课程: AI Algorithms for Game Design with Python

免费学习该课程！

今天就开通帐号，24,700 门业界名师课程任您挑！

Transposition tables

Transposition tables - Python教程

课程: AI Algorithms for Game Design with Python

开始一个月试用 购买团队方案

Transposition tables

“

- [Instructor] Another useful improvement for Minimax is a transposition table. It's simply a cache of previously seen states. This is sometimes known as memoization. An entry in this table contains some representation of the state and the resulting utility value. So when a node function is called, the first thing to do is to look up the state in the table. If it's found, we may confidently use the utility value from the table and return, and if it's not found, we may proceed with the usual recursion of Minimax. So for example, this is what a transposition table would look like for chess. It's a table where each entry contains some representation of a state and the previously found utility value for that state. It makes sense to keep highly repeating states in that table. And the lookup function must work fast because as you may imagine, this table can get crowded very fast. This may sound like a very good optimization, and it is, but it's not perfect. It has some complications. The…

内容