Zero-Shot Reward Specification via Grounded Natural Language.
Parsa MahmoudiehDeepak PathakTrevor DarrellPublished in: ICML (2022)
Keyphrases
- natural language
- formal languages
- formal language
- reinforcement learning
- natural language generation
- natural language interface
- knowledge representation
- specification language
- semantic representation
- neural network
- natural language processing
- semantic analysis
- formal specification
- language processing
- semantic interpretation
- reward function
- computer vision
- machine learning
- natural language text
- natural language understanding
- information extraction
- expert systems
- formal verification
- high level