Metraipolis › archive › dc_b98dfa5f65d2

DRAP: Dynamics-Risk Adaptive Plasticity for Continual Learning in Physical AI — arXiv:2611.05029

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.05029
License (as archived)
CC0-1.0
Archived
2026-11-04T12:41:03.871Z
SHA-256 of the archived text
ec11317fbf63008a3d93f9e5d7359a187c20f5b67ef22a685bc5ebedcc911bd7
Stable URL of this page
https://metraipolis.com/documento/dc_b98dfa5f65d2

How to cite

DRAP: Dynamics-Risk Adaptive Plasticity for Continual Learning in Physical AI — arXiv:2611.05029 — arXiv (en), https://arxiv.org/abs/2611.05029 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b98dfa5f65d2 (Metraipolis archive dc_b98dfa5f65d2, sha256 ec11317fbf63008a3d93f9e5d7359a187c20f5b67ef22a685bc5ebedcc911bd7)
BibTeX
@misc{dcb98dfa5f65d2,
  title = {DRAP: Dynamics-Risk Adaptive Plasticity for Continual Learning in Physical AI — arXiv:2611.05029},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.05029},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b98dfa5f65d2; sha256 ec11317fbf63008a3d93f9e5d7359a187c20f5b67ef22a685bc5ebedcc911bd7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05029. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Continual learning in physical AI differs from conventional continual learning because small per-step policy errors can accumulate over a trajectory. Each deviation may move the agent outside the training distribution, increase subsequent prediction errors, and ultimately cause closed-loop task failure. This work introduces Dynamics-Risk Adaptive Plasticity (DRAP), a replay-free method that estimates the historical importance of adaptable model tensors from sample-level action gradients weighted by local trajectory-amplification risk. A ridge-regularized local linear model is fitted to expert state-action transitions, and the largest singular value of its state-transition matrix provides a policy-independent, Lyapunov-inspired estimate of one-step perturbation amplification. After each task, risk-weighted squared gradients are accumulated in a tensor-level importance memory. During subsequent tasks, DRAP uses normalized cumulative importance to scale the effective AdamW learning rate of each tensor, immediately reducing the plasticity of components that supported previous behavior at dynamically sensitive trajectory points. We instantiate DRAP on a fixed KappaTune-selected subnetwork of OpenVLA and evaluate replay-free sequential adaptation across the LIBERO Spatial, Object, and Goal suites. Compared with standard AdamW fine-tuning and EWC, the results show substantially improved retention of previously learned behaviors while preserving strong new-suite acquisition.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2611.05029 — 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 2026-11-04T12:29:47.321Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).