AI RESEARCH
Accelerated Learning with Linear Temporal Logic using Differentiable Simulation
arXiv CS.LG
•
ArXi:2506.01167v2 Announce Type: replace Ensuring that reinforcement learning (RL) controllers satisfy safety and reliability constraints in real-world settings remains challenging: state-avoidance and constrained Marko decision processes often fail to capture trajectory-level requirements or induce overly conservative behavior. Formal specification languages such as linear temporal logic (LTL) offer correct-by-construction objectives, yet their rewards are typically sparse, and heuristic shaping can undermine correctness. We.