Papers › Fast-Slow Recurrent Neural Networks

Fast-Slow Recurrent Neural Networks

24 May 2017NeurIPS 2017 12arXiv:1705.08639archive 2025-07-28

Asier Mujika, Florian Meier, Angelika Steger

Processing sequential data of variable length is a major challenge in a wide range of applications, such as speech recognition, language modeling, generative image modeling and machine translation. Here, we address this challenge by proposing a novel recurrent neural network (RNN) architecture, the Fast-Slow RNN (FS-RNN). The FS-RNN incorporates the strengths of both multiscale RNNs and deep transition RNNs as it processes sequential data on different timescales and learns complex transition functions from one time step to the next. We evaluate the FS-RNN on two character level language modeling data sets, Penn Treebank and Hutter Prize Wikipedia, where we improve state of the art results to $1.19$ and $1.25$ bits-per-character (BPC), respectively. In addition, an ensemble of two FS-RNNs achieves $1.20$ BPC on Hutter Prize Wikipedia outperforming the best known compression algorithm with respect to the BPC measure. We also present an empirical investigation of the learning and network dynamics of the FS-RNN, which explains the improved performance compared to other RNN architectures. Our approach is general as any kind of RNN cell is a possible building block for the FS-RNN architecture, and thus can be flexibly applied to different tasks.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

amujika/Fast-Slow-LSTM officialmentioned in papertf report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Language ModelingLanguage ModellingMachine TranslationSpeech Recognitionspeech-recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Language Modelling Hutter Prize Large FS-LSTM-4 Bit per Character (BPC) 1.245 #15 of 18 Archive leaderboard report
Language Modelling Hutter Prize Large FS-LSTM-4 Number of params 47M #15 of 18 Archive leaderboard report
Language Modelling Hutter Prize FS-LSTM-4 Bit per Character (BPC) 1.277 #17 of 18 Archive leaderboard report
Language Modelling Hutter Prize FS-LSTM-4 Number of params 27M #17 of 18 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) FS-LSTM-4 Bit per Character (BPC) 1.190 #10 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) FS-LSTM-4 Number of params 27M #10 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) FS-LSTM-2 Bit per Character (BPC) 1.193 #12 of 20 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) FS-LSTM-2 Number of params 27M #12 of 20 Archive leaderboard report
Language Modelling enwik8 Large FS-LSTM-4 Bit per Character (BPC) 1.25 #35 of 42 Archive leaderboard report
Language Modelling enwik8 Large FS-LSTM-4 Number of params 47M #35 of 42 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections