Deep Reinforcement Learning with a Combinatorial Action Space for Predicting and Tracking Popular Discussion Threads.
Ji HeMari OstendorfXiaodong HeJianshu ChenJianfeng GaoLihong LiLi DengPublished in: CoRR (2016)
Keyphrases
- action space
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- real valued
- continuous state
- particle filter
- action selection
- stochastic processes
- continuous state spaces
- markov decision process
- function approximation
- function approximators
- optimal policy
- dynamic programming
- state action
- information retrieval
- higher education
- active learning