Reinforcement Learning Techniques for Decentralized Self-adaptive Service Assembly.
Mauro CaporuscioMirko D'AngeloVincenzo GrassiRaffaela MirandolaPublished in: ESOCC (2016)
Keyphrases
- reinforcement learning
- multi agent
- service providers
- function approximation
- peer to peer
- fully distributed
- learning algorithm
- optimal policy
- distributed systems
- service quality
- management system
- state space
- model free
- reinforcement learning algorithms
- temporal difference
- service oriented architecture
- customer service
- markov decision processes
- neural network
- cooperative
- web services
- service discovery
- dynamic programming
- learning process
- mobile services
- mobile devices
- continuous state
- decentralized control
- multiagent reinforcement learning
- robotic control