Metraipolis › archive › dc_f9c4adeed773
Neural ODEs Meet Concurrent Learning: Stable Online Learning with Lyapunov Guarantees — arXiv:2609.32289
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32289
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:39:37.955Z
- SHA-256 of the archived text
7d9f6cfbfba04d02a2775456e567202c6a7e85ebe80da7f70039c46b48a1b0d2- Stable URL of this page
https://metraipolis.com/documento/dc_f9c4adeed773
How to cite
Neural ODEs Meet Concurrent Learning: Stable Online Learning with Lyapunov Guarantees — arXiv:2609.32289 — arXiv (en), https://arxiv.org/abs/2609.32289 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f9c4adeed773 (Metraipolis archive dc_f9c4adeed773, sha256 7d9f6cfbfba04d02a2775456e567202c6a7e85ebe80da7f70039c46b48a1b0d2)
BibTeX
@misc{dcf9c4adeed773,
title = {Neural ODEs Meet Concurrent Learning: Stable Online Learning with Lyapunov Guarantees — arXiv:2609.32289},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32289},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f9c4adeed773; sha256 7d9f6cfbfba04d02a2775456e567202c6a7e85ebe80da7f70039c46b48a1b0d2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32289. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Neural ODEs learn dynamics from trajectory losses, but their adjoint gradients lack the regressor-times-parameter-error structure on which Lyapunov analyses of online adaptation rest, so training on streaming data comes without stability guarantees. We show that this structure is in fact present: the adjoint gradient decomposes exactly into a positive semi-definite trajectory operator acting on the parameter error plus a nonlinear perturbation with explicit, horizon-dependent bounds. A quadratic Lyapunov function then certifies online Neural ODE training over sliding windows under computable gain and horizon conditions, and the same certificate extends to stored data: its drift branch recovers concurrent learning, and its trajectory branch yields NODE-CL, a stored-segment Gauss-Newton method built on batched forward sensitivities that needs no state-derivative estimates. On four DeepMind Control Suite domains, NODE-CL attains the lowest median prediction error on three under velocity measurement noise, where observer-based concurrent learning degrades by up to 8x; with clean measurements it is best on the pendulum and within a factor of 1.6 of the best stored-data baseline on the cartpole and reacher. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Systems and Control (eess.SY) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.32289 — 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 2026-10-07T22:14:28.594Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).