Reinforcement Learning with Internal Reward for Multi-Agent Cooperation: A Theoretical Approach.
Fumito UwanoNaoki TatebeMasaya NakataKeiki TakadamaTim KovacsPublished in: EAI Endorsed Trans. Collab. Comput. (2016)
Keyphrases
- reinforcement learning
- multi agent cooperation
- multi agent
- function approximation
- multi agent systems
- state space
- reward function
- optimal policy
- temporal difference
- eligibility traces
- model free
- theoretical analysis
- markov decision processes
- learning algorithm
- total reward
- cooperative
- reinforcement learning algorithms
- machine learning
- transfer learning
- average reward
- transition model
- multi agent reinforcement learning
- partially observable environments
- robotic control
- game theory
- learning problems
- expected reward
- policy search
- markov decision problems
- reinforcement learning methods
- policy iteration
- partially observable
- learning capabilities
- supervised learning
- dynamic programming