Research map / Time series / Causal and generative time series
Identifying nonlinear dynamical systems with multiple time scales and long-range dependencies
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 manifold-attractor regularization to a PLRNN | PLRNN · Bursting cortical neuron DS | KL divergence | Helped |
| Add manifold-attractor regularization to a PLRNN | PLRNN · Bursting cortical neuron DS, low-frequency components (≤50 Hz) | power-spectrum MSE | Helped |
| Increase manifold-attractor regularization | PLRNN · Bursting cortical neuron DS, high-frequency components (>50 Hz) | power-spectrum MSE | No clear effect |
| Replace manifold-attractor regularization with an orthogonality constraint (ablation) | PLRNN · Bursting cortical neuron DS | power-spectrum MSE | Hurt |
| Replace manifold-attractor regularization with a plain L2 constraint on weight matrices (ablation) | PLRNN · Bursting cortical neuron DS | power-spectrum MSE | Hurt |
| Replace the proposed regularization with standard L2 regularization on a fraction of states (ablation) | PLRNN · Tasks tested | performance | Hurt |
| Replace the proposed regularization with standard L2 regularization on all weights (ablation) | PLRNN · Tasks tested | performance | Hurt |
| Remove the manifold-attractor regularization (ablation) | PLRNN · Bursting cortical neuron DS | KL divergence | Hurt |
| Vary the regularized-unit proportion across a broad range (ablation) | rPLRNN · Tasks tested | performance | No clear effect |
| Use the PLRNN structure instead of a vanilla RNN structure with identity initialization (ablation) | Identity-initialized RNN · Addition, multiplication, and sequential MNIST | performance | Helped |
| Add manifold-attractor regularization | PLRNN · Addition, multiplication, and sequential MNIST | performance | Helped |
Closest papers by meaning
- Parallel-in-Time Training of Recurrent Neural Networks for Dynamical Systems Reconstruction (NeurIPS 2026 · similarity 0.92)
- Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction (NeurIPS 2024 · similarity 0.89)
- Inferring stochastic low-rank recurrent neural networks from neural data (NeurIPS 2024 · similarity 0.89)
- Inference of Neural Dynamics Using Switching Recurrent Neural Networks (NeurIPS 2024 · similarity 0.89)
- Learning dynamics in linear recurrent neural networks (ICML 2025 · similarity 0.88)
- When are dynamical systems learned from time series data statistically accurate? (NeurIPS 2024 · similarity 0.88)
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.