Fast Adaptation of Deep Reinforcement Learning-Based Navigation Skills to Human Preference.
Jinyoung ChoiChristopher R. DanceJung-Eun KimKyungsik ParkJaehun HanJoonho SeoMinsu KimPublished in: ICRA (2020)
Keyphrases
- reinforcement learning
- human experts
- multi agent
- adaptation process
- function approximation
- high level knowledge
- personality traits
- learning capabilities
- human interaction
- markov decision processes
- state space
- indoor environments
- human subjects
- model free
- temporal difference
- function approximators
- user preferences
- machine learning