Towards the Use of Deep Reinforcement Learning with Global Policy For Query-based Extractive Summarisation.
Diego MolláPublished in: CoRR (2017)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- database
- query biased
- response time
- partially observable environments
- markov decision process
- query processing
- action selection
- function approximation
- user queries
- database queries
- partially observable
- uniform manner
- query expansion
- data sources
- reinforcement learning problems
- dynamic programming
- markov decision problems
- function approximators
- machine learning
- state and action spaces
- agent receives
- policy gradient
- control policies
- action space
- policy iteration
- keywords
- learning algorithm
- query specific
- reinforcement learning algorithms
- multi document summarization
- text summarization
- markov decision processes
- natural language processing