Match Plan Generation in Web Search with Parameterized Action Reinforcement Learning.
Ziyan LuoLinfeng ZhaoWei ChengSihao ChenQi ChenHui XueHaidong WangChuanjie LiuMao YangLintao ZhangPublished in: WWW (2021)
Keyphrases
- plan generation
- web search
- reinforcement learning
- plan execution
- action selection
- plan recognition
- search engine
- action space
- partially observable domains
- temporal planning
- state space
- state action
- reward shaping
- web pages
- planning problems
- optimal control
- machine learning
- temporal constraints
- markov decision processes
- optimal policy
- temporal information
- initial state
- dynamic programming
- agent learns