Markov Decision Process - Tree search

Posted on 2021-03-03 | In AI Planning and Decision Making

Words count in article: 60 | Reading time ≈ 1

Online search (approximate solution)

Reach a leaf node: you have a certain result of winning or losing.
If you have not reach the depth, you can add more nodes to the tree(expansion).

Trial in MCTs: selection + …
Trial in UCT: selection
select a to max Q and min N(n, a) : adaptivity

Partial Observed

Post author: HUANG Liu
Post link: https://huangliu0909.github.io/2021/03/03/AI_MDP_1/
Copyright Notice: All articles in this blog are licensed under CC BY-NC-SA 3.0 unless stating additionally.