Preprint
Oct 2026
Bridging Exploration and Loss of Plasticity in On-Policy Reinforcement Learning
This work reports multiple metrics which characterize the effect that the policy standard deviation $\sigma$ has on LoP, and suggests related explanations for the effectiveness of widely adopted LoP mitigation techniques.
A. Grieco, Evelyn D'Elia, Edoardo Caldarelli et al.
· 0 citations