Exploring with Sticky Mittens: Reinforcement Learning with Expert Interventions via Option Templates.
Souradeep DuttaKaustubh SridharOsbert BastaniEdgar DobribanJames WeimerInsup LeeJulia Parish-MorrisPublished in: CoRL (2022)
Keyphrases
- reinforcement learning
- function approximation
- learning process
- temporal difference
- human experts
- markov decision processes
- machine learning
- state space
- learning algorithm
- optimal policy
- robotic control
- domain experts
- policy search
- expert advice
- information retrieval
- multi agent
- model free
- reinforcement learning methods
- reinforcement learning algorithms
- optimal control
- option pricing
- temporal difference learning
- data sets
- expert knowledge
- case study
- transfer learning
- supervised learning