On the Geometry of Reinforcement Learning in Continuous State and Action Spaces.
Saket TiwariOmer GottesmanGeorge Dimitri KonidarisPublished in: CoRR (2023)
Keyphrases
- continuous state and action spaces
- reinforcement learning
- continuous state
- action selection
- machine learning
- function approximators
- reinforcement learning algorithms
- temporal difference
- markov decision processes
- robot navigation
- function approximation
- partially observable
- hidden state
- optimal control
- model free
- partially observable markov decision processes
- action space
- continuous action
- dynamic programming
- multi agent
- solving complex
- state space
- policy search