AI RESEARCH

Positive-Only Drifting Policy Optimization

arXiv CS.LG

ArXi:2604.16519v1 Announce Type: new In the field of online reinforcement learning (RL), traditional Gaussian policies and flow-based methods are often constrained by their unimodal expressiveness, complex gradient clipping, or stringent trust-region requirements. Moreover, they all rely on post-hoc penalization of negative samples to correct erroneous actions. This paper