Metraipolis › archive › dc_361ff367a7a5
Continual Learning of Dynamical Systems in Recurrent Neural Networks through Recyclable Unit Gating — arXiv:2609.38356
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38356
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T13:45:14.212Z
- SHA-256 of the archived text
4a2cf4c0d0f4120aa85bc94c21c2fb0baaa7bf7acc24b7423c49fe2b4701edf0- Stable URL of this page
https://metraipolis.com/documento/dc_361ff367a7a5
How to cite
Continual Learning of Dynamical Systems in Recurrent Neural Networks through Recyclable Unit Gating — arXiv:2609.38356 — arXiv (en), https://arxiv.org/abs/2609.38356 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_361ff367a7a5 (Metraipolis archive dc_361ff367a7a5, sha256 4a2cf4c0d0f4120aa85bc94c21c2fb0baaa7bf7acc24b7423c49fe2b4701edf0)
BibTeX
@misc{dc361ff367a7a5,
title = {Continual Learning of Dynamical Systems in Recurrent Neural Networks through Recyclable Unit Gating — arXiv:2609.38356},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38356},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_361ff367a7a5; sha256 4a2cf4c0d0f4120aa85bc94c21c2fb0baaa7bf7acc24b7423c49fe2b4701edf0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38356. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Dynamical Systems Reconstruction (DSR) aims to infer models from observed time series that reproduce a system's qualitative long-term behavior. Continual DSR (cDSR) requires learning new systems while preserving previously learned dynamics, yet even small parameter updates in recurrent models can qualitatively alter their behavior over long autonomous rollouts. We benchmark established continual learning (CL) methods spanning parameter regularization, replay, and parameter isolation on the fully trainable and interpretable Almost-Linear RNN (AL-RNN). Parameter isolation preserves earlier dynamics most effectively, but excessive task-specific allocations can rapidly exhaust a fixed-size network. We therefore introduce Continually-Recyclable Unit-Gating (CRUG), which conserves capacity through compact allocation and forward transfer. Differentiable gates trained with an $L_0$-based penalty select task-specific units, while unused units are recycled for subsequent tasks. Directed connections allow later tasks to reuse earlier representations without affecting the dynamics of previously committed units. CRUG achieves the strongest reconstruction--capacity trade-off among the tested methods with zero forgetting and reliably learns a heterogeneous sequence of nonlinear and chaotic systems. Furthermore, we show that forward transfer is more pronounced and useful when tasks share similar underlying dynamics. Lastly, we demonstrate that CRUG's advantages extend beyond autonomous cDSR to sequential cognitive tasks. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.38356 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T13:18:10.193Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).