A comparison of eligibility trace and momentum on SARSA in continuous state-and action-space.
Barry D. NicholsPublished in: CEEC (2017)
Keyphrases
- action space
- continuous state
- reinforcement learning
- state space
- single agent
- action selection
- markov decision processes
- function approximators
- real valued
- control policies
- policy search
- reinforcement learning algorithms
- multi agent
- stochastic processes
- finite state
- state action
- function approximation
- multiple agents
- continuous state spaces
- dynamic environments
- continuous action
- robot navigation
- temporal difference
- dynamic programming
- policy iteration
- model free
- markov decision process
- decision making
- decision problems
- partially observable
- machine learning
- planning problems
- dynamical systems
- heuristic search
- optimal policy
- markov chain