Multiagent Value Iteration Algorithms in Dynamic Programming and Reinforcement Learning.
Dimitri P. BertsekasPublished in: CoRR (2020)
Keyphrases
- dynamic programming
- reinforcement learning
- markov decision processes
- multi agent
- state space
- learning algorithm
- policy iteration
- partially observable markov decision processes
- orders of magnitude
- optimal policy
- heuristic search
- computationally efficient
- optimization problems
- computational complexity
- cooperative
- machine learning
- autonomous agents
- computational cost
- model free
- markov decision process
- stochastic games
- stochastic approximation