Scale-free adaptive planning for deterministic dynamics & discounted rewards.
Peter L. BartlettVictor GabillonJennifer HealeyMichal ValkoPublished in: ICML (2019)
Keyphrases
- scale free
- markov decision processes
- complex networks
- small world
- fully observable
- planning problems
- scale free networks
- fully connected
- power law
- reinforcement learning
- protein protein interaction networks
- small world properties
- optimal policy
- biological networks
- degree distribution
- dynamic programming
- community detection
- community structure
- power law distribution
- high precision