Efficient Ant Reinforcement Learning Using Replacing Eligibility Traces.
SeungGwan LeeSeokMi HongPublished in: ICAISC (2006)
Keyphrases
- eligibility traces
- reinforcement learning
- reinforcement learning algorithms
- policy evaluation
- reinforcement learning methods
- function approximation
- model free
- dynamic programming
- markov decision processes
- particle swarm optimization
- temporal difference
- learning speed
- least squares
- multi agent
- density estimation
- monte carlo
- optimal policy
- control problems
- learning algorithm