Researchers have developed Sanko, a new parallel depth-first proof-number solver designed to tackle the complexities of solving Go under positional superko (PSK) rules. PSK, which forbids recreating earlier whole-board positions, presents significant challenges for traditional solving tools like transposition tables, static life-and-death recognition, and safe move pruning. Sanko addresses these issues with a novel proof-transfer theorem, enabling it to efficiently solve 5x5 Go in a fraction of the time previously required and establish new game-theoretic values for numerous board configurations under PSK rules. The team has also released the solver's source code and all run data to ensure reproducibility of their findings. AI
IMPACT Introduces a novel algorithmic approach for solving complex games, potentially influencing AI research in game theory and combinatorial search.
RANK_REASON Academic paper detailing a new algorithm and its application to a specific problem domain. [lever_c_demoted from research: ic=1 ai=1.0]
- 4x7
- 5x5 Go
- arXiv
- basic-ko rule set
- Positional superko
- proof-transfer theorem
- safe move pruning
- Sanko
- static life-and-death recognition
- transposition tables
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →