Metraipolis › archive › dc_34236dc26689

Learning Transferable Policies from Action-free Time Series Through Dynamical Embeddings — arXiv:2610.03065

Source
arXiv
Original
https://arxiv.org/abs/2610.03065
License (as archived)
CC0-1.0
Archived
2026-10-05T23:10:05.856Z
SHA-256 of the archived text
c5a1dc651af307f7c77c25de12f879cd9d4fff8ecbd4d7d28a74367bd47df401
Stable URL of this page
https://metraipolis.com/documento/dc_34236dc26689

How to cite

Learning Transferable Policies from Action-free Time Series Through Dynamical Embeddings — arXiv:2610.03065 — arXiv, https://arxiv.org/abs/2610.03065 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_34236dc26689 (Metraipolis archive dc_34236dc26689, sha256 c5a1dc651af307f7c77c25de12f879cd9d4fff8ecbd4d7d28a74367bd47df401)
BibTeX
@misc{dc34236dc26689,
  title = {Learning Transferable Policies from Action-free Time Series Through Dynamical Embeddings — arXiv:2610.03065},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.03065},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_34236dc26689; sha256 c5a1dc651af307f7c77c25de12f879cd9d4fff8ecbd4d7d28a74367bd47df401},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03065. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learning control from action-free recordings is challenging because intervention effects are unobserved and policies may exploit errors in reconstructed dynamics. We present a hierarchical model-based reinforcement learning framework that uses shared structure across related systems to learn system-specific control policies from action-free recordings. A hierarchical dynamical system reconstruction model captures shared dynamics and individual variation through low-dimensional embeddings. These embeddings are then reused to parameterize shared policy and value networks, linking differences in reconstructed dynamics to differences in control. Policies are trained entirely via simulation under an explicit intervention model with additive latent perturbations. Piecewise-linear recurrent neural networks enable mechanistic analyses of the controlled dynamics, while decoder-based constraints make the immediate effects of interventions interpretable in observation space and permit interventions on one modality while protecting another from direct manipulation. On Lorenz-63 and double-pendulum systems, hierarchical policies improve transfer over independently trained policies. On Lorenz-63, they also achieve a higher mean reward than repeated planning with the same reconstructed models, perform comparably to methods trained with controlled interactions, and generalize to systems absent from policy training after embedding inference alone. Applications to neural-behavioral recordings demonstrate suppression of predicted movement under constrained neural perturbations. Together, these findings show how shared dynamical representations support transferable control and mechanistic hypothesis generation from action-free recordings.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.03065 — 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 2026-10-05T23:01:38.666Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).