Pragmatic Feature Preferences: Learning Reward-Relevant Preferences from Human Input.
Andi PengYuying SunTianmin ShuDavid AbelPublished in: CoRR (2024)
Keyphrases
- preference learning
- learning process
- learning algorithm
- user preferences
- human behavior
- learning tasks
- learning systems
- human learning
- online learning
- learning environment
- learning problems
- multi agent
- soft constraints
- learning preferences
- learning agent
- language acquisition
- learning capabilities
- long run
- multi attribute
- human experts
- utility function
- unsupervised learning
- input data
- training data