Loss and Reward Weighing for increased learning in Distributed Reinforcement Learning

no code implementations • 25 Apr 2023 • Martin Holen, Per-Arne Andersen, Kristian Muri Knausgård, Morten Goodwin

This paper introduces two learning schemes for distributed agents in Reinforcement Learning (RL) environments, namely Reward-Weighted (R-Weighted) and Loss-Weighted (L-Weighted) gradient merger.

reinforcement-learning Reinforcement Learning (RL)