Unsupervised Paraphrasing via Deep Reinforcement Learning.
A. B. SiddiqueSamet OymakVagelis HristidisPublished in: KDD (2020)
Keyphrases
- reinforcement learning
- supervised learning
- deep architectures
- unsupervised learning
- function approximation
- context sensitive
- deep learning
- learning algorithm
- model free
- machine learning
- reinforcement learning algorithms
- markov decision processes
- context specific
- state space
- unsupervised manner
- action space
- learning problems
- supervised classification
- data sets
- dynamic programming
- multi agent
- temporal difference
- question answering
- optimal policy
- data driven
- semi supervised
- policy gradient
- robotic control