Learning When Not to Answer: a Ternary Reward Structure for Reinforcement Learning Based Question Answering.
Fréderic GodinAnjishnu KumarArpit MittalPublished in: NAACL-HLT (2) (2019)
Keyphrases
- reinforcement learning
- question answering
- learning algorithm
- question classification
- learning process
- natural language questions
- question answering systems
- answering questions
- information extraction
- answer extraction
- artificial intelligence
- answer validation
- machine learning
- reward function
- named entities
- natural language processing
- information retrieval