Loss and Reward Weighing for increased learning in Distributed Reinforcement Learning.
Martin HolenPer-Arne AndersenKristian Muri KnausgårdMorten GoodwinPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- learning algorithm
- learning process
- multi agent
- active learning
- distributed learning
- function approximation
- online learning
- learning capabilities
- reinforcement learning algorithms
- supervised learning
- eligibility traces
- learning agents
- learning problems
- learning tasks
- learning systems
- optimal policy
- knowledge acquisition
- distributed systems
- inductive inference
- state space
- learning agent
- learned knowledge
- prior knowledge
- optimal control
- semi supervised
- partially observable
- robot control
- dynamic programming
- reinforcement learning methods
- neural network