Spike-Based Reinforcement Learning in Continuous State and Action Space: When Policy Gradient Methods Fail.
Eleni VasilakiNicolas FrémauxRobert UrbanczikWalter SennWulfram GerstnerPublished in: PLoS Comput. Biol. (2009)
Keyphrases
- action space
- continuous state
- reinforcement learning
- function approximators
- state space
- markov decision processes
- real valued
- policy search
- policy gradient
- state action
- action selection
- control policies
- reinforcement learning methods
- stochastic processes
- continuous action
- continuous state spaces
- function approximation
- temporal difference
- model free
- markov decision process
- markov decision problems
- optimal control
- single agent
- control problems
- search space