Deep Reinforcement Learning with Graph-based State Representations.
Vikram WaradpandeDaniel KudenkoMegha KhoslaPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- function approximation
- state space
- multi agent
- model free
- robotic control
- machine learning
- graph theoretic
- supervised learning
- graph model
- neural network
- reinforcement learning algorithms
- learning problems
- markov decision processes
- learning algorithm
- temporal difference
- optimal control
- dynamical systems
- dynamic programming
- learning process
- partially observable
- data sets
- direct policy search