Metraipolis › archive › dc_e86c1996a7ee

Regularised RL improves sim-to-real transfer — arXiv:2610.32855

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.32855
License (as archived)
CC0-1.0
Archived
2026-10-26T20:13:12.086Z
SHA-256 of the archived text
e9a2ca9cfad17819fdf641b2dc2346d96a9ec35a439c428954d8701ec04130ab
Stable URL of this page
https://metraipolis.com/documento/dc_e86c1996a7ee

How to cite

Regularised RL improves sim-to-real transfer — arXiv:2610.32855 — arXiv (en), https://arxiv.org/abs/2610.32855 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e86c1996a7ee (Metraipolis archive dc_e86c1996a7ee, sha256 e9a2ca9cfad17819fdf641b2dc2346d96a9ec35a439c428954d8701ec04130ab)
BibTeX
@misc{dce86c1996a7ee,
  title = {Regularised RL improves sim-to-real transfer — arXiv:2610.32855},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.32855},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e86c1996a7ee; sha256 e9a2ca9cfad17819fdf641b2dc2346d96a9ec35a439c428954d8701ec04130ab},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32855. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Regularised reinforcement learning (RL) has provable robustness guarantees that should ease the transfer of policies from simulation to robots, known as sim-to-real. Yet, weakly regularised algorithms like Proximal Policy Optimisation (PPO) remain the de facto choice in sim-to-real pipelines, while the practical benefits of regularisation for transfer remain poorly characterised. In this work, we study the sim-to-real performance of commonly used RL algorithms, with the aim of characterising the effects of regularisation on robotic performance. To that end, we derive Soft MellowMax Actor-Critic (SMM-AC), a novel regularised RL algorithm based on the soft mellowmax value function, which carries strong robustness guarantees. On MuJoCo continuous control benchmarks, SMM-AC is competitive with baselines like Soft Actor-Critic (SAC). On real robots, SMM-AC achieves higher return, while exhibiting greater robustness to perturbations in observations, the environment, and dynamics than baselines. Together, these results demonstrate that regularised RL improves real-world policy deployment through better robustness to the sim-to-real gap, and changing real-world conditions.

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/2610.32855 — 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 2026-10-26T19:49:03.435Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).