Unsupervised Paraphrasing via Deep Reinforcement Learning.
A. B. SiddiqueSamet OymakVagelis HristidisPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- supervised learning
- unsupervised learning
- deep architectures
- function approximation
- state space
- machine learning
- question answering
- semi supervised
- deep learning
- data driven
- multi agent
- function approximators
- context sensitive
- robotic control
- unsupervised manner
- completely unsupervised
- robot control
- markov decision processes
- learning process
- learning algorithm
- weakly supervised
- learning capabilities
- temporal difference
- context specific
- genetic algorithm
- information retrieval
- neural network