Reinforcement Learning for solving Russian Puzzle

于承鑫 游柏晏、曾懷恩
國立中正大學
yu910703@yahoo.com.tw
「領域:物理數學與數值方法」
「對象:大一」
「類型一:教案」
「類型二:實驗-學生成果」
摘要內容
To efficiently find the solution to the Russian puzzle, we introduce machine learning to assist us. Starting from the Random method of mechanical thinking, and then introducing the Score matrix to give it the ability to select puzzles intelligently. Finally through the Q-Table method, the computer have memory can record successes and failures, thus allowing it to avoid repeated failures.

課程內容
留言板
請登入會員後查看