- Keywords: ranking, search engine, reinforcement learning
- Abstract: Ranking items in large-scale item search engines such as Amazon and Taobao is a typical multi-step decision-making problem. Due to the interactive nature between the human user and the search engine, reinforcement learning is a natural solution to this problem. In this project, we use Virtual-Taobao as the environment and some effective methods to solve this problem. Experimental results show that TD3 performs best on this problem.