Unnamed: 0
int64
0
41k
title
stringlengths
4
274
category
stringlengths
5
18
summary
stringlengths
22
3.66k
theme
stringclasses
8 values
1,000
A Theory of Local Learning, the Learning Channel, and the Optimality of Backpropagation
cs.LG
In a physical neural system, where storage and processing are intimately intertwined, the rules for adjusting the synaptic weights can only depend on variables that are available locally, such as the activity of the pre- and post-synaptic neurons, resulting in local learning rules. A systematic framework for studying t...
computer science
1,001
Beating the Perils of Non-Convexity: Guaranteed Training of Neural Networks using Tensor Methods
cs.LG
Training neural networks is a challenging non-convex optimization problem, and backpropagation or gradient descent can get stuck in spurious local optima. We propose a novel algorithm based on tensor decomposition for guaranteed training of two-layer neural networks. We provide risk bounds for our proposed method, with...
computer science
1,002
Natural Neural Networks
stat.ML
We introduce Natural Neural Networks, a novel family of algorithms that speed up convergence by adapting their internal representation during training to improve conditioning of the Fisher matrix. In particular, we show a specific example that employs a simple and efficient reparametrization of the neural network weigh...
computer science
1,003
Semi-Supervised Learning with Ladder Networks
cs.NE
We combine supervised learning with unsupervised learning in deep neural networks. The proposed model is trained to simultaneously minimize the sum of supervised and unsupervised cost functions by backpropagation, avoiding the need for layer-wise pre-training. Our work builds on the Ladder network proposed by Valpola (...
computer science
1,004
OCReP: An Optimally Conditioned Regularization for Pseudoinversion Based Neural Training
cs.NE
In this paper we consider the training of single hidden layer neural networks by pseudoinversion, which, in spite of its popularity, is sometimes affected by numerical instability issues. Regularization is known to be effective in such cases, so that we introduce, in the framework of Tikhonov regularization, a matricia...
computer science
1,005
Partitioning Large Scale Deep Belief Networks Using Dropout
stat.ML
Deep learning methods have shown great promise in many practical applications, ranging from speech recognition, visual object recognition, to text processing. However, most of the current deep learning methods suffer from scalability problems for large-scale applications, forcing researchers or users to focus on small-...
computer science
1,006
Hessian-free Optimization for Learning Deep Multidimensional Recurrent Neural Networks
cs.LG
Multidimensional recurrent neural networks (MDRNNs) have shown a remarkable performance in the area of speech and handwriting recognition. The performance of an MDRNN is improved by further increasing its depth, and the difficulty of learning the deeper network is overcome by using Hessian-free (HF) optimization. Given...
computer science
1,007
On the Expressive Power of Deep Learning: A Tensor Analysis
cs.NE
It has long been conjectured that hypotheses spaces suitable for data that is compositional in nature, such as text or images, may be more efficiently represented with deep hierarchical networks than with shallow ones. Despite the vast empirical evidence supporting this belief, theoretical justifications to date are li...
computer science
1,008
A review of learning vector quantization classifiers
cs.LG
In this work we present a review of the state of the art of Learning Vector Quantization (LVQ) classifiers. A taxonomy is proposed which integrates the most relevant LVQ approaches to date. The main concepts associated with modern LVQ approaches are defined. A comparison is made among eleven LVQ classifiers using one r...
computer science
1,009
Provable approximation properties for deep neural networks
stat.ML
We discuss approximation of functions using deep neural nets. Given a function $f$ on a $d$-dimensional manifold $\Gamma \subset \mathbb{R}^m$, we construct a sparsely-connected depth-4 neural network and bound its error in approximating $f$. The size of the network depends on dimension and curvature of the manifold $\...
computer science
1,010
Semantics, Representations and Grammars for Deep Learning
cs.LG
Deep learning is currently the subject of intensive study. However, fundamental concepts such as representations are not formally defined -- researchers "know them when they see them" -- and there is no common language for describing and analyzing algorithms. This essay proposes an abstract framework that identifies th...
computer science
1,011
Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree
stat.ML
We seek to improve deep neural networks by generalizing the pooling operations that play a central role in current architectures. We pursue a careful exploration of approaches to allow pooling to learn and to adapt to complex and variable patterns. The two primary directions lie in (1) learning a pooling function via (...
computer science
1,012
Batch Normalized Recurrent Neural Networks
stat.ML
Recurrent Neural Networks (RNNs) are powerful models for sequential data that have the potential to learn long-term dependencies. However, they are computationally expensive to train and difficult to parallelize. Recent work has shown that normalizing intermediate representations of neural networks can significantly im...
computer science
1,013
Optimizing and Contrasting Recurrent Neural Network Architectures
stat.ML
Recurrent Neural Networks (RNNs) have long been recognized for their potential to model complex time series. However, it remains to be determined what optimization techniques and recurrent architectures can be used to best realize this potential. The experiments presented take a deep look into Hessian free optimization...
computer science
1,014
An Exploration of Softmax Alternatives Belonging to the Spherical Loss Family
cs.NE
In a multi-class classification problem, it is standard to model the output of a neural network as a categorical distribution conditioned on the inputs. The output must therefore be positive and sum to one, which is traditionally enforced by a softmax. This probabilistic mapping allows to use the maximum likelihood pri...
computer science
1,015
Understanding Adversarial Training: Increasing Local Stability of Neural Nets through Robust Optimization
stat.ML
We propose a general framework for increasing local stability of Artificial Neural Nets (ANNs) using Robust Optimization (RO). We achieve this through an alternating minimization-maximization procedure, in which the loss of the network is minimized over perturbed examples that are generated at each parameter update. We...
computer science
1,016
Unitary Evolution Recurrent Neural Networks
cs.LG
Recurrent neural networks (RNNs) are notoriously difficult to train. When the eigenvalues of the hidden to hidden weight matrix deviate from absolute value 1, optimization becomes difficult due to the well studied issue of vanishing and exploding gradients, especially when trying to learn long-term dependencies. To cir...
computer science
1,017
The Limitations of Deep Learning in Adversarial Settings
cs.CR
Deep learning takes advantage of large datasets and computationally efficient training algorithms to outperform other approaches at various machine learning tasks. However, imperfections in the training phase of deep neural networks make them vulnerable to adversarial samples: inputs crafted by adversaries with the int...
computer science
1,018
The Power of Depth for Feedforward Neural Networks
cs.LG
We show that there is a simple (approximately radial) function on $\reals^d$, expressible by a small 3-layer feedforward neural networks, which cannot be approximated by any 2-layer network, to more than a certain constant accuracy, unless its width is exponential in the dimension. The result holds for virtually all kn...
computer science
1,019
Structured Pruning of Deep Convolutional Neural Networks
cs.NE
Real time application of deep learning algorithms is often hindered by high computational complexity and frequent memory accesses. Network pruning is a promising technique to solve this problem. However, pruning usually results in irregular network connections that not only demand extra representation efforts but also ...
computer science
1,020
Common Variable Learning and Invariant Representation Learning using Siamese Neural Networks
stat.ML
We consider the statistical problem of learning common source of variability in data which are synchronously captured by multiple sensors, and demonstrate that Siamese neural networks can be naturally applied to this problem. This approach is useful in particular in exploratory, data-driven applications, where neither ...
computer science
1,021
Stochastic Neural Networks with Monotonic Activation Functions
stat.ML
We propose a Laplace approximation that creates a stochastic unit from any smooth monotonic activation function, using only Gaussian noise. This paper investigates the application of this stochastic approximation in training a family of Restricted Boltzmann Machines (RBM) that are closely linked to Bregman divergences....
computer science
1,022
Single-Solution Hypervolume Maximization and its use for Improving Generalization of Neural Networks
cs.LG
This paper introduces the hypervolume maximization with a single solution as an alternative to the mean loss minimization. The relationship between the two problems is proved through bounds on the cost function when an optimal solution to one of the problems is evaluated on the other, with a hyperparameter to control t...
computer science
1,023
Poor starting points in machine learning
cs.LG
Poor (even random) starting points for learning/training/optimization are common in machine learning. In many settings, the method of Robbins and Monro (online stochastic gradient descent) is known to be optimal for good starting points, but may not be optimal for poor starting points -- indeed, for poor starting point...
computer science
1,024
Benefits of depth in neural networks
cs.LG
For any positive integer $k$, there exist neural networks with $\Theta(k^3)$ layers, $\Theta(1)$ nodes per layer, and $\Theta(1)$ distinct parameters which can not be approximated by networks with $\mathcal{O}(k)$ layers unless they are exponentially large --- they must possess $\Omega(2^k)$ nodes. This result is prove...
computer science
1,025
Streaming PCA: Matching Matrix Bernstein and Near-Optimal Finite Sample Guarantees for Oja's Algorithm
cs.LG
This work provides improved guarantees for streaming principle component analysis (PCA). Given $A_1, \ldots, A_n\in \mathbb{R}^{d\times d}$ sampled independently from distributions satisfying $\mathbb{E}[A_i] = \Sigma$ for $\Sigma \succeq \mathbf{0}$, this work provides an $O(d)$-space linear-time single-pass streaming...
computer science
1,026
Practical Riemannian Neural Networks
cs.NE
We provide the first experimental results on non-synthetic datasets for the quasi-diagonal Riemannian gradient descents for neural networks introduced in [Ollivier, 2015]. These include the MNIST, SVHN, and FACE datasets as well as a previously unpublished electroencephalogram dataset. The quasi-diagonal Riemannian alg...
computer science
1,027
Scalable and Sustainable Deep Learning via Randomized Hashing
stat.ML
Current deep learning architectures are growing larger in order to learn from complex datasets. These architectures require giant matrix multiplication operations to train millions of parameters. Conversely, there is another growing trend to bring deep learning to low-power, embedded devices. The matrix operations, ass...
computer science
1,028
Noisy Activation Functions
cs.LG
Common nonlinear activation functions used in neural networks can cause training difficulties due to the saturation behavior of the activation function, which may hide dependencies that are not visible to vanilla-SGD (using first order gradients only). Gating mechanisms that use softly saturating activation functions t...
computer science
1,029
Training Input-Output Recurrent Neural Networks through Spectral Methods
cs.LG
We consider the problem of training input-output recurrent neural networks (RNN) for sequence labeling tasks. We propose a novel spectral approach for learning the network parameters. It is based on decomposition of the cross-moment tensor between the output and a non-linear transformation of the input, based on score ...
computer science
1,030
Turing learning: a metric-free approach to inferring behavior and its application to swarms
stat.ML
We propose Turing Learning, a novel system identification method for inferring the behavior of natural or artificial systems. Turing Learning simultaneously optimizes two populations of computer programs, one representing models of the behavior of the system under investigation, and the other representing classifiers. ...
computer science
1,031
Stochastic Variance Reduction for Nonconvex Optimization
math.OC
We study nonconvex finite-sum problems and analyze stochastic variance reduced gradient (SVRG) methods for them. SVRG and related methods have recently surged into prominence for convex optimization given their edge over stochastic gradient descent (SGD); but their theoretical analysis almost exclusively assumes convex...
computer science
1,032
Variational Autoencoders for Feature Detection of Magnetic Resonance Imaging Data
cs.LG
Independent component analysis (ICA), as an approach to the blind source-separation (BSS) problem, has become the de-facto standard in many medical imaging settings. Despite successes and a large ongoing research effort, the limitation of ICA to square linear transformations have not been overcome, so that general INFO...
computer science
1,033
A guide to convolution arithmetic for deep learning
stat.ML
We introduce a guide to help deep learning practitioners understand and manipulate convolutional neural network architectures. The guide clarifies the relationship between various properties (input shape, kernel shape, zero padding, strides and output shape) of convolutional, pooling and transposed convolutional layers...
computer science
1,034
Acceleration of Deep Neural Network Training with Resistive Cross-Point Devices
cs.LG
In recent years, deep neural networks (DNN) have demonstrated significant business impact in large scale analysis and classification tasks such as speech recognition, visual object detection, pattern extraction, etc. Training of large DNNs, however, is universally considered as time consuming and computationally intens...
computer science
1,035
Biologically Inspired Radio Signal Feature Extraction with Sparse Denoising Autoencoders
stat.ML
Automatic modulation classification (AMC) is an important task for modern communication systems; however, it is a challenging problem when signal features and precise models for generating each modulation may be unknown. We present a new biologically-inspired AMC method without the need for models or manually specified...
computer science
1,036
Learning activation functions from data using cubic spline interpolation
stat.ML
Neural networks require a careful design in order to perform properly on a given task. In particular, selecting a good activation function (possibly in a data-dependent fashion) is a crucial step, which remains an open problem in the research community. Despite a large amount of investigations, most current implementat...
computer science
1,037
On Optimality Conditions for Auto-Encoder Signal Recovery
stat.ML
Auto-Encoders are unsupervised models that aim to learn patterns from observed data by minimizing a reconstruction cost. The useful representations learned are often found to be sparse and distributed. On the other hand, compressed sensing and sparse coding assume a data generating process, where the observed data is g...
computer science
1,038
Recurrent Neural Networks for Multivariate Time Series with Missing Values
cs.LG
Multivariate time series data in practical applications, such as health care, geoscience, and biology, are characterized by a variety of missing values. In time series prediction and other related tasks, it has been noted that missing values and their missing patterns are often correlated with the target labels, a.k.a....
computer science
1,039
Modeling Missing Data in Clinical Time Series with RNNs
cs.LG
We demonstrate a simple strategy to cope with missing data in sequential inputs, addressing the task of multilabel classification of diagnoses given clinical time series. Collected from the pediatric intensive care unit (PICU) at Children's Hospital Los Angeles, our data consists of multivariate time series of observat...
computer science
1,040
Improving Power Generation Efficiency using Deep Neural Networks
stat.ML
Recently there has been significant research on power generation, distribution and transmission efficiency especially in the case of renewable resources. The main objective is reduction of energy losses and this requires improvements on data acquisition and analysis. In this paper we address these concerns by using con...
computer science
1,041
Faster Training of Very Deep Networks Via p-Norm Gates
stat.ML
A major contributing factor to the recent advances in deep neural networks is structural units that let sensory information and gradients to propagate easily. Gating is one such structure that acts as a flow control. Gates are employed in many recent state-of-the-art recurrent models such as LSTM and GRU, and feedforwa...
computer science
1,042
BBQ-Networks: Efficient Exploration in Deep Reinforcement Learning for Task-Oriented Dialogue Systems
cs.LG
We present a new algorithm that significantly improves the efficiency of exploration for deep Q-learning agents in dialogue systems. Our agents explore via Thompson sampling, drawing Monte Carlo samples from a Bayes-by-Backprop neural network. Our algorithm learns much faster than common exploration strategies such as ...
computer science
1,043
Why does deep and cheap learning work so well?
cs.LG
We show how the success of deep learning could depend not only on mathematics but also on physics: although well-known mathematical theorems guarantee that neural networks can approximate arbitrary functions well, the class of functions of practical interest can frequently be approximated through "cheap learning" with ...
computer science
1,044
Distribution-Specific Hardness of Learning Neural Networks
cs.LG
Although neural networks are routinely and successfully trained in practice using simple gradient-based methods, most existing theoretical results are negative, showing that learning such networks is difficult, in a worst-case sense over all data distributions. In this paper, we take a more nuanced view, and consider w...
computer science
1,045
Sampling Generative Networks
cs.NE
We introduce several techniques for sampling and visualizing the latent spaces of generative models. Replacing linear interpolation with spherical linear interpolation prevents diverging from a model's prior distribution and produces sharper samples. J-Diagrams and MINE grids are introduced as visualizations of manifol...
computer science
1,046
A Cheap Linear Attention Mechanism with Fast Lookups and Fixed-Size Representations
cs.LG
The softmax content-based attention mechanism has proven to be very beneficial in many applications of recurrent neural networks. Nevertheless it suffers from two major computational limitations. First, its computations for an attention lookup scale linearly in the size of the attended sequence. Second, it does not enc...
computer science
1,047
Regularized Dynamic Boltzmann Machine with Delay Pruning for Unsupervised Learning of Temporal Sequences
cs.LG
We introduce Delay Pruning, a simple yet powerful technique to regularize dynamic Boltzmann machines (DyBM). The recently introduced DyBM provides a particularly structured Boltzmann machine, as a generative model of a multi-dimensional time-series. This Boltzmann machine can have infinitely many layers of units but al...
computer science
1,048
Tractable Generative Convolutional Arithmetic Circuits
cs.LG
Casting neural networks in generative frameworks is a highly sought-after endeavor these days. Existing methods, such as Generative Adversarial Networks, capture some of the generative capabilities, but not all. To truly leverage the power of generative models, tractable marginalization is needed, a feature outside the...
computer science
1,049
Using Fast Weights to Attend to the Recent Past
stat.ML
Until recently, research on artificial neural networks was largely restricted to systems with only two types of variable: Neural activities that represent the current or recent input and weights that learn to capture regularities among inputs, outputs and payoffs. There is no good reason for this restriction. Synapses ...
computer science
1,050
Hybrid clustering-classification neural network in the medical diagnostics of reactive arthritis
cs.LG
The hybrid clustering-classification neural network is proposed. This network allows increasing a quality of information processing under the condition of overlapping classes due to the rational choice of a learning rate parameter and introducing a special procedure of fuzzy reasoning in the clustering process, which o...
computer science
1,051
Full-Capacity Unitary Recurrent Neural Networks
stat.ML
Recurrent neural networks are powerful models for processing sequential data, but they are generally plagued by vanishing and exploding gradient problems. Unitary recurrent neural networks (uRNNs), which use unitary recurrence matrices, have recently been proposed as a means to avoid these issues. However, in previous ...
computer science
1,052
Demystifying ResNet
cs.NE
The Residual Network (ResNet), proposed in He et al. (2015), utilized shortcut connections to significantly reduce the difficulty of training, which resulted in great performance boosts in terms of both training and generalization error. It was empirically observed in He et al. (2015) that stacking more layers of res...
computer science
1,053
Combating Reinforcement Learning's Sisyphean Curse with Intrinsic Fear
cs.LG
Many practical environments contain catastrophic states that an optimal agent would visit infrequently or never. Even on toy problems, Deep Reinforcement Learning (DRL) agents tend to periodically revisit these states upon forgetting their existence under a new policy. We introduce intrinsic fear (IF), a learned reward...
computer science
1,054
Adversarial Ladder Networks
cs.NE
The use of unsupervised data in addition to supervised data in training discriminative neural networks has improved the performance of this clas- sification scheme. However, the best results were achieved with a training process that is divided in two parts: first an unsupervised pre-training step is done for initializ...
computer science
1,055
Neural Taylor Approximations: Convergence and Exploration in Rectifier Networks
cs.LG
Modern convolutional networks, incorporating rectifiers and max-pooling, are neither smooth nor convex. Standard guarantees therefore do not apply. Nevertheless, methods from convex optimization such as gradient descent and Adam are widely used as building blocks for deep learning algorithms. This paper provides the fi...
computer science
1,056
Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders
cs.LG
We study a variant of the variational autoencoder model (VAE) with a Gaussian mixture as a prior distribution, with the goal of performing unsupervised clustering through deep generative models. We observe that the known problem of over-regularisation that has been shown to arise in regular VAEs also manifests itself i...
computer science
1,057
Identity Matters in Deep Learning
cs.LG
An emerging design principle in deep learning is that each layer of a deep artificial neural network should be able to easily express the identity transformation. This idea not only motivated various normalization techniques, such as \emph{batch normalization}, but was also key to the immense success of \emph{residual ...
computer science
1,058
Deep Learning with Sets and Point Clouds
stat.ML
We introduce a simple permutation equivariant layer for deep learning with set structure.This type of layer, obtained by parameter-sharing, has a simple implementation and linear-time complexity in the size of each set. We use deep permutation-invariant networks to perform point-could classification and MNIST-digit sum...
computer science
1,059
Compacting Neural Network Classifiers via Dropout Training
stat.ML
We introduce dropout compaction, a novel method for training feed-forward neural networks which realizes the performance gains of training a large model with dropout regularization, yet extracts a compact neural network for run-time efficiency. In the proposed method, we introduce a sparsity-inducing prior on the per u...
computer science
1,060
Spikes as regularizers
cs.NE
We present a confidence-based single-layer feed-forward learning algorithm SPIRAL (Spike Regularized Adaptive Learning) relying on an encoding of activation spikes. We adaptively update a weight vector relying on confidence estimates and activation offsets relative to previous activity. We regularize updates proportion...
computer science
1,061
Local minima in training of neural networks
stat.ML
There has been a lot of recent interest in trying to characterize the error surface of deep models. This stems from a long standing question. Given that deep networks are highly nonlinear systems optimized by local gradient methods, why do they not seem to be affected by bad local minima? It is widely believed that tra...
computer science
1,062
Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline
cs.LG
We propose a simple but strong baseline for time series classification from scratch with deep neural networks. Our proposed baseline models are pure end-to-end without any heavy preprocessing on the raw data or feature crafting. The proposed Fully Convolutional Network (FCN) achieves premium performance to other state-...
computer science
1,063
Tunable Sensitivity to Large Errors in Neural Network Training
stat.ML
When humans learn a new concept, they might ignore examples that they cannot make sense of at first, and only later focus on such examples, when they are more useful for learning. We propose incorporating this idea of tunable sensitivity for hard examples in neural network learning, using a new generalization of the cr...
computer science
1,064
Efficient Convolutional Auto-Encoding via Random Convexification and Frequency-Domain Minimization
stat.ML
The omnipresence of deep learning architectures such as deep convolutional neural networks (CNN)s is fueled by the synergistic combination of ever-increasing labeled datasets and specialized hardware. Despite the indisputable success, the reliance on huge amounts of labeled data and specialized hardware can be a limiti...
computer science
1,065
Memory Augmented Neural Networks with Wormhole Connections
cs.LG
Recent empirical results on long-term dependency tasks have shown that neural networks augmented with an external memory can learn the long-term dependency tasks more easily and achieve better generalization than vanilla recurrent neural networks (RNN). We suggest that memory augmented neural networks can reduce the ef...
computer science
1,066
Toward the automated analysis of complex diseases in genome-wide association studies using genetic programming
cs.NE
Machine learning has been gaining traction in recent years to meet the demand for tools that can efficiently analyze and make sense of the ever-growing databases of biomedical data in health care systems around the world. However, effectively using machine learning methods requires considerable domain expertise, which ...
computer science
1,067
Deep Learning with Dynamic Computation Graphs
cs.NE
Neural networks that compute over graph structures are a natural fit for problems in a variety of domains, including natural language (parse trees) and cheminformatics (molecular graphs). However, since the computation graph has a different shape and size for every input, such networks do not directly support batched t...
computer science
1,068
Deep Kernelized Autoencoders
stat.ML
In this paper we introduce the deep kernelized autoencoder, a neural network model that allows an explicit approximation of (i) the mapping from an input space to an arbitrary, user-specified kernel space and (ii) the back-projection from such a kernel space to input space. The proposed method is based on traditional a...
computer science
1,069
A Deterministic and Generalized Framework for Unsupervised Learning with Restricted Boltzmann Machines
cs.LG
Restricted Boltzmann machines (RBMs) are energy-based neural-networks which are commonly used as the building blocks for deep architectures neural architectures. In this work, we derive a deterministic framework for the training, evaluation, and use of RBMs based upon the Thouless-Anderson-Palmer (TAP) mean-field appro...
computer science
1,070
Robust Stochastic Configuration Networks with Kernel Density Estimation
cs.NE
Neural networks have been widely used as predictive models to fit data distribution, and they could be implemented through learning a collection of samples. In many applications, however, the given dataset may contain noisy samples or outliers which may result in a poor learner model in terms of generalization. This pa...
computer science
1,071
Generative Temporal Models with Memory
cs.LG
We consider the general problem of modeling temporal data with long-range dependencies, wherein new observations are fully or partially predictable based on temporally-distant, past observations. A sufficiently powerful temporal model should separate predictable elements of the sequence from unpredictable elements, exp...
computer science
1,072
On the Origin of Deep Learning
cs.LG
This paper is a review of the evolutionary history of deep learning models. It covers from the genesis of neural networks when associationism modeling of the brain is studied, to the models that dominate the last decade of research in deep learning like convolutional neural networks, deep belief networks, and recurrent...
computer science
1,073
The Shattered Gradients Problem: If resnets are the answer, then what is the question?
cs.NE
A long-standing obstacle to progress in deep learning is the problem of vanishing and exploding gradients. The problem has largely been overcome through the introduction of carefully constructed initializations and batch normalization. Nevertheless, architectures incorporating skip-connections such as resnets perform m...
computer science
1,074
Generalization and Equilibrium in Generative Adversarial Nets (GANs)
cs.LG
We show that training of generative adversarial network (GAN) may not have good generalization properties; e.g., training may appear successful but the trained distribution may be far from target distribution in standard metrics. However, generalization does occur for a weaker metric called neural net distance. It is a...
computer science
1,075
Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
cs.NE
The recently proposed Temporal Ensembling has achieved state-of-the-art results in several semi-supervised learning benchmarks. It maintains an exponential moving average of label predictions on each training example, and penalizes predictions that are inconsistent with this target. However, because the targets change ...
computer science
1,076
On the Expressive Power of Overlapping Architectures of Deep Learning
cs.LG
Expressive efficiency refers to the relation between two architectures A and B, whereby any function realized by B could be replicated by A, but there exists functions realized by A, which cannot be replicated by B unless its size grows significantly larger. For example, it is known that deep networks are exponentially...
computer science
1,077
Learned Optimizers that Scale and Generalize
cs.LG
Learning to learn has emerged as an important direction for achieving artificial intelligence. Two of the primary barriers to its adoption are an inability to scale to larger problems and a limited ability to generalize to new tasks. We introduce a learned gradient descent optimizer that generalizes well to new tasks, ...
computer science
1,078
An Automated Auto-encoder Correlation-based Health-Monitoring and Prognostic Method for Machine Bearings
cs.LG
This paper studies an intelligent ultimate technique for health-monitoring and prognostic of common rotary machine components, particularly bearings. During a run-to-failure experiment, rich unsupervised features from vibration sensory data are extracted by a trained sparse auto-encoder. Then, the correlation of the ex...
computer science
1,079
Ensemble representation learning: an analysis of fitness and survival for wrapper-based genetic programming methods
cs.NE
Recently we proposed a general, ensemble-based feature engineering wrapper (FEW) that was paired with a number of machine learning methods to solve regression problems. Here, we adapt FEW for supervised classification and perform a thorough analysis of fitness and survival methods within this framework. Our tests demon...
computer science
1,080
Failures of Gradient-Based Deep Learning
cs.LG
In recent years, Deep Learning has become the go-to solution for a broad range of applications, often outperforming state-of-the-art. However, it is important, for both theoreticians and practitioners, to gain a deeper understanding of the difficulties and limitations associated with common approaches and algorithms. W...
computer science
1,081
A Neural Representation of Sketch Drawings
cs.NE
We present sketch-rnn, a recurrent neural network (RNN) able to construct stroke-based drawings of common objects. The model is trained on thousands of crude human-drawn images representing hundreds of classes. We outline a framework for conditional and unconditional sketch generation, and describe new robust training ...
computer science
1,082
Diagonal RNNs in Symbolic Music Modeling
cs.NE
In this paper, we propose a new Recurrent Neural Network (RNN) architecture. The novelty is simple: We use diagonal recurrent matrices instead of full. This results in better test likelihood and faster convergence compared to regular full RNNs in most of our experiments. We show the benefits of using diagonal recurrent...
computer science
1,083
DropIn: Making Reservoir Computing Neural Networks Robust to Missing Inputs by Dropout
cs.LG
The paper presents a novel, principled approach to train recurrent neural networks from the Reservoir Computing family that are robust to missing part of the input features at prediction time. By building on the ensembling properties of Dropout regularization, we propose a methodology, named DropIn, which efficiently t...
computer science
1,084
Training Deep Convolutional Neural Networks with Resistive Cross-Point Devices
cs.LG
In a previous work we have detailed the requirements to obtain a maximal performance benefit by implementing fully connected deep neural networks (DNN) in form of arrays of resistive devices for deep learning. This concept of Resistive Processing Unit (RPU) devices we extend here towards convolutional neural networks (...
computer science
1,085
SuperSpike: Supervised learning in multi-layer spiking neural networks
cs.LG
A vast majority of computation in the brain is performed by spiking neural networks. Despite the ubiquity of such spiking, we currently lack an understanding of how biological spiking neural circuits learn and compute in-vivo, as well as how we can instantiate such capabilities in artificial spiking circuits in-silico....
computer science
1,086
Are Saddles Good Enough for Deep Learning?
stat.ML
Recent years have seen a growing interest in understanding deep neural networks from an optimization perspective. It is understood now that converging to low-cost local minima is sufficient for such models to become effective in practice. However, in this work, we propose a new hypothesis based on recent theoretical fi...
computer science
1,087
Gated Orthogonal Recurrent Units: On Learning to Forget
cs.LG
We present a novel recurrent neural network (RNN) based model that combines the remembering ability of unitary RNNs with the ability of gated RNNs to effectively forget redundant/irrelevant information in its memory. We achieve this by extending unitary RNNs with a gating mechanism. Our model is able to outperform LSTM...
computer science
1,088
Neural networks and rational functions
cs.LG
Neural networks and rational functions efficiently approximate each other. In more detail, it is shown here that for any ReLU network, there exists a rational function of degree $O(\text{polylog}(1/\epsilon))$ which is $\epsilon$-close, and similarly for any rational function there exists a ReLU network of size $O(\tex...
computer science
1,089
Fatiguing STDP: Learning from Spike-Timing Codes in the Presence of Rate Codes
cs.NE
Spiking neural networks (SNNs) could play a key role in unsupervised machine learning applications, by virtue of strengths related to learning from the fine temporal structure of event-based signals. However, some spike-timing-related strengths of SNNs are hindered by the sensitivity of spike-timing-dependent plasticit...
computer science
1,090
Sparse Neural Networks Topologies
cs.LG
We propose Sparse Neural Network architectures that are based on random or structured bipartite graph topologies. Sparse architectures provide compression of the models learned and speed-ups of computations, they can also surpass their unstructured or fully connected counterparts. As we show, even more compact topologi...
computer science
1,091
Towards Deep Learning Models Resistant to Adversarial Attacks
stat.ML
Recent work has demonstrated that neural networks are vulnerable to adversarial examples, i.e., inputs that are almost indistinguishable from natural data and yet classified incorrectly by the network. In fact, some of the latest findings suggest that the existence of adversarial attacks may be an inherent weakness of ...
computer science
1,092
Spectrally-normalized margin bounds for neural networks
cs.LG
This paper presents a margin-based multiclass generalization bound for neural networks that scales with their margin-normalized "spectral complexity": their Lipschitz constant, meaning the product of the spectral norms of the weight matrices, times a certain correction factor. This bound is empirically investigated for...
computer science
1,093
Model compression as constrained optimization, with application to neural nets. Part I: general framework
cs.LG
Compressing neural nets is an active research problem, given the large size of state-of-the-art nets for tasks such as object recognition, and the computational limits imposed by mobile devices. We give a general formulation of model compression as constrained optimization. This includes many types of compression: quan...
computer science
1,094
Model compression as constrained optimization, with application to neural nets. Part II: quantization
cs.LG
We consider the problem of deep neural net compression by quantization: given a large, reference net, we want to quantize its real-valued weights using a codebook with $K$ entries so that the training loss of the quantized net is minimal. The codebook can be optimally learned jointly with the net, or fixed, as for bina...
computer science
1,095
Tikhonov Regularization for Long Short-Term Memory Networks
cs.LG
It is a well-known fact that adding noise to the input data often improves network performance. While the dropout technique may be a cause of memory loss, when it is applied to recurrent connections, Tikhonov regularization, which can be regarded as the training with additive noise, avoids this issue naturally, though ...
computer science
1,096
Neural Expectation Maximization
cs.LG
Many real world tasks such as reasoning and physical interaction require identification and manipulation of conceptual entities. A first step towards solving these tasks is the automated discovery of distributed symbol-like representations. In this paper, we explicitly formalize this problem as inference in a spatial m...
computer science
1,097
Classification via Tensor Decompositions of Echo State Networks
cs.LG
This work introduces a tensor-based method to perform supervised classification on spatiotemporal data processed in an echo state network. Typically when performing supervised classification tasks on data processed in an echo state network, the entire collection of hidden layer node states from the training dataset is ...
computer science
1,098
Deep learning with convolutional neural networks for decoding and visualization of EEG pathology
cs.LG
We apply convolutional neural networks (ConvNets) to the task of distinguishing pathological from normal EEG recordings in the Temple University Hospital EEG Abnormal Corpus. We use two basic, shallow and deep ConvNet architectures recently shown to decode task-related information from EEG at least as well as establish...
computer science
1,099
A New Learning Paradigm for Random Vector Functional-Link Network: RVFL+
stat.ML
In school, a teacher plays an important role in various classroom teaching patterns. Likewise to this human learning activity, the learning using privileged information (LUPI) paradigm provides additional information generated by the teacher to 'teach' learning algorithms during the training stage. Therefore, this nove...
computer science