Papers › Gating Revisited: Deep Multi-layer RNNs That Can Be Trained

Gating Revisited: Deep Multi-layer RNNs That Can Be Trained

25 Nov 2019arXiv:1911.11033archive 2025-07-28

Mehmet Ozgur Turkoglu, Stefano D'Aronco, Jan Dirk Wegner, Konrad Schindler

We propose a new STAckable Recurrent cell (STAR) for recurrent neural networks (RNNs), which has fewer parameters than widely used LSTM and GRU while being more robust against vanishing or exploding gradients. Stacking recurrent units into deep architectures suffers from two major limitations: (i) many recurrent cells (e.g., LSTMs) are costly in terms of parameters and computation resources; and (ii) deep RNNs are prone to vanishing or exploding gradients during training. We investigate the training of multi-layer RNNs and examine the magnitude of the gradients as they propagate through the network in the "vertical" direction. We show that, depending on the structure of the basic recurrent unit, the gradients are systematically attenuated or amplified. Based on our analysis we design a new type of gated cell that better preserves gradient magnitude. We validate our design on a large number of sequence modelling tasks and demonstrate that the proposed STAR cell allows to build and train deeper recurrent architectures, ultimately leading to improved performance while being computationally more efficient.

PaperPDFCode

Code

0zgur0/STAR_Network officialmentioned in papermentioned on GitHubtf report
croros/STAR_Network_Pytorch mentioned on GitHubpytorch report
titu1994/tf_star_rnn mentioned on GitHubtf 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

Action RecognitionAction Recognition In VideosLanguage ModellingMusic ModelingSequential Image Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Recognition In Videos Jester (Gesture Recognition) convSTAR Val 92.7 #6 of 9 Archive leaderboard report
Language Modelling Penn Treebank (Character Level) STAR Bit per Character (BPC) 1.30 #17 of 20 Archive leaderboard report
Sequential Image Classification Sequential MNIST STAR Unpermuted Accuracy 99.4% #29 of 30 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.

Methods

GRULSTMSigmoid ActivationTanh Activation

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