Constrained Flow Policy Updates: A Generalized Schr\"odinger Bridge View
This work builds on the density-free kinetic-energy regularizer of FLAC, a recent reward-only method, and proposes Reparameterized Augmented-Lagrangian Flow Actor with Least Energy (RAFALE), an off-policy actor-critic method for safe RL.
Bo-Yan Li, Matthew Kim, Sylvia L. Herbert
· 0 citations