Using Reinforcement Learning for Multi-policy Optimization in Decentralized Autonomic Systems - An Experimental Evaluation.
Ivana DusparicVinny CahillPublished in: ATC (2009)
Keyphrases
- autonomic systems
- experimental evaluation
- reinforcement learning
- autonomic computing systems
- optimal policy
- policy search
- multi agent
- optimization problems
- markov decision process
- optimization algorithm
- cooperative
- software systems
- reward function
- partially observable
- action selection
- partially observable environments
- state space
- policy gradient
- function approximation
- machine learning
- reinforcement learning algorithms
- action space
- state and action spaces
- reinforcement learning problems
- high level
- actor critic
- learning algorithm
- case study
- decentralized control
- continuous state spaces
- policy evaluation
- markov decision problems
- control policy
- software engineering
- temporal difference
- distributed systems