Research map / Time series / Causal and generative time series
Bifurcations and loss jumps in RNN training
Lukas Eisenmann, Zahra Monfared, Niclas Alexander Göring, Daniel Durstewitz
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 SCYFI instead of exhaustive search to find k-cycles | exhaustive search over linear subregions · PLRNN, M=2, cycle order k | number of subregions searched | Improved |
| add generalized teacher forcing (alpha=0.1) to gradient-descent PLRNN training | PLRNN (M=10) trained by gradient descent with clipping · 2d cycle task | loss curve smoothness / bifurcation avoidance | Improved |
| training trajectory crosses a bifurcation curve in parameter space | PLRNN trained by BPTT on a 16-cycle · 2d toy example | loss | Hurt |
Closest papers by meaning
- Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction (NeurIPS 2024 · similarity 0.89)
- Mechanistic Interpretability of RNNs emulating Hidden Markov Models (NeurIPS 2025 · similarity 0.88)
- Learning Dynamics of RNNs in Closed-Loop Environments (NeurIPS 2025 · similarity 0.88)
- Learning dynamics in linear recurrent neural networks (ICML 2025 · similarity 0.88)
- Inference of Neural Dynamics Using Switching Recurrent Neural Networks (NeurIPS 2024 · similarity 0.88)
- Parallel-in-Time Training of Recurrent Neural Networks for Dynamical Systems Reconstruction (NeurIPS 2026 · 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.
