Adrian Remonda, Jiajun Xi, Nicklas Hansen, Marcus Greiff, Thomas Lew, John Talbot, John Subosits, Xiaolong Wang · 2026 · arXiv
DOI: 10.48550/arXiv.2610.26757 · open-access PDF
Verstärkendes Lernen ermöglicht das Lernen von Hochleistungskontrollrichtlinien, aber seine Abhängigkeit von umfangreicher Interaktion beschränkt die Bereitstellung auf physischen Systemen, in denen die Datenerhebung teuer ist und Fehler unsicher sind. Wir führen Pre-Informed Adaptation (PIA) ein, ein offline-to-online-modellbasiertes RL-Framework, das strukturierte Simulations-Priors verwendet, um sich schnell anzupassen.
Machine translation (Meta M2M100); the English original follows.
Reinforcement learning enables learning high-performance control policies, but its reliance on extensive interaction limits deployment on physical systems where data collection is costly, and failures are unsafe. We introduce Prior-Informed Adaptation (PIA), an offline-to-online model-based RL framework that uses structured simulation priors for rapid adaptation. PIA pretrains a world model on heterogeneous low- and high-fidelity simulated trajectories, then adapts it with limited target-domain interaction using uncertainty-regularized planning. Rather than requiring an accurate simulator for
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26757, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange