Quantum Policy Iteration via Amplitude Estimation and Grover Search - Towards Quantum Advantage for Reinforcement Learning.
Simon WiedemannDaniel HeinSteffen UdluftChristian B. MendlPublished in: Trans. Mach. Learn. Res. (2023)
Keyphrases
- policy iteration
- reinforcement learning
- markov decision processes
- model free
- optimal policy
- temporal difference
- approximate dynamic programming
- stochastic approximation
- policy evaluation
- sample path
- markov decision process
- least squares
- fixed point
- state space
- search algorithm
- actor critic
- learning algorithm
- infinite horizon
- finite state
- function approximation
- search space
- action space
- machine learning
- reinforcement learning algorithms
- average reward
- temporal difference learning
- convergence rate
- dynamic programming
- multi agent
- approximate policy iteration