Scholarlink

NoGo is a version of Go where stones are never removed from the board, once played. Strong computer players have been created for NoGo. However, the game properties and optimal play strategies are not well studied. We introduce CGTSolver, a search algorithm that applies concepts from combinatorial game theory (CGT) in order to solve Linear NoGo. We develop several decomposition strategies and simplification rules for this game. Our results show that CGTSolver is much more efficient than previous solvers, and as the board size increases, the performance gap widens. With this new approach we solved all NoGo boards up to \(1\times 39\) —twelve boards more than in previous work.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Solving Linear NoGo with Combinatorial Game Theory

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Haoyu Du,
  • Martin Müller

摘要

NoGo is a version of Go where stones are never removed from the board, once played. Strong computer players have been created for NoGo. However, the game properties and optimal play strategies are not well studied. We introduce CGTSolver, a search algorithm that applies concepts from combinatorial game theory (CGT) in order to solve Linear NoGo. We develop several decomposition strategies and simplification rules for this game. Our results show that CGTSolver is much more efficient than previous solvers, and as the board size increases, the performance gap widens. With this new approach we solved all NoGo boards up to \(1\times 39\) —twelve boards more than in previous work.