Modified action decoder using Bayesian reasoning for multi-agent deep reinforcement learning.
Wei DuShifei DingChenglong ZhangShuying DuPublished in: Int. J. Mach. Learn. Cybern. (2021)
Keyphrases
- bayesian reasoning
- reinforcement learning
- multi agent
- action selection
- function approximation
- action space
- state action
- single agent
- state space
- partially observable domains
- reward shaping
- reinforcement learning algorithms
- model free
- markov decision processes
- machine learning
- multi agent systems
- agent learns
- region matching
- reinforcement learning agents
- dynamic programming
- multi agent environments
- extensive form games
- fitted q iteration
- multiagent reinforcement learning
- learning agents
- function approximators
- pattern matching
- d objects
- object recognition
- multiscale