The cost of passing -- using deep learning AIs to expand our understanding of the ancient game of Go

TL;DR

This paper introduces a new numerical measure called the 'cost of passing' derived from deep learning AI engines to analyze and understand the game of Go more deeply, providing insights into move urgency and game features.

Contribution

It develops a novel, context-sensitive numerical tool for evaluating Go moves and recognizing game features based on AI engine outputs.

Findings

The 'cost of passing' effectively measures move urgency.

The measure reveals new insights into game dynamics.

Applications include improved move analysis and game understanding.

Abstract

AI engines utilizing deep learning neural networks provide excellent tools for analyzing traditional board games. Here we are interested in gaining new insights into the ancient game of Go. For that purpose, we need to define new numerical measures based on the raw output of the engines. In this paper, we develop a numerical tool for automated move-by-move performance evaluation in a context-sensitive manner and for recognizing game features. We measure the urgency of a move by the cost of passing, which is the score value difference between the current configuration of stones and after a hypothetical pass in the same board position. Here we investigate the properties of this measure and describe some applications.