Research map / Time series / Long-term forecasting
Nonlinear Dynamic Boltzmann Machines for Time-Series Prediction
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 RNN hidden layer controlling bias to Gaussian DyBM (ablation) | G-DyBM · Retail-price prediction | best test RMSE: 0.0718 -> 0.0564 | Helped |
| add RNN hidden layer controlling bias to Gaussian DyBM (ablation) | G-DyBM · Sunspots prediction | best test RMSE: 0.1327 -> 0.0770 | Helped |
| replace VAR with RNN-G-DyBM (eligibility traces plus nonlinear RNN) | VAR (d=2) · Retail-price prediction | test RMSE: 0.0846 -> 0.0580 | Helped |
| replace VAR with RNN-G-DyBM | VAR (d=3) · Sunspots prediction | test RMSE: 0.1859 -> 0.0770 | Helped |
| replace LSTM with RNN-G-DyBM | LSTM · Retail-price prediction | test RMSE: 0.0674 -> 0.0564 | Helped |
| replace LSTM with RNN-G-DyBM | LSTM · Sunspots prediction | test RMSE: 0.07342 -> 0.0770 | Hurt |
| replace LSTM with RNN-G-DyBM | LSTM · Sunspots prediction | training time per epoch (s): 11.2132 -> 0.7014 | Helped |
| increase eligibility-trace decay rate lambda (ablation) | RNN-G-DyBM · Retail-price prediction | test RMSE | Mixed |
| add eligibility traces and RNN to VAR (lambda>0) (ablation) | VAR · Noisy sine and NARMA-30 synthetic | MSE: reduces error by more than 30 % | Helped |
Closest papers by meaning
- STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction (ICLR 2024 · similarity 0.85)
- Parallel-in-Time Training of Recurrent Neural Networks for Dynamical Systems Reconstruction (NeurIPS 2026 · similarity 0.84)
- Learning with Restricted Boltzmann Machines: Asymptotics of AMP and GD in High Dimensions (NeurIPS 2025 · similarity 0.84)
- DDN: Dual-domain Dynamic Normalization for Non-stationary Time Series Forecasting (NeurIPS 2024 · similarity 0.84)
- xLSTM-Mixer: Multivariate Time Series Forecasting by Mixing via Scalar Memories (NeurIPS 2025 · similarity 0.83)
- Dynamical Diffusion: Learning Temporal Dynamics with Diffusion Models (ICLR 2025 · similarity 0.83)
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.