Metraipolis › archive › dc_38fee209baaa

Inferring Transferable Rewards via Active Inverse Reinforcement Learning — arXiv:2611.03296

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03296
License (as archived)
CC0-1.0
Archived
2026-11-03T18:35:04.398Z
SHA-256 of the archived text
0319044ba38512bed122a9e72a60ff1654ab76d67f22d4f4b2fa752c1ccb941a
Stable URL of this page
https://metraipolis.com/documento/dc_38fee209baaa

How to cite

Inferring Transferable Rewards via Active Inverse Reinforcement Learning — arXiv:2611.03296 — arXiv (en), https://arxiv.org/abs/2611.03296 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_38fee209baaa (Metraipolis archive dc_38fee209baaa, sha256 0319044ba38512bed122a9e72a60ff1654ab76d67f22d4f4b2fa752c1ccb941a)
BibTeX
@misc{dc38fee209baaa,
  title = {Inferring Transferable Rewards via Active Inverse Reinforcement Learning — arXiv:2611.03296},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03296},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_38fee209baaa; sha256 0319044ba38512bed122a9e72a60ff1654ab76d67f22d4f4b2fa752c1ccb941a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03296. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Inverse Reinforcement Learning (IRL) aims to recover expert rewards that transfer across environments with differing dynamics. Prior work has shown that such transfer is possible when gathering expert demonstrations across sufficiently different dynamics. In practice, however, expert demonstrations are costly, making it critical to select the right dynamics to reduce the quantity of expert data required for learning transferable rewards. We propose a two-phase approach: we first cycle through the available dynamics to obtain a reliable reward estimate, then actively select dynamics to reduce our uncertainty about the expert reward. Crucially, this yields novel sample complexity guarantees on transfer that hold for any dynamics, with a rate independent of the number of available dynamics. Empirically, our method significantly reduces the number of expert demonstrations needed to learn a transferable reward.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.HC, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Human-Computer Interaction (cs.HC).
Pagina del lavoro: https://arxiv.org/abs/2611.03296 — 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 2026-11-03T18:12:37.999Z dalla pagina https://arxiv.org/list/cs.HC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).