← Back to Benchmarks
simmediummanipulation-datametric · varies

Physics Informed Viscous Value Representations

Description

Offline goal-conditioned reinforcement learning (GCRL) learns goal-conditioned policies from static pre-collected datasets. However, accurate value estimation remains a challenge due to the limited coverage of the state-action space. Recent physics-informed approaches have sought to address this by imposing physical and geometric constraints on the value function through regularization defined over first-order partial differential equations (PDEs), such as the Eikonal equation. However, these fo

Source

http://arxiv.org/abs/2602.23280v1