Searcharxiv⌕ Search

arXiv subjects

Yoshua Bengio

Publications and source records attributed to Yoshua Bengio.

At least 523 records · Page 29Linked to original sources

Batch-normalized joint training for DNN-based distant speech recognition

Improving distant speech recognition is a crucial step towards flexible human-machine interfaces. Current technology, however, still exhibits a lack of robustness, especially when adverse acoustic conditions are met. Despite the significant progress made in the last years on both speech enhancement and speech recognition, one potential limitation of state-of-the-art technology lies in composing modules that are not well matched because they are not trained jointly. To address this concern, a promising approach consists in concatenating a speech enhancement and a speech recognition deep neural network and to jointly update their parameters as if they were within a single bigger network. Unfortunately, joint training can be difficult because the output distribution of the speech enhancement system may change substantially during the optimization procedure. The speech recognition module would have to deal with an input distribution that is non-stationary and unnormalized. To mitigate this issue, we propose a joint training approach based on a fully batch-normalized architecture. Experiments, conducted using different datasets, tasks and acoustic conditions, revealed that the proposed framework significantly overtakes other competitive solutions, especially in challenging environments.

cs.CL↗

A network of deep neural networks for distant speech recognition

Despite the remarkable progress recently made in distant speech recognition, state-of-the-art technology still suffers from a lack of robustness, especially when adverse acoustic conditions characterized by non-stationary noises and reverberation are met. A prominent limitation of current systems lies in the lack of matching and communication between the various technologies involved in the distant speech recognition process. The speech enhancement and speech recognition modules are, for instance, often trained independently. Moreover, the speech enhancement normally helps the speech recognizer, but the output of the latter is not commonly used, in turn, to improve the speech enhancement. To address both concerns, we propose a novel architecture based on a network of deep neural networks, where all the components are jointly trained and better cooperate with each other thanks to a full communication scheme between them. Experiments, conducted using different datasets, tasks and acoustic conditions, revealed that the proposed framework can overtake other competitive solutions, including recent joint training approaches.

cs.CL↗

Independently Controllable Features

Finding features that disentangle the different causes of variation in real data is a difficult task, that has nonetheless received considerable attention in static domains like natural images. Interactive environments, in which an agent can deliberately take actions, offer an opportunity to tackle this task better, because the agent can experiment with different actions and observe their effects. We introduce the idea that in interactive environments, latent factors that control the variation in observed data can be identified by figuring out what the agent can control. We propose a naive method to find factors that explain or measure the effect of the actions of a learner, and test it in illustrative experiments.

cs.LG↗

Dynamic Neural Turing Machine with Soft and Hard Addressing Schemes

We extend neural Turing machine (NTM) model into a dynamic neural Turing machine (D-NTM) by introducing a trainable memory addressing scheme. This addressing scheme maintains for each memory cell two separate vectors, content and address vectors. This allows the D-NTM to learn a wide variety of location-based addressing strategies including both linear and nonlinear ones. We implement the D-NTM with both continuous, differentiable and discrete, non-differentiable read/write mechanisms. We investigate the mechanisms and effects of learning to read and write into a memory through experiments on Facebook bAbI tasks using both a feedforward and GRUcontroller. The D-NTM is evaluated on a set of Facebook bAbI tasks and shown to outperform NTM and LSTM baselines. We have done extensive analysis of our model and different variations of NTM on bAbI task. We also provide further experimental results on sequential pMNIST, Stanford Natural Language Inference, associative recall and copy tasks.

cs.LG↗

Diet Networks: Thin Parameters for Fat Genomics

Learning tasks such as those involving genomic data often poses a serious challenge: the number of input features can be orders of magnitude larger than the number of training examples, making it difficult to avoid overfitting, even when using the known regularization techniques. We focus here on tasks in which the input is a description of the genetic variation specific to a patient, the single nucleotide polymorphisms (SNPs), yielding millions of ternary inputs. Improving the ability of deep learning to handle such datasets could have an important impact in precision medicine, where high-dimensional data regarding a particular patient is used to make predictions of interest. Even though the amount of data for such tasks is increasing, this mismatch between the number of examples and the number of inputs remains a concern. Naive implementations of classifier neural networks involve a huge number of free parameters in their first layer: each input feature is associated with as many parameters as there are hidden units. We propose a novel neural network parametrization which considerably reduces the number of free parameters. It is based on the idea that we can first learn or provide a distributed representation for each input feature (e.g. for each position in the genome where variations are observed), and then learn (with another neural network called the parameter prediction network) how to map a feature's distributed representation to the vector of parameters specific to that feature in the classifier neural network (the weights which link the value of the feature to each of the hidden units). We show experimentally on a population stratification task of interest to medical studies that the proposed approach can significantly reduce both the number of parameters and the error rate of the classifier.

cs.LG↗

Hierarchical Multiscale Recurrent Neural Networks

Learning both hierarchical and temporal representation has been among the long-standing challenges of recurrent neural networks. Multiscale recurrent neural networks have been considered as a promising approach to resolve this issue, yet there has been a lack of empirical evidence showing that this type of models can actually capture the temporal dependencies by discovering the latent hierarchical structure of the sequence. In this paper, we propose a novel multiscale approach, called the hierarchical multiscale recurrent neural networks, which can capture the latent hierarchical structure in the sequence by encoding the temporal dependencies with different timescales using a novel update mechanism. We show some evidence that our proposed multiscale architecture can discover underlying hierarchical structure in the sequences without using explicit boundary information. We evaluate our proposed model on character-level language modelling and handwriting sequence modelling.

cs.LG↗

A Structured Self-attentive Sentence Embedding

This paper proposes a new model for extracting an interpretable sentence embedding by introducing self-attention. Instead of using a vector, we use a 2-D matrix to represent the embedding, with each row of the matrix attending on a different part of the sentence. We also propose a self-attention mechanism and a special regularization term for the model. As a side effect, the embedding comes with an easy way of visualizing what specific parts of the sentence are encoded into the embedding. We evaluate our model on 3 different tasks: author profiling, sentiment classification, and textual entailment. Results show that our model yields a significant performance gain compared to other sentence embedding methods in all of the 3 tasks.

cs.CL↗

An Actor-Critic Algorithm for Sequence Prediction

We present an approach to training neural networks to generate sequences using actor-critic methods from reinforcement learning (RL). Current log-likelihood training methods are limited by the discrepancy between their training and testing modes, as models must generate tokens conditioned on their previous guesses rather than the ground-truth tokens. We address this problem by introducing a \textit{critic} network that is trained to predict the value of an output token, given the policy of an \textit{actor} network. This results in a training procedure that is much closer to the test phase, and allows us to directly optimize for a task-specific score such as BLEU. Crucially, since we leverage these techniques in the supervised learning setting rather than the traditional RL setting, we condition the critic network on the ground-truth output. We show that our method leads to improved performance on both a synthetic task, and for German-English machine translation. Our analysis paves the way for such methods to be applied in natural language generation tasks, such as machine translation, caption generation, and dialogue modelling.

cs.LG↗

A Neural Knowledge Language Model

Current language models have a significant limitation in the ability to encode and decode factual knowledge. This is mainly because they acquire such knowledge from statistical co-occurrences although most of the knowledge words are rarely observed. In this paper, we propose a Neural Knowledge Language Model (NKLM) which combines symbolic knowledge provided by the knowledge graph with the RNN language model. By predicting whether the word to generate has an underlying fact or not, the model can generate such knowledge-related words by copying from the description of the predicted fact. In experiments, we show that the NKLM significantly improves the performance while generating a much smaller number of unknown words.

cs.CL↗

Mode Regularized Generative Adversarial Networks

Although Generative Adversarial Networks achieve state-of-the-art results on a variety of generative tasks, they are regarded as highly unstable and prone to miss modes. We argue that these bad behaviors of GANs are due to the very particular functional shape of the trained discriminators in high dimensional spaces, which can easily make training stuck or push probability mass in the wrong direction, towards that of higher concentration than that of the data generating distribution. We introduce several ways of regularizing the objective, which can dramatically stabilize the training of GAN models. We also show that our regularizers can help the fair distribution of probability mass across the modes of the data generating distribution, during the early phases of training and thus providing a unified solution to the missing modes problem.

cs.LG↗

A Robust Adaptive Stochastic Gradient Method for Deep Learning

Stochastic gradient algorithms are the main focus of large-scale optimization problems and led to important successes in the recent advancement of the deep learning algorithms. The convergence of SGD depends on the careful choice of learning rate and the amount of the noise in stochastic estimates of the gradients. In this paper, we propose an adaptive learning rate algorithm, which utilizes stochastic curvature information of the loss function for automatically tuning the learning rates. The information about the element-wise curvature of the loss function is estimated from the local statistics of the stochastic first order gradients. We further propose a new variance reduction technique to speed up the convergence. In our experiments with deep neural networks, we obtained better performance compared to the popular stochastic gradient algorithms.

cs.LG↗

Recurrent Neural Networks With Limited Numerical Precision

Recurrent Neural Networks (RNNs) produce state-of-art performance on many machine learning tasks but their demand on resources in terms of memory and computational power are often high. Therefore, there is a great interest in optimizing the computations performed with these models especially when considering development of specialized low-power hardware for deep networks. One way of reducing the computational needs is to limit the numerical precision of the network weights and biases. This has led to different proposed rounding methods which have been applied so far to only Convolutional Neural Networks and Fully-Connected Networks. This paper addresses the question of how to best reduce weight precision during training in the case of RNNs. We present results from the use of different stochastic and deterministic reduced precision training methods applied to three major RNN types which are then tested on several datasets. The results show that the weight binarization methods do not work with the RNNs. However, the stochastic and deterministic ternarization, and pow2-ternarization methods gave rise to low-precision RNNs that produce similar and even higher accuracy on certain datasets therefore providing a path towards training more efficient implementations of RNNs in specialized hardware.

cs.NE↗

Recurrent Neural Networks With Limited Numerical Precision

Recurrent Neural Networks (RNNs) produce state-of-art performance on many machine learning tasks but their demand on resources in terms of memory and computational power are often high. Therefore, there is a great interest in optimizing the computations performed with these models especially when considering development of specialized low-power hardware for deep networks. One way of reducing the computational needs is to limit the numerical precision of the network weights and biases, and this will be addressed for the case of RNNs. We present results from the use of different stochastic and deterministic reduced precision training methods applied to two major RNN types, which are then tested on three datasets. The results show that the stochastic and deterministic ternarization, pow2- ternarization, and exponential quantization methods gave rise to low-precision RNNs that produce similar and even higher accuracy on certain datasets, therefore providing a path towards training more efficient implementations of RNNs in specialized hardware.

cs.NE↗

Maximum-Likelihood Augmented Discrete Generative Adversarial Networks

Despite the successes in capturing continuous distributions, the application of generative adversarial networks (GANs) to discrete settings, like natural language tasks, is rather restricted. The fundamental reason is the difficulty of back-propagation through discrete random variables combined with the inherent instability of the GAN training objective. To address these problems, we propose Maximum-Likelihood Augmented Discrete Generative Adversarial Networks. Instead of directly optimizing the GAN objective, we derive a novel and low-variance objective using the discriminator's output that follows corresponds to the log-likelihood. Compared with the original, the new objective is proved to be consistent in theory and beneficial in practice. The experimental results on various discrete datasets demonstrate the effectiveness of the proposed approach.

cs.AI↗

Learning Normalized Inputs for Iterative Estimation in Medical Image Segmentation

In this paper, we introduce a simple, yet powerful pipeline for medical image segmentation that combines Fully Convolutional Networks (FCNs) with Fully Convolutional Residual Networks (FC-ResNets). We propose and examine a design that takes particular advantage of recent advances in the understanding of both Convolutional Neural Networks as well as ResNets. Our approach focuses upon the importance of a trainable pre-processing when using FC-ResNets and we show that a low-capacity FCN model can serve as a pre-processor to normalize medical input data. In our image segmentation pipeline, we use FCNs to obtain normalized images, which are then iteratively refined by means of a FC-ResNet to generate a segmentation prediction. As in other fully convolutional approaches, our pipeline can be used off-the-shelf on different image modalities. We show that using this pipeline, we exhibit state-of-the-art performance on the challenging Electron Microscopy benchmark, when compared to other 2D methods. We improve segmentation results on CT images of liver lesions, when contrasting with standard FCN methods. Moreover, when applying our 2D pipeline on a challenging 3D MRI prostate segmentation challenge we reach results that are competitive even when compared to 3D methods. The obtained results illustrate the strong potential and versatility of the pipeline by achieving highly accurate results on multi-modality images from different anatomical regions and organs.

cs.CV↗

SampleRNN: An Unconditional End-to-End Neural Audio Generation Model

In this paper we propose a novel model for unconditional audio generation based on generating one audio sample at a time. We show that our model, which profits from combining memory-less modules, namely autoregressive multilayer perceptrons, and stateful recurrent neural networks in a hierarchical structure is able to capture underlying sources of variations in the temporal sequences over very long time spans, on three datasets of different nature. Human evaluation on the generated samples indicate that our model is preferred over competing models. We also show how each component of the model contributes to the exhibited performance.

cs.SD↗

Memory Augmented Neural Networks with Wormhole Connections

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 effects of vanishing gradients by creating shortcut (or wormhole) connections. Based on this observation, we propose a novel memory augmented neural network model called TARDIS (Temporal Automatic Relation Discovery in Sequences). The controller of TARDIS can store a selective set of embeddings of its own previous hidden states into an external memory and revisit them as and when needed. For TARDIS, memory acts as a storage for wormhole connections to the past to propagate the gradients more effectively and it helps to learn the temporal dependencies. The memory structure of TARDIS has similarities to both Neural Turing Machines (NTM) and Dynamic Neural Turing Machines (D-NTM), but both read and write operations of TARDIS are simpler and more efficient. We use discrete addressing for read/write operations which helps to substantially to reduce the vanishing gradient problem with very long sequences. Read and write operations in TARDIS are tied with a heuristic once the memory becomes full, and this makes the learning problem simpler when compared to NTM or D-NTM type of architectures. We provide a detailed analysis on the gradient propagation in general for MANNs. We evaluate our models on different long-term dependency tasks and report competitive results in all of them.

cs.LG↗

On Random Weights for Texture Generation in One Layer Neural Networks

Recent work in the literature has shown experimentally that one can use the lower layers of a trained convolutional neural network (CNN) to model natural textures. More interestingly, it has also been experimentally shown that only one layer with random filters can also model textures although with less variability. In this paper we ask the question as to why one layer CNNs with random filters are so effective in generating textures? We theoretically show that one layer convolutional architectures (without a non-linearity) paired with the an energy function used in previous literature, can in fact preserve and modulate frequency coefficients in a manner so that random weights and pretrained weights will generate the same type of images. Based on the results of this analysis we question whether similar properties hold in the case where one uses one convolution layer with a non-linearity. We show that in the case of ReLu non-linearity there are situations where only one input will give the minimum possible energy whereas in the case of no nonlinearity, there are always infinite solutions that will give the minimum possible energy. Thus we can show that in certain situations adding a ReLu non-linearity generates less variable images.

cs.CV↗