A Multi-Agent Approach to POMDPs using off-Policy Reinforcement Learning and Genetic Algorithms.
Samuel O. ObadanZenghui WangPublished in: Int. J. Comput. (2020)
Keyphrases
- reinforcement learning
- multi agent
- genetic algorithm
- partially observable markov decision processes
- neural network
- learning classifier systems
- function approximation
- markov decision processes
- simulated annealing
- state space
- model free
- partially observable
- continuous state
- artificial neural networks
- learning algorithm
- reinforcement learning algorithms
- evolutionary computation
- multi agent environments
- multi objective
- evolutionary algorithm
- multi agent systems
- cooperative
- machine learning
- reinforcement learning methods
- learning agents
- temporal difference
- optimal policy
- action selection
- markov decision problems
- fuzzy logic
- multiagent systems
- single agent
- dynamic programming
- genetic programming
- policy search
- multi agent reinforcement learning
- control policies
- partially observable stochastic games
- actor critic
- partial observability
- rl algorithms
- metaheuristic
- autonomous agents