Swarm Reinforcement Learning using DEEPSO-Q with Advantage for Operational Planning of Energy Plants.
Kenjiro TakahashiYoshikazu FukuyamaPublished in: ICAIIC (2020)
Keyphrases
- real robot
- reinforcement learning
- macro actions
- function approximation
- energy consumption
- action selection
- planning problems
- state space
- energy minimization
- reward shaping
- decision making
- deterministic domains
- partially observable
- motion planning
- decision support
- markov decision problems
- reinforcement learning algorithms
- cooperative
- swarm intelligence
- stochastic domains
- goal oriented
- ai planning
- heuristic search
- model free
- markov decision processes
- optimal policy
- classical planning
- planning systems
- reinforcement learning methods
- planning domains
- reinforcement learning problems
- learning classifier systems
- particle swarm optimization
- learning process