Short-Term Memory Optimization in Recurrent Neural Networks by Autoencoder-based Initialization
Abstract
Training RNNs to learn long-term dependencies is difficult due to vanishing gradients. We explore an alternative solution based on explicit memorization using linear autoencoders for sequences, which allows to maximize the short-term memory and that can be solved with a closed-form solution without backpropagation. We introduce an initialization schema that pretrains the weights of a recurrent neural network to approximate the linear autoencoder of the input sequences and we show how such pretraining can better support solving hard classification tasks with long sequences. We test our approach on sequential and permuted MNIST. We show that the proposed approach achieves a much lower reconstruction error for long sequences and a better gradient propagation during the finetuning phase.
Cite
@article{arxiv.2011.02886,
title = {Short-Term Memory Optimization in Recurrent Neural Networks by Autoencoder-based Initialization},
author = {Antonio Carta and Alessandro Sperduti and Davide Bacciu},
journal= {arXiv preprint arXiv:2011.02886},
year = {2020}
}
Comments
Accepted at NeurIPS 2020 workshop "Beyond Backpropagation: Novel Ideas for Training Neural Architectures"