A Model-Free Approach to Meta-Level Control of Anytime Algorithms.
Justin SvegliatoPrakhar SharmaShlomo ZilbersteinPublished in: ICRA (2020)
Keyphrases
- model free
- anytime algorithms
- dynamic programming
- reinforcement learning
- data streams
- real time systems
- solution quality
- function approximation
- policy iteration
- markov decision problems
- reinforcement learning algorithms
- heuristic search algorithms
- optimal solution
- temporal difference
- simulated annealing
- neural network
- learning process