Earticle

Solving life and death problems is a key aspect of computer Baduk program. An approach to solve life and death problems, which is based on search, is presented. With this approach the program SharpSense is implemented. Experiment shows that SharpSense has the capability of professional Baduk players in solving enclosed problems. SharpSense detects two errors in All about Life and Death, which is the classic in this domain. And SharpSense can successfully solve some complex situations about life and death arising from real Baduk games.

In 1970 Albert Zobrist published his Ph.D. thesis describing the first computer program for Go. Judging by the progress made in the past 35years, it will likely be many decades before there is a program capable of playing at an expert level. This paper gives a highly personal overview of some results in computer and mathematical Go with references to the details. In some sections, I give an overview of my own work, including plans for the future.

In ths paper I shall suggest that a study of belief revision in Baduk could be an unexpectedly fertile ground of research for logicians, philosophers, conlputer scientists, and ganle theorists as well as for Baduk players. The study of belief revision turns out to be one of the most distinguished success stories in the recent history of logic. Starting from the legendary AGM postulates published in 1925, it has expanded its scope far beyond epistemic logic and philosophy of science up to computer science, artificial intelligence researches, and economics. Nevertheless, there are many serious open problems, which are largely philosophcal in character, in current theories of belief revision. By examining some typical situations of belief revision in Baduk, I propose to shed light on one of those philosophical problems, i.e., the principle of informational economy in AGM theory.

바둑은 몇 가지 매우 독특한 형태의 ‘빅’을 가지고 있다. 독립적으로 조건에 관계 없이 살아있는 경우보다 빅으로 사는 경우가 더 많은 것이다. 우리는 여기서 이미 발표된 빅의 형태들을 종류별로 구분하기 위한 대략의 그림을 제시 할 것이다. 여기에는 예를 들어 ‘젖힘 빅’과 같이 잡는 것이 가능하기는 하지만 그리 바람직하지는 않은 형태들도 포함되어 있다. 우리는 가능한 구성 요소들을 설명하고, 우리가 어떻게 이러한 구성 요소들을 사용하여 빅의 유형들을 체계적으로 정리하였는가를 예를 통해 제시할 것이다.

The game of Go [weiqi (Chinese/C)/ baduk (Korean/K)] has a number of distinct types of shared life [seki (Japanese/J)/ shuangl huo2 (C)/ bik (K)]. There are more kinds of shared life than there are of independent, unconditional, life. We provide an overview, and outline an approach towards a full classification of shared life, while referring to the relatively few known publications. We include positions in which capture is possible, but not desirable - e.g. hanezeki, and generalized nakade. We describe the components that are available, and indicate how we can systematically construct seki configurations using these components, and give some examples. We indicate what work is required to catalogue fully all seki configurations.