Recurrent highway networks

Julian Georg Zilly, Rupesh Kumar Srivastava, Jan Koutnik, Jürgen Schmidhuber

Research output: Chapter in Book/Report/Conference proceedingConference contribution

50 Scopus citations

Abstract

Many sequential processing tasks require complex nonlinear transition functions from one step to the next. However, recurrent neural networks with "deep" transition functions remain difficult to train, even when using Long Short-Term Memory (LSTM) networks. We introduce a novel theoretical analysis of recurrent networks based on Geršgorin's circle theorem that illuminates several modeling and optimization issues and improves our understanding of the LSTM cell. Based on this analysis we propose Recurrent Highway Networks, which extend the LSTM architecture to allow step-to-step transition depths larger than one. Several language modeling experiments demonstrate that the proposed architecture results in powerful and efficient models. On the Penn Treebank corpus, solely increasing the transition depth from 1 to 10 improves word-level perplexity from 90.6 to 65.4 using the same number of parameters. On the larger Wikipedia datasets for character prediction (text8 and enwik8), RHNs outperform all previous results and achieve an entropy of 1.27 bits per character.
Original languageEnglish (US)
Title of host publication34th International Conference on Machine Learning, ICML 2017
PublisherInternational Machine Learning Society (IMLS)rasmussen@ptd.net
Pages6346-6357
Number of pages12
ISBN (Print)9781510855144
StatePublished - Jan 1 2017
Externally publishedYes

Fingerprint

Dive into the research topics of 'Recurrent highway networks'. Together they form a unique fingerprint.

Cite this