Metraipolis › archive › dc_0404462e45e2

Arrive and Survive: Scaling Safe Goal-Conditioned Policy Learning from One-Bit Failure Signals — arXiv:2608.26571

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.26571
License (as archived)
CC0-1.0
Archived
2026-10-17T16:09:39.817Z
SHA-256 of the archived text
49129b42b8bb203ec3dc274a63f8a088f78a1d94b81cab791d3a5320d2ce58bf
Stable URL of this page
https://metraipolis.com/documento/dc_0404462e45e2

How to cite

Arrive and Survive: Scaling Safe Goal-Conditioned Policy Learning from One-Bit Failure Signals — arXiv:2608.26571 — arXiv (en), https://arxiv.org/abs/2608.26571 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0404462e45e2 (Metraipolis archive dc_0404462e45e2, sha256 49129b42b8bb203ec3dc274a63f8a088f78a1d94b81cab791d3a5320d2ce58bf)
BibTeX
@misc{dc0404462e45e2,
  title = {Arrive and Survive: Scaling Safe Goal-Conditioned Policy Learning from One-Bit Failure Signals — arXiv:2608.26571},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.26571},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0404462e45e2; sha256 49129b42b8bb203ec3dc274a63f8a088f78a1d94b81cab791d3a5320d2ce58bf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26571. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Contrastive reinforcement learning (CRL) scales effectively in goal-conditioned tasks by casting policy learning into a self-supervised contrastive objective. However, in a failure-terminated Markov decision process, established CRL considers pre-failure future goals only when constructing positive samples, without accounting for the probability mass removed by failure termination. Our theoretical analysis shows that this omission induces a systematic overestimation bias in goal-reaching values. Consequently, near-failure trajectories provide disproportionately strong supervision of success despite retaining little future occupancy. Unsafe actions can thereby be reinforced through catastrophic failure bootstrapping, leading to failed policy learning and unsustainable goal-reaching behaviours. To address this problem, we introduce two minimal yet strong corrections: mass-weighted InfoNCE corrects the overweighting of short surviving futures in critic learning, and a log-survival-mass score restores the missing survival mass in policy optimization. The resulting method, Safe Contrastive Reinforcement Learning (Safe-CRL), requires only the one-bit signal provided by failure termination to scale safe goal-conditioned policy learning. Across twelve failure-prone robot navigation and locomotion tasks, Safe-CRL consistently improves survival and substantially outperforms the Scaling-CRL baseline in goal-reaching performance. Additionally, deep Safe-CRL policies exhibit complex failure-avoidance behaviours. This study completes the CRL theory under failure termination and provides a scalable safe RL framework. The code is available via this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2608.26571 — 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 2026-10-17T15:34:49.910Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).