Research map / Time series / Long-term forecasting
Memory-Gated Recurrent 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 |
|---|---|---|---|
| Use separate marginal-memory and joint-memory components | mGRN · Simulated time series | MSE: 21.93 -> 21.88 | Helped |
| Use a total variable split instead of two groups (ablation) | mGRN · Simulated time series | MSE: 21.91 -> 21.88 | Helped |
| Use a total variable split instead of two groups (ablation) | Channel-wise LSTM · Simulated time series | MSE: 22.06 -> 21.93 | Helped |
| Use mGRN instead of the strongest reported baseline | MIMIC-III · In-hospital mortality | AUC-PR: 0.515 -> 0.523 | Helped |
| Use mGRN instead of channel-wise LSTM | MIMIC-III · Decompensation | AUC-ROC: 0.906 -> 0.911 | Helped |
| Use mGRN instead of channel-wise LSTM | MIMIC-III · Decompensation | AUC-PR: 0.333 -> 0.347 | Helped |
| Use mGRN instead of channel-wise LSTM | MIMIC-III · Length of stay | Kappa: 0.442 -> 0.447 | Helped |
| Use mGRN instead of LSTM | MIMIC-III · Length of stay | MAD: 123.1 -> 124.6 | Hurt |
| Use mGRN instead of the strongest reported baseline | UEA · Character trajectories (CT) | accuracy: 0.997 -> 0.995* | Hurt |
| Use mGRN instead of TapNet | UEA · Face detection (FD) | accuracy: 0.556* -> 0.596 | Helped |
| Use mGRN instead of the strongest reported baseline | UEA · LSST | accuracy: 0.590* -> 0.614 | Helped |
Closest papers by meaning
- Hierarchically Gated Recurrent Neural Network for Sequence Modeling (NeurIPS 2023 · similarity 0.88)
- Variable Computation in Recurrent Neural Networks (ICLR 2017 · similarity 0.88)
- Can recurrent neural networks warp time? (ICLR 2018 · similarity 0.88)
- On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis (ICLR 2021 · similarity 0.88)
- Combining Recurrent, Convolutional, and Continuous-time Models with Linear State Space Layers (NeurIPS 2021 · similarity 0.88)
- On the Convergence Rate of Training Recurrent Neural Networks (NeurIPS 2019 · similarity 0.87)
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.