Deep Reinforcement Learning with Distributional Semantic Rewards for Abstractive Summarization.
Siyao LiDeren LeiPengda QinWilliam Yang WangPublished in: EMNLP/IJCNLP (1) (2019)
Keyphrases
- reinforcement learning
- markov decision processes
- function approximation
- state space
- model free
- reinforcement learning algorithms
- semantic search
- temporal difference
- learning process
- neural network
- semantic network
- optimal policy
- semantic web
- domain specific
- co occurrence
- natural language
- sports video
- learning algorithm
- reward shaping
- multiarmed bandit
- semantic features
- optimal control
- transfer learning
- domain ontology
- dynamic programming
- high level
- semantic web technologies
- reward function
- semantic content
- multi document summarization
- partially observable
- multi agent
- semantic description
- document summarization
- total reward
- machine learning