Reinforcement Learning with Converging Goal Space and Binary Reward Function.
Wooseok RolWonseok JeonHamid BamshadHyunseok YangPublished in: CASE (2020)
Keyphrases
- reward function
- reinforcement learning
- reinforcement learning algorithms
- initially unknown
- markov decision processes
- state space
- transition model
- optimal policy
- policy search
- inverse reinforcement learning
- reward signal
- partially observable
- hierarchical reinforcement learning
- multiple agents
- action space
- function approximation
- transition probabilities
- markov decision process
- model free
- agent learns
- temporal difference
- state action
- machine learning
- state variables
- dynamic programming
- search space
- multi agent
- function approximators
- learning agent
- objective function
- learning algorithm