Emergent Policy Discovery for Visual Reinforcement Learning Through Tangled Program Graphs: A Tutorial.
Stephen KellyRobert J. SmithMalcolm I. HeywoodPublished in: GPTP (2018)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- visual information
- markov decision process
- action selection
- partially observable environments
- state and action spaces
- partially observable
- reinforcement learning algorithms
- reward function
- visual features
- graph theory
- low level
- graph matching
- control policies
- state action
- control policy
- function approximators
- action space
- policy iteration
- knowledge discovery
- state space
- graph theoretic
- continuous state spaces
- reinforcement learning problems
- approximate dynamic programming
- learning algorithm
- function approximation
- community discovery
- rl algorithms
- continuous state
- actor critic
- high level
- data mining
- policy gradient
- multi agent
- model free
- infinite horizon
- graph representation
- graph mining
- graph structure
- pattern discovery