AutoSafe is a policy architecture that integrates structured safety monitoring for continuous safe online RL on continuous-control tasks and a physical cart-pole.
That is, there exist b(s)∈Randg(s)∈Rm such that ˆ∆(s,a) =b(s) +g(s) ⊤a+εlin(s,a),|εlin(s,a)|≤ϵlin(s)(22) for all actionsaon this segment
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Safe Online Learning via Smooth Safety-Structured Policy Composition
AutoSafe is a policy architecture that integrates structured safety monitoring for continuous safe online RL on continuous-control tasks and a physical cart-pole.