Research map / Time series / Causal and generative time series
Learning to Adaptively Scale Recurrent Neural Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add context-dependent adaptive scale selection | LSTM (ASLSTM) · Low density signal type identification | accuracy: 81.3 -> 97.7 | Helped |
| Add context-dependent adaptive scale selection | GRU (ASGRU) · Low density signal type identification | accuracy: 84.1 -> 98.0 | Helped |
| Add adaptive rather than fixed scale selection | LSTM · MNIST unpermuted | accuracy: 97.4 -> 98.3 | Helped |
| Add adaptive rather than fixed scale selection | LSTM · MNIST permuted | accuracy: 87.7 -> 90.8 | Helped |
| Add adaptive rather than fixed scale selection | GRU · MNIST unpermuted | accuracy: 97.0 -> 98.1 | Helped |
| Add adaptive rather than fixed scale selection | GRU · MNIST permuted | accuracy: 89.8 -> 91.2 | Helped |
| Add adaptive rather than fixed scale selection | LSTM · FMA-small MFCC | accuracy: 37.7 -> 40.9 | Helped |
| Add adaptive rather than fixed scale selection | GRU · FMA-small MFCC | accuracy: 38.5 -> 42.4 | Helped |
| Add adaptive rather than fixed scale selection | LSTM · FMA-small raw audio | accuracy: 18.9 -> 20.1 | Helped |
| Add adaptive rather than fixed scale selection | GRU · FMA-small raw audio | accuracy: 18.4 -> 19.5 | Helped |
| Add adaptive rather than fixed scale selection | LSTM · WikiText-2 | perplexity: 97.7 -> 93.8 | Helped |
| Add adaptive rather than fixed scale selection | GRU · WikiText-2 | perplexity: 95.4 -> 92.6 | Helped |
Closest papers by meaning
- Identifying nonlinear dynamical systems with multiple time scales and long-range dependencies (ICLR 2021 · similarity 0.87)
- Hierarchical Multiscale Recurrent Neural Networks (ICLR 2017 · similarity 0.87)
- AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networks (ICLR 2019 · similarity 0.87)
- Time Adaptive Recurrent Neural Network (CVPR 2021 · similarity 0.87)
- Learning dynamics in linear recurrent neural networks (ICML 2025 · similarity 0.87)
- Recurrent neural networks: vanishing and exploding gradients are not the end of the story (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.