithms | Recommender systems help users find relevant items of interest, for example
on e-commerce or media streaming sites. Most academic research is concerned
with approaches that personalize the recommendations according to long-term
user profiles. In many real-world applications, however, such long-term
profiles often do not exist and recommendations therefore have to be made
solely based on the observed behavior of a user during an ongoing session.
Given the high practical relevance of the problem, an increased interest in
this problem can be observed in recent years, leading to a number of proposals
for session-based recommendation algorithms that typically aim to predict the
user's immediate next actions. In this work, we present the results of an
in-depth performance comparison of a number of such algorithms, using a variety
of datasets and evaluation measures. Our comparison includes the most recent
approaches based on recurrent neural networks like GRU4REC, factorized Markov
model approaches such as FISM or FOSSIL, as well as simpler methods based,
e.g., on nearest neighbor schemes. Our experiments reveal that algorithms of
this latter class, despite their sometimes almost trivial nature, often perform
equally well or significantly better than today's more complex approaches based
on deep neural networks. Our results therefore suggest that there is
substantial room for improvement regarding the development of more
sophisticated session-based recommendation algorithms.
|
Similarity based hierarchical clustering of physiological parameters for
the identification of health states - a feasibility study | This paper introduces a new unsupervised method for the clustering of
physiological data into health states based on their similarity. We propose an
iterative hierarchical clustering approach that combines health states
according to a similarity constraint to new arbitrary health states. We applied
method to experimental data in which the physical strain of subjects was
systematically varied. We derived health states based on parameters extracted
from ECG data. The occurrence of health states shows a high temporal
correlation to the experimental phases of the physical exercise. We compared
our method to other clustering algorithms and found a significantly higher
accuracy with respect to the identification of health states.
|
Learning the Multiple Traveling Salesmen Problem with Permutation
Invariant Pooling Networks | While there are optimal TSP solvers, as well as recent learning-based
approaches, the generalization of the TSP to the Multiple Traveling Salesmen
Problem is much less studied. Here, we design a neural network solution that
treats the salesmen, cities and depot as three different sets of varying
cardinalities. We apply a novel technique that combines elements from recent
architectures that were developed for sets, as well as elements from graph
networks. Coupled with new constraint enforcing output layers, a dedicated
loss, and a search method, our solution is shown to outperform all the
meta-heuristics of the leading solver in the field.
|
Metric Learning with Dynamically Generated Pairwise Constraints for Ear
Recognition | Ear recognition task is known as predicting whether two ear images belong to
the same person or not. In this paper, we present a novel metric learning
method for ear recognition. This method is formulated as a pairwise constrained
optimization problem. In each training cycle, this method selects the nearest
similar and dissimilar neighbors of each sample to construct the pairwise
constraints, and then solve the optimization problem by the iterated Bregman
projections. Experiments are conducted on AMI, USTB II and WPUT databases. The
results show that the proposed approach can achieve promising recognition rates
in ear recognition, and its training process is much more efficient than the
other competing metric learning methods.
|
On the Limitation of Local Intrinsic Dimensionality for Characterizing
the Subspaces of Adversarial Examples | Understanding and characterizing the subspaces of adversarial examples aid in
studying the robustness of deep neural networks (DNNs) to adversarial
perturbations. Very recently, Ma et al. (ICLR 2018) proposed to use local
intrinsic dimensionality (LID) in layer-wise hidden representations of DNNs to
study adversarial subspaces. It was demonstrated that LID can be used to
characterize the adversarial subspaces associated with different attack
methods, e.g., the Carlini and Wagner's (C&W) attack and the fast gradient sign
attack.
In this paper, we use MNIST and CIFAR-10 to conduct two new sets of
experiments that are absent in existing LID analysis and report the limitation
of LID in characterizing the corresponding adversarial subspaces, which are (i)
oblivious attacks and LID analysis using adversarial examples with different
confidence levels; and (ii) black-box transfer attacks. For (i), we find that
the performance of LID is very sensitive to the confidence parameter deployed
by an attack, and the LID learned from ensembles of adversarial examples with
varying confidence levels surprisingly gives poor performance. For (ii), we
find that when adversarial examples are crafted from another DNN model, LID is
ineffective in characterizing their adversarial subspaces. These two findings
together suggest the limited capability of LID in characterizing the subspaces
of adversarial examples.
|
BAGAN: Data Augmentation with Balancing GAN | Image classification datasets are often imbalanced, characteristic that
negatively affects the accuracy of deep-learning classifiers. In this work we
propose balancing GAN (BAGAN) as an augmentation tool to restore balance in
imbalanced datasets. This is challenging because the few minority-class images
may not be enough to train a GAN. We overcome this issue by including during
the adversarial training all available images of majority and minority classes.
The generative model learns useful features from majority classes and uses
these to generate images for minority classes. We apply class conditioning in
the latent space to drive the generation process towards a target class. The
generator in the GAN is initialized with the encoder module of an autoencoder
that enables us to learn an accurate class-conditioning in the latent space. We
compare the proposed methodology with state-of-the-art GANs and demonstrate
that BAGAN generates images of superior quality when trained with an imbalanced
dataset.
|
Flow From Motion: A Deep Learning Approach | Wearable devices have the potential to enhance sports performance, yet they
are not fulfilling this promise. Our previous studies with 6 professional
tennis coaches and 20 players indicate that this could be due the lack of
psychological or mental state feedback, which the coaches claim to provide.
Towards this end, we propose to detect the flow state, mental state of optimal
performance, using wearables data to be later used in training. We performed a
study with a professional tennis coach and two players. The coach provided
labels about the players' flow state while each player had a wearable device on
their racket holding wrist. We trained multiple models using the wearables data
and the coach labels. Our deep neural network models achieved around 98%
testing accuracy for a variety of conditions. This suggests that the flow state
or what coaches recognize as flow, can be detected using wearables data in
tennis which is a novel result. The implication for the HCI community is that
having access to such information would allow for design of novel hardware and
interaction paradigms that would be helpful in professional athlete training.
|
HAMLET: Interpretable Human And Machine co-LEarning Technique | Efficient label acquisition processes are key to obtaining robust
classifiers. However, data labeling is often challenging and subject to high
levels of label noise. This can arise even when classification targets are well
defined, if instances to be labeled are more difficult than the prototypes used
to define the class, leading to disagreements among the expert community. Here,
we enable efficient training of deep neural networks. From low-confidence
labels, we iteratively improve their quality by simultaneous learning of
machines and experts. We call it Human And Machine co-LEarning Technique
(HAMLET). Throughout the process, experts become more consistent, while the
algorithm provides them with explainable feedback for confirmation. HAMLET uses
a neural embedding function and a memory module filled with diverse reference
embeddings from different classes. Its output includes classification labels
and highly relevant reference embeddings as explanation. We took the study of
brain monitoring at intensive care unit (ICU) as an application of HAMLET on
continuous electroencephalography (cEEG) data. Although cEEG monitoring yields
large volumes of data, labeling costs and difficulty make it hard to build a
classifier. Additionally, while experts agree on the labels of clear-cut
examples of cEEG patterns, labeling many real-world cEEG data can be extremely
challenging. Thus, a large minority of sequences might be mislabeled. HAMLET
has shown significant performance gain against deep learning and other
baselines, increasing accuracy from 7.03% to 68.75% on challenging inputs.
Besides improved performance, clinical experts confirmed the interpretability
of those reference embeddings in helping explaining the classification results
by HAMLET.
|
MOrdReD: Memory-based Ordinal Regression Deep Neural Networks for Time
Series Forecasting | Time series forecasting is ubiquitous in the modern world. Applications range
from health care to astronomy, and include climate modelling, financial trading
and monitoring of critical engineering equipment. To offer value over this
range of activities, models must not only provide accurate forecasts, but also
quantify and adjust their uncertainty over time. In this work, we directly
tackle this task with a novel, fully end-to-end deep learning method for time
series forecasting. By recasting time series forecasting as an ordinal
regression task, we develop a principled methodology to assess long-term
predictive uncertainty and describe rich multimodal, non-Gaussian behaviour,
which arises regularly in applied settings.
Notably, our framework is a wholly general-purpose approach that requires
little to no user intervention to be used. We showcase this key feature in a
large-scale benchmark test with 45 datasets drawn from both, a wide range of
real-world application domains, as well as a comprehensive list of synthetic
maps. This wide comparison encompasses state-of-the-art methods in both the
Machine Learning and Statistics modelling literature, such as the Gaussian
Process. We find that our approach does not only provide excellent predictive
forecasts, shadowing true future values, but also allows us to infer valuable
information, such as the predictive distribution of the occurrence of critical
events of interest, accurately and reliably even over long time horizons.
|
Domain transfer convolutional attribute embedding | In this paper, we study the problem of transfer learning with the attribute
data. In the transfer learning problem, we want to leverage the data of the
auxiliary and the target domains to build an effective model for the
classification problem in the target domain. Meanwhile, the attributes are
naturally stable cross different domains. This strongly motives us to learn
effective domain transfer attribute representations. To this end, we proposed
to embed the attributes of the data to a common space by using the powerful
convolutional neural network (CNN) model. The convolutional representations of
the data points are mapped to the corresponding attributes so that they can be
effective embedding of the attributes. We also represent the data of different
domains by a domain-independent CNN, ant a domain-specific CNN, and combine
their outputs with the attribute embedding to build the classification model.
An joint learning framework is constructed to minimize the classification
errors, the attribute mapping error, the mismatching of the domain-independent
representations cross different domains, and to encourage the the neighborhood
smoothness of representations in the target domain. The minimization problem is
solved by an iterative algorithm based on gradient descent. Experiments over
benchmark data sets of person re-identification, bankruptcy prediction, and
spam email detection, show the effectiveness of the proposed method.
|
DJAM: distributed Jacobi asynchronous method for learning personal
models | Processing data collected by a network of agents often boils down to solving
an optimization problem. The distributed nature of these problems calls for
methods that are, themselves, distributed. While most collaborative learning
problems require agents to reach a common (or consensus) model, there are
situations in which the consensus solution may not be optimal. For instance,
agents may want to reach a compromise between agreeing with their neighbors and
minimizing a personal loss function. We present DJAM, a Jacobi-like distributed
algorithm for learning personalized models. This method is
implementation-friendly: it has no hyperparameters that need tuning, it is
asynchronous, and its updates only require single-neighbor interactions. We
prove that DJAM converges with probability one to the solution, provided that
the personal loss functions are strongly convex and have Lipschitz gradient. We
then give evidence that DJAM is on par with state-of-the-art methods: our
method reaches a solution with error similar to the error of a carefully tuned
ADMM in about the same number of single-neighbor interactions.
|
Predicting the Future with Transformational States | An intelligent observer looks at the world and sees not only what is, but
what is moving and what can be moved. In other words, the observer sees how the
present state of the world can transform in the future. We propose a model that
predicts future images by learning to represent the present state and its
transformation given only a sequence of images. To do so, we introduce an
architecture with a latent state composed of two components designed to capture
(i) the present image state and (ii) the transformation between present and
future states, respectively. We couple this latent state with a recurrent
neural network (RNN) core that predicts future frames by transforming past
states into future states by applying the accumulated state transformation with
a learned operator. We describe how this model can be integrated into an
encoder-decoder convolutional neural network (CNN) architecture that uses
weighted residual connections to integrate representations of the past with
representations of the future. Qualitatively, our approach generates image
sequences that are stable and capture realistic motion over multiple predicted
frames, without requiring adversarial training. Quantitatively, our method
achieves prediction results comparable to state-of-the-art results on standard
image prediction benchmarks (Moving MNIST, KTH, and UCF101).
|
Quantum Entanglement in Deep Learning Architectures | Modern deep learning has enabled unprecedented achievements in various
domains. Nonetheless, employment of machine learning for wave function
representations is focused on more traditional architectures such as restricted
Boltzmann machines (RBMs) and fully-connected neural networks. In this letter,
we establish that contemporary deep learning architectures, in the form of deep
convolutional and recurrent networks, can efficiently represent highly
entangled quantum systems. By constructing Tensor Network equivalents of these
architectures, we identify an inherent reuse of information in the network
operation as a key trait which distinguishes them from standard Tensor Network
based representations, and which enhances their entanglement capacity. Our
results show that such architectures can support volume-law entanglement
scaling, polynomially more efficiently than presently employed RBMs. Thus,
beyond a quantification of the entanglement capacity of leading deep learning
architectures, our analysis formally motivates a shift of trending
neural-network based wave function representations closer to the
state-of-the-art in machine learning.
|
A Common Framework for Natural Gradient and Taylor based Optimisation
using Manifold Theory | This technical report constructs a theoretical framework to relate standard
Taylor approximation based optimisation methods with Natural Gradient (NG), a
method which is Fisher efficient with probabilistic models. Such a framework
will be shown to also provide mathematical justification to combine higher
order methods with the method of NG.
|
A disciplined approach to neural network hyper-parameters: Part 1 --
learning rate, batch size, momentum, and weight decay | Although deep learning has produced dazzling successes for applications of
image, speech, and video processing in the past few years, most trainings are
with suboptimal hyper-parameters, requiring unnecessarily long training times.
Setting the hyper-parameters remains a black art that requires years of
experience to acquire. This report proposes several efficient ways to set the
hyper-parameters that significantly reduce training time and improves
performance. Specifically, this report shows how to examine the training
validation/test loss function for subtle clues of underfitting and overfitting
and suggests guidelines for moving toward the optimal balance point. Then it
discusses how to increase/decrease the learning rate/momentum to speed up
training. Our experiments show that it is crucial to balance every manner of
regularization for each dataset and architecture. Weight decay is used as a
sample regularizer to show how its optimal value is tightly coupled with the
learning rates and momentums. Files to help replicate the results reported here
are available.
|
Low-Shot Learning for the Semantic Segmentation of Remote Sensing
Imagery | Recent advances in computer vision using deep learning with RGB imagery
(e.g., object recognition and detection) have been made possible thanks to the
development of large annotated RGB image datasets. In contrast, multispectral
image (MSI) and hyperspectral image (HSI) datasets contain far fewer labeled
images, in part due to the wide variety of sensors used. These annotations are
especially limited for semantic segmentation, or pixel-wise classification, of
remote sensing imagery because it is labor intensive to generate image
annotations. Low-shot learning algorithms can make effective inferences despite
smaller amounts of annotated data. In this paper, we study low-shot learning
using self-taught feature learning for semantic segmentation. We introduce 1)
an improved self-taught feature learning framework for HSI and MSI data and 2)
a semi-supervised classification algorithm. When these are combined, they
achieve state-of-the-art performance on remote sensing datasets that have
little annotated training data available. These low-shot learning frameworks
will reduce the manual image annotation burden and improve semantic
segmentation performance for remote sensing imagery.
|
A Decision Tree Approach to Predicting Recidivism in Domestic Violence | Domestic violence (DV) is a global social and public health issue that is
highly gendered. Being able to accurately predict DV recidivism, i.e.,
re-offending of a previously convicted offender, can speed up and improve risk
assessment procedures for police and front-line agencies, better protect
victims of DV, and potentially prevent future re-occurrences of DV. Previous
work in DV recidivism has employed different classification techniques,
including decision tree (DT) induction and logistic regression, where the main
focus was on achieving high prediction accuracy. As a result, even the diagrams
of trained DTs were often too difficult to interpret due to their size and
complexity, making decision-making challenging. Given there is often a
trade-off between model accuracy and interpretability, in this work our aim is
to employ DT induction to obtain both interpretable trees as well as high
prediction accuracy. Specifically, we implement and evaluate different
approaches to deal with class imbalance as well as feature selection. Compared
to previous work in DV recidivism prediction that employed logistic regression,
our approach can achieve comparable area under the ROC curve results by using
only 3 of 11 available features and generating understandable decision trees
that contain only 4 leaf nodes.
|
Bypassing Feature Squeezing by Increasing Adversary Strength | Feature Squeezing is a recently proposed defense method which reduces the
search space available to an adversary by coalescing samples that correspond to
many different feature vectors in the original space into a single sample. It
has been shown that feature squeezing defenses can be combined in a joint
detection framework to achieve high detection rates against state-of-the-art
attacks. However, we demonstrate on the MNIST and CIFAR-10 datasets that by
increasing the adversary strength of said state-of-the-art attacks, one can
bypass the detection framework with adversarial examples of minimal visual
distortion. These results suggest for proposed defenses to validate against
stronger attack configurations.
|
DRACO: Byzantine-resilient Distributed Training via Redundant Gradients | Distributed model training is vulnerable to byzantine system failures and
adversarial compute nodes, i.e., nodes that use malicious updates to corrupt
the global model stored at a parameter server (PS). To guarantee some form of
robustness, recent work suggests using variants of the geometric median as an
aggregation rule, in place of gradient averaging. Unfortunately, median-based
rules can incur a prohibitive computational overhead in large-scale settings,
and their convergence guarantees often require strong assumptions. In this
work, we present DRACO, a scalable framework for robust distributed training
that uses ideas from coding theory. In DRACO, each compute node evaluates
redundant gradients that are used by the parameter server to eliminate the
effects of adversarial updates. DRACO comes with problem-independent robustness
guarantees, and the model that it trains is identical to the one trained in the
adversary-free setup. We provide extensive experiments on real datasets and
distributed setups across a variety of large-scale models, where we show that
DRACO is several times, to orders of magnitude faster than median-based
approaches.
|
MLE-induced Likelihood for Markov Random Fields | Due to the intractable partition function, the exact likelihood function for
a Markov random field (MRF), in many situations, can only be approximated.
Major approximation approaches include pseudolikelihood and Laplace
approximation. In this paper, we propose a novel way of approximating the
likelihood function through first approximating the marginal likelihood
functions of individual parameters and then reconstructing the joint likelihood
function from these marginal likelihood functions. For approximating the
marginal likelihood functions, we derive a particular likelihood function from
a modified scenario of coin tossing which is useful for capturing how one
parameter interacts with the remaining parameters in the likelihood function.
For reconstructing the joint likelihood function, we use an appropriate copula
to link up these marginal likelihood functions. Numerical investigation
suggests the superior performance of our approach. Especially as the size of
the MRF increases, both the numerical performance and the computational cost of
our approach remain consistently satisfactory, whereas Laplace approximation
deteriorates and pseudolikelihood becomes computationally unbearable.
|
Entropy based Independent Learning in Anonymous Multi-Agent Settings | Efficient sequential matching of supply and demand is a problem of interest
in many online to offline services. For instance, Uber, Lyft, Grab for matching
taxis to customers; Ubereats, Deliveroo, FoodPanda etc for matching restaurants
to customers. In these online to offline service problems, individuals who are
responsible for supply (e.g., taxi drivers, delivery bikes or delivery van
drivers) earn more by being at the "right" place at the "right" time. We are
interested in developing approaches that learn to guide individuals to be in
the "right" place at the "right" time (to maximize revenue) in the presence of
other similar "learning" individuals and only local aggregated observation of
other agents states (e.g., only number of other taxis in same zone as current
agent).
A key characteristic of the domains of interest is that the interactions
between individuals are anonymous, i.e., the outcome of an interaction
(competing for demand) is dependent only on the number and not on the identity
of the agents. We model these problems using the Anonymous MARL (AyMARL) model.
The key contribution of this paper is in employing principle of maximum entropy
to provide a general framework of independent learning that is both empirically
effective (even with only local aggregated information of agent population
distribution) and theoretically justified.
Finally, our approaches provide a significant improvement with respect to
joint and individual revenue on a generic simulator for online to offline
services and a real world taxi problem over existing approaches. More
importantly, this is achieved while having the least variance in revenues
earned by the learning individuals, an indicator of fairness.
|
Complex-Valued Restricted Boltzmann Machine for Direct Speech
Parameterization from Complex Spectra | This paper describes a novel energy-based probabilistic distribution that
represents complex-valued data and explains how to apply it to direct feature
extraction from complex-valued spectra. The proposed model, the complex-valued
restricted Boltzmann machine (CRBM), is designed to deal with complex-valued
visible units as an extension of the well-known restricted Boltzmann machine
(RBM). Like the RBM, the CRBM learns the relationships between visible and
hidden units without having connections between units in the same layer, which
dramatically improves training efficiency by using Gibbs sampling or
contrastive divergence (CD). Another important characteristic is that the CRBM
also has connections between real and imaginary parts of each of the
complex-valued visible units that help represent the data distribution in the
complex domain. In speech signal processing, classification and generation
features are often based on amplitude spectra (e.g., MFCC, cepstra, and
mel-cepstra) even if they are calculated from complex spectra, and they ignore
phase information. In contrast, the proposed feature extractor using the CRBM
directly encodes the complex spectra (or another complex-valued representation
of the complex spectra) into binary-valued latent features (hidden units).
Since the visible-hidden connections are undirected, we can also recover
(decode) the complex spectra from the latent features directly. Our speech
coding experiments demonstrated that the CRBM outperformed other speech coding
methods, such as methods using the conventional RBM, the mel-log spectrum
approximate (MLSA) decoder, etc.
|
Learning Synergies between Pushing and Grasping with Self-supervised
Deep Reinforcement Learning | Skilled robotic manipulation benefits from complex synergies between
non-prehensile (e.g. pushing) and prehensile (e.g. grasping) actions: pushing
can help rearrange cluttered objects to make space for arms and fingers;
likewise, grasping can help displace objects to make pushing movements more
precise and collision-free. In this work, we demonstrate that it is possible to
discover and learn these synergies from scratch through model-free deep
reinforcement learning. Our method involves training two fully convolutional
networks that map from visual observations to actions: one infers the utility
of pushes for a dense pixel-wise sampling of end effector orientations and
locations, while the other does the same for grasping. Both networks are
trained jointly in a Q-learning framework and are entirely self-supervised by
trial and error, where rewards are provided from successful grasps. In this
way, our policy learns pushing motions that enable future grasps, while
learning grasps that can leverage past pushes. During picking experiments in
both simulation and real-world scenarios, we find that our system quickly
learns complex behaviors amid challenging cases of clutter, and achieves better
grasping success rates and picking efficiencies than baseline alternatives
after only a few hours of training. We further demonstrate that our method is
capable of generalizing to novel objects. Qualitative results (videos), code,
pre-trained models, and simulation environments are available at
http://vpg.cs.princeton.edu
|
Reinforcement Learning for Fair Dynamic Pricing | Unfair pricing policies have been shown to be one of the most negative
perceptions customers can have concerning pricing, and may result in long-term
losses for a company. Despite the fact that dynamic pricing models help
companies maximize revenue, fairness and equality should be taken into account
in order to avoid unfair price differences between groups of customers. This
paper shows how to solve dynamic pricing by using Reinforcement Learning (RL)
techniques so that prices are maximized while keeping a balance between revenue
and fairness. We demonstrate that RL provides two main features to support
fairness in dynamic pricing: on the one hand, RL is able to learn from recent
experience, adapting the pricing policy to complex market environments; on the
other hand, it provides a trade-off between short and long-term objectives,
hence integrating fairness into the model's core. Considering these two
features, we propose the application of RL for revenue optimization, with the
additional integration of fairness as part of the learning procedure by using
Jain's index as a metric. Results in a simulated environment show a significant
improvement in fairness while at the same time maintaining optimisation of
revenue.
|
Hiding in the Crowd: A Massively Distributed Algorithm for Private
Averaging with Malicious Adversaries | The amount of personal data collected in our everyday interactions with
connected devices offers great opportunities for innovative services fueled by
machine learning, as well as raises serious concerns for the privacy of
individuals. In this paper, we propose a massively distributed protocol for a
large set of users to privately compute averages over their joint data, which
can then be used to learn predictive models. Our protocol can find a solution
of arbitrary accuracy, does not rely on a third party and preserves the privacy
of users throughout the execution in both the honest-but-curious and malicious
adversary models. Specifically, we prove that the information observed by the
adversary (the set of maliciours users) does not significantly reduce the
uncertainty in its prediction of private values compared to its prior belief.
The level of privacy protection depends on a quantity related to the Laplacian
matrix of the network graph and generally improves with the size of the graph.
Furthermore, we design a verification procedure which offers protection against
malicious users joining the service with the goal of manipulating the outcome
of the algorithm.
|
Cross-validation in high-dimensional spaces: a lifeline for
least-squares models and multi-class LDA | Least-squares models such as linear regression and Linear Discriminant
Analysis (LDA) are amongst the most popular statistical learning techniques.
However, since their computation time increases cubically with the number of
features, they are inefficient in high-dimensional neuroimaging datasets.
Fortunately, for k-fold cross-validation, an analytical approach has been
developed that yields the exact cross-validated predictions in least-squares
models without explicitly training the model. Its computation time grows with
the number of test samples. Here, this approach is systematically investigated
in the context of cross-validation and permutation testing. LDA is used
exemplarily but results hold for all other least-squares methods. Furthermore,
a non-trivial extension to multi-class LDA is formally derived. The analytical
approach is evaluated using complexity calculations, simulations, and
permutation testing of an EEG/MEG dataset. Depending on the ratio between
features and samples, the analytical approach is up to 10,000x faster than the
standard approach (retraining the model on each training set). This allows for
a fast cross-validation of least-squares models and multi-class LDA in
high-dimensional data, with obvious applications in multi-dimensional datasets,
Representational Similarity Analysis, and permutation testing.
|
Fast Parametric Learning with Activation Memorization | Neural networks trained with backpropagation often struggle to identify
classes that have been observed a small number of times. In applications where
most class labels are rare, such as language modelling, this can become a
performance bottleneck. One potential remedy is to augment the network with a
fast-learning non-parametric model which stores recent activations and class
labels into an external memory. We explore a simplified architecture where we
treat a subset of the model parameters as fast memory stores. This can help
retain information over longer time intervals than a traditional memory, and
does not require additional space or compute. In the case of image
classification, we display faster binding of novel classes on an Omniglot image
curriculum task. We also show improved performance for word-based language
models on news reports (GigaWord), books (Project Gutenberg) and Wikipedia
articles (WikiText-103) --- the latter achieving a state-of-the-art perplexity
of 29.2.
|
Automated Speed and Lane Change Decision Making using Deep Reinforcement
Learning | This paper introduces a method, based on deep reinforcement learning, for
automatically generating a general purpose decision making function. A Deep
Q-Network agent was trained in a simulated environment to handle speed and lane
change decisions for a truck-trailer combination. In a highway driving case, it
is shown that the method produced an agent that matched or surpassed the
performance of a commonly used reference model. To demonstrate the generality
of the method, the exact same algorithm was also tested by training it for an
overtaking case on a road with oncoming traffic. Furthermore, a novel way of
applying a convolutional neural network to high level input that represents
interchangeable objects is also introduced.
|
Building state-of-the-art distant speech recognition using the CHiME-4
challenge with a setup of speech enhancement baseline | This paper describes a new baseline system for automatic speech recognition
(ASR) in the CHiME-4 challenge to promote the development of noisy ASR in
speech processing communities by providing 1) state-of-the-art system with a
simplified single system comparable to the complicated top systems in the
challenge, 2) publicly available and reproducible recipe through the main
repository in the Kaldi speech recognition toolkit. The proposed system adopts
generalized eigenvalue beamforming with bidirectional long short-term memory
(LSTM) mask estimation. We also propose to use a time delay neural network
(TDNN) based on the lattice-free version of the maximum mutual information
(LF-MMI) trained with augmented all six microphones plus the enhanced data
after beamforming. Finally, we use a LSTM language model for lattice and n-best
re-scoring. The final system achieved 2.74\% WER for the real test set in the
6-channel track, which corresponds to the 2nd place in the challenge. In
addition, the proposed baseline recipe includes four different speech
enhancement measures, short-time objective intelligibility measure (STOI),
extended STOI (eSTOI), perceptual evaluation of speech quality (PESQ) and
speech distortion ratio (SDR) for the simulation test set. Thus, the recipe
also provides an experimental platform for speech enhancement studies with
these performance measures.
|
World Models | We explore building generative neural network models of popular reinforcement
learning environments. Our world model can be trained quickly in an
unsupervised manner to learn a compressed spatial and temporal representation
of the environment. By using features extracted from the world model as inputs
to an agent, we can train a very compact and simple policy that can solve the
required task. We can even train our agent entirely inside of its own
hallucinated dream generated by its world model, and transfer this policy back
into the actual environment.
An interactive version of this paper is available at
https://worldmodels.github.io/
|
Task Agnostic Continual Learning Using Online Variational Bayes | Catastrophic forgetting is the notorious vulnerability of neural networks to
the change of the data distribution while learning. This phenomenon has long
been considered a major obstacle for allowing the use of learning agents in
realistic continual learning settings. A large body of continual learning
research assumes that task boundaries are known during training. However,
research for scenarios in which task boundaries are unknown during training has
been lacking. In this paper we present, for the first time, a method for
preventing catastrophic forgetting (BGD) for scenarios with task boundaries
that are unknown during training --- task-agnostic continual learning. Code of
our algorithm is available at https://github.com/igolan/bgd.
|
Stein Points | An important task in computational statistics and machine learning is to
approximate a posterior distribution $p(x)$ with an empirical measure supported
on a set of representative points $\{x_i\}_{i=1}^n$. This paper focuses on
methods where the selection of points is essentially deterministic, with an
emphasis on achieving accurate approximation when $n$ is small. To this end, we
present `Stein Points'. The idea is to exploit either a greedy or a conditional
gradient method to iteratively minimise a kernel Stein discrepancy between the
empirical measure and $p(x)$. Our empirical results demonstrate that Stein
Points enable accurate approximation of the posterior at modest computational
cost. In addition, theoretical results are provided to establish convergence of
the method.
|
Distributed Adaptive Sampling for Kernel Matrix Approximation | Most kernel-based methods, such as kernel or Gaussian process regression,
kernel PCA, ICA, or $k$-means clustering, do not scale to large datasets,
because constructing and storing the kernel matrix $\mathbf{K}_n$ requires at
least $\mathcal{O}(n^2)$ time and space for $n$ samples. Recent works show that
sampling points with replacement according to their ridge leverage scores (RLS)
generates small dictionaries of relevant points with strong spectral
approximation guarantees for $\mathbf{K}_n$. The drawback of RLS-based methods
is that computing exact RLS requires constructing and storing the whole kernel
matrix. In this paper, we introduce SQUEAK, a new algorithm for kernel
approximation based on RLS sampling that sequentially processes the dataset,
storing a dictionary which creates accurate kernel matrix approximations with a
number of points that only depends on the effective dimension $d_{eff}(\gamma)$
of the dataset. Moreover since all the RLS estimations are efficiently
performed using only the small dictionary, SQUEAK is the first RLS sampling
algorithm that never constructs the whole matrix $\mathbf{K}_n$, runs in linear
time $\widetilde{\mathcal{O}}(nd_{eff}(\gamma)^3)$ w.r.t. $n$, and requires
only a single pass over the dataset. We also propose a parallel and distributed
version of SQUEAK that linearly scales across multiple machines, achieving
similar accuracy in as little as
$\widetilde{\mathcal{O}}(\log(n)d_{eff}(\gamma)^3)$ time.
|
Forward-Backward Reinforcement Learning | Goals for reinforcement learning problems are typically defined through
hand-specified rewards. To design such problems, developers of learning
algorithms must inherently be aware of what the task goals are, yet we often
require agents to discover them on their own without any supervision beyond
these sparse rewards. While much of the power of reinforcement learning derives
from the concept that agents can learn with little guidance, this requirement
greatly burdens the training process. If we relax this one restriction and
endow the agent with knowledge of the reward function, and in particular of the
goal, we can leverage backwards induction to accelerate training. To achieve
this, we propose training a model to learn to take imagined reversal steps from
known goal states. Rather than training an agent exclusively to determine how
to reach a goal while moving forwards in time, our approach travels backwards
to jointly predict how we got there. We evaluate our work in Gridworld and
Towers of Hanoi and empirically demonstrate that it yields better performance
than standard DDQN.
|
Demystifying Differentiable Programming: Shift/Reset the Penultimate
Backpropagator | Deep learning has seen tremendous success over the past decade in computer
vision, machine translation, and gameplay. This success rests in crucial ways
on gradient-descent optimization and the ability to learn parameters of a
neural network by backpropagating observed errors. However, neural network
architectures are growing increasingly sophisticated and diverse, which
motivates an emerging quest for even more general forms of differentiable
programming, where arbitrary parameterized computations can be trained by
gradient descent. In this paper, we take a fresh look at automatic
differentiation (AD) techniques, and especially aim to demystify the
reverse-mode form of AD that generalizes backpropagation in neural networks.
We uncover a tight connection between reverse-mode AD and delimited
continuations, which permits implementing reverse-mode AD purely via operator
overloading and without any auxiliary data structures. We further show how this
formulation of AD can be fruitfully combined with multi-stage programming
(staging), leading to a highly efficient implementation that combines the
performance benefits of deep learning frameworks based on explicit reified
computation graphs (e.g., TensorFlow) with the expressiveness of pure library
approaches (e.g., PyTorch).
|
Safe end-to-end imitation learning for model predictive control | We propose the use of Bayesian networks, which provide both a mean value and
an uncertainty estimate as output, to enhance the safety of learned control
policies under circumstances in which a test-time input differs significantly
from the training set. Our algorithm combines reinforcement learning and
end-to-end imitation learning to simultaneously learn a control policy as well
as a threshold over the predictive uncertainty of the learned model, with no
hand-tuning required. Corrective action, such as a return of control to the
model predictive controller or human expert, is taken when the uncertainty
threshold is exceeded. We validate our method on fully-observable and
vision-based partially-observable systems using cart-pole and autonomous
driving simulations using deep convolutional Bayesian neural networks. We
demonstrate that our method is robust to uncertainty resulting from varying
system dynamics as well as from partial state observability.
|
Incremental Training of Deep Convolutional Neural Networks | We propose an incremental training method that partitions the original
network into sub-networks, which are then gradually incorporated in the running
network during the training process. To allow for a smooth dynamic growth of
the network, we introduce a look-ahead initialization that outperforms the
random initialization. We demonstrate that our incremental approach reaches the
reference network baseline accuracy. Additionally, it allows to identify
smaller partitions of the original state-of-the-art network, that deliver the
same final accuracy, by using only a fraction of the global number of
parameters. This allows for a potential speedup of the training time of several
factors. We report training results on CIFAR-10 for ResNet and VGGNet.
|
Disease-Atlas: Navigating Disease Trajectories with Deep Learning | Joint models for longitudinal and time-to-event data are commonly used in
longitudinal studies to forecast disease trajectories over time. While there
are many advantages to joint modeling, the standard forms suffer from
limitations that arise from a fixed model specification, and computational
difficulties when applied to high-dimensional datasets. In this paper, we
propose a deep learning approach to address these limitations, enhancing
existing methods with the inherent flexibility and scalability of deep neural
networks, while retaining the benefits of joint modeling. Using longitudinal
data from a real-world medical dataset, we demonstrate improvements in
performance and scalability, as well as robustness in the presence of
irregularly sampled data.
|
Privacy-preserving Prediction | Ensuring differential privacy of models learned from sensitive user data is
an important goal that has been studied extensively in recent years. It is now
known that for some basic learning problems, especially those involving
high-dimensional data, producing an accurate private model requires much more
data than learning without privacy. At the same time, in many applications it
is not necessary to expose the model itself. Instead users may be allowed to
query the prediction model on their inputs only through an appropriate
interface. Here we formulate the problem of ensuring privacy of individual
predictions and investigate the overheads required to achieve it in several
standard models of classification and regression.
We first describe a simple baseline approach based on training several models
on disjoint subsets of data and using standard private aggregation techniques
to predict. We show that this approach has nearly optimal sample complexity for
(realizable) PAC learning of any class of Boolean functions. At the same time,
without strong assumptions on the data distribution, the aggregation step
introduces a substantial overhead. We demonstrate that this overhead can be
avoided for the well-studied class of thresholds on a line and for a number of
standard settings of convex regression. The analysis of our algorithm for
learning thresholds relies crucially on strong generalization guarantees that
we establish for all differentially private prediction algorithms.
|
A Study of Clustering Techniques and Hierarchical Matrix Formats for
Kernel Ridge Regression | We present memory-efficient and scalable algorithms for kernel methods used
in machine learning. Using hierarchical matrix approximations for the kernel
matrix the memory requirements, the number of floating point operations, and
the execution time are drastically reduced compared to standard dense linear
algebra routines. We consider both the general $\mathcal{H}$ matrix
hierarchical format as well as Hierarchically Semi-Separable (HSS) matrices.
Furthermore, we investigate the impact of several preprocessing and clustering
techniques on the hierarchical matrix compression. Effective clustering of the
input leads to a ten-fold increase in efficiency of the compression. The
algorithms are implemented using the STRUMPACK solver library. These results
confirm that --- with correct tuning of the hyperparameters --- classification
using kernel ridge regression with the compressed matrix does not lose
prediction accuracy compared to the exact --- not compressed --- kernel matrix
and that our approach can be extended to $\mathcal{O}(1M)$ datasets, for which
computation with the full kernel matrix becomes prohibitively expensive. We
present numerical experiments in a distributed memory environment up to 1,024
processors of the NERSC's Cori supercomputer using well-known datasets to the
machine learning community that range from dimension 8 up to 784.
|
Canonical Correlation Analysis of Datasets with a Common Source Graph | Canonical correlation analysis (CCA) is a powerful technique for discovering
whether or not hidden sources are commonly present in two (or more) datasets.
Its well-appreciated merits include dimensionality reduction, clustering,
classification, feature selection, and data fusion. The standard CCA however,
does not exploit the geometry of the common sources, which may be available
from the given data or can be deduced from (cross-) correlations. In this
paper, this extra information provided by the common sources generating the
data is encoded in a graph, and is invoked as a graph regularizer. This leads
to a novel graph-regularized CCA approach, that is termed graph (g) CCA. The
novel gCCA accounts for the graph-induced knowledge of common sources, while
minimizing the distance between the wanted canonical variables. Tailored for
diverse practical settings where the number of data is smaller than the data
vector dimensions, the dual formulation of gCCA is also developed. One such
setting includes kernels that are incorporated to account for nonlinear data
dependencies. The resultant graph-kernel (gk) CCA is also obtained in closed
form. Finally, corroborating image classification tests over several real
datasets are presented to showcase the merits of the novel linear, dual, and
kernel approaches relative to competing alternatives.
|
How Developers Iterate on Machine Learning Workflows -- A Survey of the
Applied Machine Learning Literature | Machine learning workflow development is anecdotally regarded to be an
iterative process of trial-and-error with humans-in-the-loop. However, we are
not aware of quantitative evidence corroborating this popular belief. A
quantitative characterization of iteration can serve as a benchmark for machine
learning workflow development in practice, and can aid the development of
human-in-the-loop machine learning systems. To this end, we conduct a
small-scale survey of the applied machine learning literature from five
distinct application domains. We collect and distill statistics on the role of
iteration within machine learning workflow development, and report preliminary
trends and insights from our investigation, as a starting point towards this
benchmark. Based on our findings, we finally describe desiderata for effective
and versatile human-in-the-loop machine learning systems that can cater to
users in diverse domains.
|
Classification of crystallization outcomes using deep convolutional
neural networks | The Machine Recognition of Crystallization Outcomes (MARCO) initiative has
assembled roughly half a million annotated images of macromolecular
crystallization experiments from various sources and setups. Here,
state-of-the-art machine learning algorithms are trained and tested on
different parts of this data set. We find that more than 94% of the test images
can be correctly labeled, irrespective of their experimental origin. Because
crystal recognition is key to high-density screening and the systematic
analysis of crystallization experiments, this approach opens the door to both
industrial and fundamental research applications.
|
ClickBAIT-v2: Training an Object Detector in Real-Time | Modern deep convolutional neural networks (CNNs) for image classification and
object detection are often trained offline on large static datasets. Some
applications, however, will require training in real-time on live video streams
with a human-in-the-loop. We refer to this class of problem as time-ordered
online training (ToOT). These problems will require a consideration of not only
the quantity of incoming training data, but the human effort required to
annotate and use it. We demonstrate and evaluate a system tailored to training
an object detector on a live video stream with minimal input from a human
operator. We show that we can obtain bounding box annotation from
weakly-supervised single-point clicks through interactive segmentation.
Furthermore, by exploiting the time-ordered nature of the video stream through
object tracking, we can increase the average training benefit of human
interactions by 3-4 times.
|
Smoothed Online Convex Optimization in High Dimensions via Online
Balanced Descent | We study Smoothed Online Convex Optimization, a version of online convex
optimization where the learner incurs a penalty for changing her actions
between rounds. Given a $\Omega(\sqrt{d})$ lower bound on the competitive ratio
of any online algorithm, where $d$ is the dimension of the action space, we ask
under what conditions this bound can be beaten. We introduce a novel
algorithmic framework for this problem, Online Balanced Descent (OBD), which
works by iteratively projecting the previous point onto a carefully chosen
level set of the current cost function so as to balance the switching costs and
hitting costs. We demonstrate the generality of the OBD framework by showing
how, with different choices of "balance," OBD can improve upon state-of-the-art
performance guarantees for both competitive ratio and regret, in particular,
OBD is the first algorithm to achieve a dimension-free competitive ratio, $3 +
O(1/\alpha)$, for locally polyhedral costs, where $\alpha$ measures the
"steepness" of the costs. We also prove bounds on the dynamic regret of OBD
when the balance is performed in the dual space that are dimension-free and
imply that OBD has sublinear static regret.
|
Reinforcement learning for non-prehensile manipulation: Transfer from
simulation to physical system | Reinforcement learning has emerged as a promising methodology for training
robot controllers. However, most results have been limited to simulation due to
the need for a large number of samples and the lack of automated-yet-safe data
collection methods. Model-based reinforcement learning methods provide an
avenue to circumvent these challenges, but the traditional concern has been the
mismatch between the simulator and the real world. Here, we show that control
policies learned in simulation can successfully transfer to a physical system,
composed of three Phantom robots pushing an object to various desired target
positions. We use a modified form of the natural policy gradient algorithm for
learning, applied to a carefully identified simulation model. The resulting
policies, trained entirely in simulation, work well on the physical system
without additional training. In addition, we show that training with an
ensemble of models makes the learned policies more robust to modeling errors,
thus compensating for difficulties in system identification.
|
Topic Modeling Based Multi-modal Depression Detection | Major depressive disorder is a common mental disorder that affects almost 7%
of the adult U.S. population. The 2017 Audio/Visual Emotion Challenge (AVEC)
asks participants to build a model to predict depression levels based on the
audio, video, and text of an interview ranging between 7-33 minutes. Since
averaging features over the entire interview will lose most temporal
information, how to discover, capture, and preserve useful temporal details for
such a long interview are significant challenges. Therefore, we propose a novel
topic modeling based approach to perform context-aware analysis of the
recording. Our experiments show that the proposed approach outperforms
context-unaware methods and the challenge baselines for all metrics.
|
Supervising Unsupervised Learning with Evolutionary Algorithm in Deep
Neural Network | A method to control results of gradient descent unsupervised learning in a
deep neural network by using evolutionary algorithm is proposed. To process
crossover of unsupervisedly trained models, the algorithm evaluates pointwise
fitness of individual nodes in neural network. Labeled training data is
randomly sampled and breeding process selects nodes by calculating degree of
their consistency on different sets of sampled data. This method supervises
unsupervised training by evolutionary process. We also introduce modified
Restricted Boltzmann Machine which contains repulsive force among nodes in a
neural network and it contributes to isolate network nodes each other to avoid
accidental degeneration of nodes by evolutionary process. These new methods are
applied to document classification problem and it results better accuracy than
a traditional fully supervised classifier implemented with linear regression
algorithm.
|
A Better Resource Allocation Algorithm with Semi-Bandit Feedback | We study a sequential resource allocation problem between a fixed number of
arms. On each iteration the algorithm distributes a resource among the arms in
order to maximize the expected success rate. Allocating more of the resource to
a given arm increases the probability that it succeeds, yet with a cut-off. We
follow Lattimore et al. (2014) and assume that the probability increases
linearly until it equals one, after which allocating more of the resource is
wasteful. These cut-off values are fixed and unknown to the learner. We present
an algorithm for this problem and prove a regret upper bound of $O(\log n)$
improving over the best known bound of $O(\log^2 n)$. Lower bounds we prove
show that our upper bound is tight. Simulations demonstrate the superiority of
our algorithm.
|
Graphite: Iterative Generative Modeling of Graphs | Graphs are a fundamental abstraction for modeling relational data. However,
graphs are discrete and combinatorial in nature, and learning representations
suitable for machine learning tasks poses statistical and computational
challenges. In this work, we propose Graphite, an algorithmic framework for
unsupervised learning of representations over nodes in large graphs using deep
latent variable generative models. Our model parameterizes variational
autoencoders (VAE) with graph neural networks, and uses a novel iterative graph
refinement strategy inspired by low-rank approximations for decoding. On a wide
variety of synthetic and benchmark datasets, Graphite outperforms competing
approaches for the tasks of density estimation, link prediction, and node
classification. Finally, we derive a theoretical connection between message
passing in graph neural networks and mean-field variational inference.
|
What deep learning can tell us about higher cognitive functions like
mindreading? | Can deep learning (DL) guide our understanding of computations happening in
biological brain? We will first briefly consider how DL has contributed to the
research on visual object recognition. In the main part we will assess whether
DL could also help us to clarify the computations underlying higher cognitive
functions such as Theory of Mind. In addition, we will compare the objectives
and learning signals of brains and machines, leading us to conclude that simply
scaling up the current DL algorithms will most likely not lead to human level
Theory of Mind.
|
Quantum algorithms for training Gaussian Processes | Gaussian processes (GPs) are important models in supervised machine learning.
Training in Gaussian processes refers to selecting the covariance functions and
the associated parameters in order to improve the outcome of predictions, the
core of which amounts to evaluating the logarithm of the marginal likelihood
(LML) of a given model. LML gives a concrete measure of the quality of
prediction that a GP model is expected to achieve. The classical computation of
LML typically carries a polynomial time overhead with respect to the input
size. We propose a quantum algorithm that computes the logarithm of the
determinant of a Hermitian matrix, which runs in logarithmic time for sparse
matrices. This is applied in conjunction with a variant of the quantum linear
system algorithm that allows for logarithmic time computation of the form
$\mathbf{y}^TA^{-1}\mathbf{y}$, where $\mathbf{y}$ is a dense vector and $A$ is
the covariance matrix. We hence show that quantum computing can be used to
estimate the LML of a GP with exponentially improved efficiency under certain
conditions.
|
Machine Speech Chain with One-shot Speaker Adaptation | In previous work, we developed a closed-loop speech chain model based on deep
learning, in which the architecture enabled the automatic speech recognition
(ASR) and text-to-speech synthesis (TTS) components to mutually improve their
performance. This was accomplished by the two parts teaching each other using
both labeled and unlabeled data. This approach could significantly improve
model performance within a single-speaker speech dataset, but only a slight
increase could be gained in multi-speaker tasks. Furthermore, the model is
still unable to handle unseen speakers. In this paper, we present a new speech
chain mechanism by integrating a speaker recognition model inside the loop. We
also propose extending the capability of TTS to handle unseen speakers by
implementing one-shot speaker adaptation. This enables TTS to mimic voice
characteristics from one speaker to another with only a one-shot speaker
sample, even from a text without any speaker information. In the speech chain
loop mechanism, ASR also benefits from the ability to further learn an
arbitrary speaker's characteristics from the generated speech waveform,
resulting in a significant improvement in the recognition rate.
|
Joint PLDA for Simultaneous Modeling of Two Factors | Probabilistic linear discriminant analysis (PLDA) is a method used for
biometric problems like speaker or face recognition that models the variability
of the samples using two latent variables, one that depends on the class of the
sample and another one that is assumed independent across samples and models
the within-class variability. In this work, we propose a generalization of PLDA
that enables joint modeling of two sample-dependent factors: the class of
interest and a nuisance condition. The approach does not change the basic form
of PLDA but rather modifies the training procedure to consider the dependency
across samples of the latent variable that models within-class variability.
While the identity of the nuisance condition is needed during training, it is
not needed during testing since we propose a scoring procedure that
marginalizes over the corresponding latent variable. We show results on a
multilingual speaker-verification task, where the language spoken is considered
a nuisance condition. We show that the proposed joint PLDA approach leads to
significant performance gains in this task for two different datasets, in
particular when the training data contains mostly or only monolingual speakers.
|
Normalization of Neural Networks using Analytic Variance Propagation | We address the problem of estimating statistics of hidden units in a neural
network using a method of analytic moment propagation. These statistics are
useful for approximate whitening of the inputs in front of saturating
non-linearities such as a sigmoid function. This is important for
initialization of training and for reducing the accumulated scale and bias
dependencies (compensating covariate shift), which presumably eases the
learning. In batch normalization, which is currently a very widely applied
technique, sample estimates of statistics of hidden units over a batch are
used. The proposed estimation uses an analytic propagation of mean and variance
of the training set through the network. The result depends on the network
structure and its current weights but not on the specific batch input. The
estimates are suitable for initialization and normalization, efficient to
compute and independent of the batch size. The experimental verification well
supports these claims. However, the method does not share the generalization
properties of BN, to which our experiments give some additional insight.
|
Stochastic Variational Inference with Gradient Linearization | Variational inference has experienced a recent surge in popularity owing to
stochastic approaches, which have yielded practical tools for a wide range of
model classes. A key benefit is that stochastic variational inference obviates
the tedious process of deriving analytical expressions for closed-form variable
updates. Instead, one simply needs to derive the gradient of the log-posterior,
which is often much easier. Yet for certain model classes, the log-posterior
itself is difficult to optimize using standard gradient techniques. One such
example are random field models, where optimization based on gradient
linearization has proven popular, since it speeds up convergence significantly
and can avoid poor local optima. In this paper we propose stochastic
variational inference with gradient linearization (SVIGL). It is similarly
convenient as standard stochastic variational inference - all that is required
is a local linearization of the energy gradient. Its benefit over stochastic
variational inference with conventional gradient methods is a clear improvement
in convergence speed, while yielding comparable or even better variational
approximations in terms of KL divergence. We demonstrate the benefits of SVIGL
in three applications: Optical flow estimation, Poisson-Gaussian denoising, and
3D surface reconstruction.
|
Feed-forward Uncertainty Propagation in Belief and Neural Networks | We propose a feed-forward inference method applicable to belief and neural
networks. In a belief network, the method estimates an approximate factorized
posterior of all hidden units given the input. In neural networks the method
propagates uncertainty of the input through all the layers. In neural networks
with injected noise, the method analytically takes into account uncertainties
resulting from this noise. Such feed-forward analytic propagation is
differentiable in parameters and can be trained end-to-end. Compared to
standard NN, which can be viewed as propagating only the means, we propagate
the mean and variance. The method can be useful in all scenarios that require
knowledge of the neuron statistics, e.g. when dealing with uncertain inputs,
considering sigmoid activations as probabilities of Bernoulli units, training
the models regularized by injected noise (dropout) or estimating activation
statistics over the dataset (as needed for normalization methods). In the
experiments we show the possible utility of the method in all these tasks as
well as its current limitations.
|
On Learning Graphs with Edge-Detecting Queries | We consider the problem of learning a general graph $G=(V,E)$ using
edge-detecting queries, where the number of vertices $|V|=n$ is given to the
learner. The information theoretic lower bound gives $m\log n$ for the number
of queries, where $m=|E|$ is the number of edges. In case the number of edges
$m$ is also given to the learner, Angluin-Chen's Las Vegas algorithm
\cite{AC08} runs in $4$ rounds and detects the edges in $O(m\log n)$ queries.
In the other harder case where the number of edges $m$ is unknown, their
algorithm runs in $5$ rounds and asks $O(m\log n+\sqrt{m}\log^2 n)$ queries.
There have been two open problems: \emph{(i)} can the number of queries be
reduced to $O(m\log n)$ in the second case, and, \emph{(ii)} can the number of
rounds be reduced without substantially increasing the number of queries (in
both cases). For the first open problem (when $m$ is unknown) we give two
algorithms. The first is an $O(1)$-round Las Vegas algorithm that asks $m\log
n+\sqrt{m}(\log^{[k]}n)\log n$ queries for any constant $k$ where
$\log^{[k]}n=\log \stackrel{k}{\cdots} \log n$. The second is an
$O(\log^*n)$-round Las Vegas algorithm that asks $O(m\log n)$ queries. This
solves the first open problem for any practical $n$, for example,
$n<2^{65536}$. We also show that no deterministic algorithm can solve this
problem in a constant number of rounds. To solve the second problem we study
the case when $m$ is known. We first show that any non-adaptive Monte Carlo
algorithm (one-round) must ask at least $\Omega(m^2\log n)$ queries, and any
two-round Las Vegas algorithm must ask at least $m^{4/3-o(1)}\log n$ queries on
average. We then give two two-round Monte Carlo algorithms, the first asks
$O(m^{4/3}\log n)$ queries for any $n$ and $m$, and the second asks $O(m\log
n)$ queries when $n>2^m$. Finally, we give a $3$-round Monte Carlo algorithm
that asks $O(m\log n)$ queries for any $n$ and $m$.
|
Active Metric Learning for Supervised Classification | Clustering and classification critically rely on distance metrics that
provide meaningful comparisons between data points. We present mixed-integer
optimization approaches to find optimal distance metrics that generalize the
Mahalanobis metric extensively studied in the literature. Additionally, we
generalize and improve upon leading methods by removing reliance on
pre-designated "target neighbors," "triplets," and "similarity pairs." Another
salient feature of our method is its ability to enable active learning by
recommending precise regions to sample after an optimal metric is computed to
improve classification performance. This targeted acquisition can significantly
reduce computational burden by ensuring training data completeness,
representativeness, and economy. We demonstrate classification and
computational performance of the algorithms through several simple and
intuitive examples, followed by results on real image and medical datasets.
|
Semi-supervised learning for structured regression on partially observed
attributed graphs | Conditional probabilistic graphical models provide a powerful framework for
structured regression in spatio-temporal datasets with complex correlation
patterns. However, in real-life applications a large fraction of observations
is often missing, which can severely limit the representational power of these
models. In this paper we propose a Marginalized Gaussian Conditional Random
Fields (m-GCRF) structured regression model for dealing with missing labels in
partially observed temporal attributed graphs. This method is aimed at learning
with both labeled and unlabeled parts and effectively predicting future values
in a graph. The method is even capable of learning from nodes for which the
response variable is never observed in history, which poses problems for many
state-of-the-art models that can handle missing data. The proposed model is
characterized for various missingness mechanisms on 500 synthetic graphs. The
benefits of the new method are also demonstrated on a challenging application
for predicting precipitation based on partial observations of climate variables
in a temporal graph that spans the entire continental US. We also show that the
method can be useful for optimizing the costs of data collection in climate
applications via active reduction of the number of weather stations to
consider. In experiments on these real-world and synthetic datasets we show
that the proposed model is consistently more accurate than alternative
semi-supervised structured models, as well as models that either use imputation
to deal with missing values or simply ignore them altogether.
|
Intertwiners between Induced Representations (with Applications to the
Theory of Equivariant Neural Networks) | Group equivariant and steerable convolutional neural networks (regular and
steerable G-CNNs) have recently emerged as a very effective model class for
learning from signal data such as 2D and 3D images, video, and other data where
symmetries are present. In geometrical terms, regular G-CNNs represent data in
terms of scalar fields ("feature channels"), whereas the steerable G-CNN can
also use vector or tensor fields ("capsules") to represent data. In algebraic
terms, the feature spaces in regular G-CNNs transform according to a regular
representation of the group G, whereas the feature spaces in Steerable G-CNNs
transform according to the more general induced representations of G. In order
to make the network equivariant, each layer in a G-CNN is required to
intertwine between the induced representations associated with its input and
output space.
In this paper we present a general mathematical framework for G-CNNs on
homogeneous spaces like Euclidean space or the sphere. We show, using
elementary methods, that the layers of an equivariant network are convolutional
if and only if the input and output feature spaces transform according to an
induced representation. This result, which follows from G.W. Mackey's abstract
theory on induced representations, establishes G-CNNs as a universal class of
equivariant network architectures, and generalizes the important recent work of
Kondor & Trivedi on the intertwiners between regular representations.
|
Pseudo-marginal Bayesian inference for supervised Gaussian process
latent variable models | We introduce a Bayesian framework for inference with a supervised version of
the Gaussian process latent variable model. The framework overcomes the high
correlations between latent variables and hyperparameters by using an unbiased
pseudo estimate for the marginal likelihood that approximately integrates over
the latent variables. This is used to construct a Markov Chain to explore the
posterior of the hyperparameters. We demonstrate the procedure on simulated and
real examples, showing its ability to capture uncertainty and multimodality of
the hyperparameters and improved uncertainty quantification in predictions when
compared with variational inference.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.