English

Layer Flexible Adaptive Computational Time

Machine Learning 2021-01-05 v5 Machine Learning

Abstract

Deep recurrent neural networks perform well on sequence data and are the model of choice. However, it is a daunting task to decide the structure of the networks, i.e. the number of layers, especially considering different computational needs of a sequence. We propose a layer flexible recurrent neural network with adaptive computation time, and expand it to a sequence to sequence model. Different from the adaptive computation time model, our model has a dynamic number of transmission states which vary by step and sequence. We evaluate the model on a financial data set and Wikipedia language modeling. Experimental results show the performance improvement of 7\% to 12\% and indicate the model's ability to dynamically change the number of layers along with the computational steps.

Keywords

Cite

@article{arxiv.1812.02335,
  title  = {Layer Flexible Adaptive Computational Time},
  author = {Lida Zhang and Abdolghani Ebrahimi and Diego Klabjan},
  journal= {arXiv preprint arXiv:1812.02335},
  year   = {2021}
}

Comments

11 pages, 5 figures