Approximate Temporal Difference Learning is a Gradient Descent for Reversible Policies.
Yann OllivierPublished in: CoRR (2018)
Keyphrases
- temporal difference learning
- approximate value iteration
- markov decision process
- function approximation
- fixed point
- reinforcement learning
- game playing
- evaluation function
- temporal difference
- optimal policy
- markov chain
- state space
- markov decision processes
- cost function
- reinforcement learning algorithms
- infinite horizon
- reward function
- objective function
- monte carlo
- function approximators