Journal ArticleDOI
Learning long-term dependencies with gradient descent is difficult
TLDR
This work shows why gradient based learning algorithms face an increasingly difficult problem as the duration of the dependencies to be captured increases, and exposes a trade-off between efficient learning by gradient descent and latching on information for long periods.Abstract:
Recurrent neural networks can be used to map input sequences to output sequences, such as for recognition, production or prediction problems. However, practical difficulties have been reported in training recurrent neural networks to perform tasks in which the temporal contingencies present in the input/output sequences span long intervals. We show why gradient based learning algorithms face an increasingly difficult problem as the duration of the dependencies to be captured increases. These results expose a trade-off between efficient learning by gradient descent and latching on information for long periods. Based on an understanding of this problem, alternatives to standard gradient descent are considered. >read more
Citations
More filters
Posted Content
Convolutional Neural Networks Analyzed via Convolutional Sparse Coding
TL;DR: This work proposes a novel multi-layer model, ML-CSC, in which signals are assumed to emerge from a cascade of CSC layers, and presents an alternative to the forward pass, which is connected to deconvolutional, recurrent and residual networks, and has better theoretical guarantees.
Journal ArticleDOI
Beyond brain size: Uncovering the neural correlates of behavioral and cognitive specialization
Corina J. Logan,Shahar Avin,Neeltje J. Boogert,Andrew Buskell,Fiona R. Cross,Adrian Currie,Sarah A. Jelbert,Dieter Lukas,Rafael Mares,Ana F. Navarrete,Shuichi Shigeno,Stephen H. Montgomery +11 more
TL;DR: In this paper, the funders of the workshop on which this article is based were the Isaac Newton Trust and the Leverhulme Trust, who provided an Early Career Fellowship to CJL and an Independent Research Fellowship to SHM.
Journal ArticleDOI
Evaluation of Deep Learning Models for Multi-Step Ahead Time Series Prediction
TL;DR: In this paper, the performance of deep learning models for multi-step-ahead time series prediction was evaluated and the results showed that the bidirectional and encoder-decoder LSTM networks provided the best performance in accuracy for the given time series problems.
Posted Content
A Tutorial on Deep Learning for Music Information Retrieval.
TL;DR: The basic principles and prominent works in deep learning for MIR are laid out and the network structures that have been successful in MIR problems are outlined to facilitate the selection of building blocks for the problems at hand.
Journal ArticleDOI
A Context-Aware Recurrent Encoder for Neural Machine Translation
TL;DR: This paper proposes a novel context-aware recurrent encoder (CAEncoder), as an alternative to the widely-used bidirectional encoder, such that the future and history contexts can be fully incorporated into the learned source representations.
References
More filters
Journal ArticleDOI
Optimization by Simulated Annealing
TL;DR: There is a deep and useful connection between statistical mechanics and multivariate or combinatorial optimization (finding the minimum of a given function depending on many parameters), and a detailed analogy with annealing in solids provides a framework for optimization of very large and complex systems.
Book ChapterDOI
Learning internal representations by error propagation
TL;DR: This chapter contains sections titled: The Problem, The Generalized Delta Rule, Simulation Results, Some Further Generalizations, Conclusion.
Book
Learning internal representations by error propagation
TL;DR: In this paper, the problem of the generalized delta rule is discussed and the Generalized Delta Rule is applied to the simulation results of simulation results in terms of the generalized delta rule.
Journal ArticleDOI
A learning algorithm for continually running fully recurrent neural networks
Ronald J. Williams,David Zipser +1 more
TL;DR: The exact form of a gradient-following learning algorithm for completely recurrent networks running in continually sampled time is derived and used as the basis for practical algorithms for temporal supervised learning tasks.
Journal ArticleDOI
Minimizing multimodal functions of continuous variables with the “simulated annealing” algorithm—Corrigenda for this article is available here
TL;DR: A new global optimization algorithm for functions of continuous variables is presented, derived from the “Simulated Annealing” algorithm recently introduced in combinatorial optimization, which is quite costly in terms of function evaluations, but its cost can be predicted in advance, depending only slightly on the starting point.