-
Ryohei NAKATSU
Article type: Preface
1997 Volume 12 Issue 6 Pages
809
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Shigenobu KOBAYASHI
Article type: Cover article
1997 Volume 12 Issue 6 Pages
810
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Kazuteru MIYAZAKI, Shigenobu KOBAYASHI
Article type: Special issue
1997 Volume 12 Issue 6 Pages
811-821
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Hajime KIMURA, Leslie Pack KAELBLING
Article type: Special issue
1997 Volume 12 Issue 6 Pages
822-830
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Minoru ASADA
Article type: Special issue
1997 Volume 12 Issue 6 Pages
831-836
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Akimoto KAMIYA
Article type: Special issue
1997 Volume 12 Issue 6 Pages
837-844
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Sadayoshi MIKAMI
Article type: Special issue
1997 Volume 12 Issue 6 Pages
845-849
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Yeun-Bae KIM
Article type: Corner article
1997 Volume 12 Issue 6 Pages
850-860
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Mamiko HATAYAMA, Kiyoshi AKAMA, Eiichi MIYAMOTO
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
861-869
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
Programs that deal with high level knowledge processing tend to get large and complicated, and it is difficult to begin by constructing a system that has all the required functions. Hence, it is necessary to progressively improve the system by gradually adding to its capabilities. The improvements are not always new functions but may be new reasoning methods or faster processing schemes. The problem is added cost. To completely rewrite the system or make major changes becomes very expensive. Therefore, such a construction method is important that permits the addition of capabilities or an improvement in efficiency by a small change of programs. However, various existing program construction methods and programming languages may not be suitable for this kind of successive program improvement. In this paper, we propose a system construction method based on equivalent transformation as a knowledge processing system construction method by successive improvement. We use a computational framework called "Rule Based Equivalent Transformation". This uses the formalism of translating a given problem into a declarative program and solving the problem by repeatedly applying equivalent transformation rules to the program. We constructed three natural language understanding systems for the domain of Japanese chess: first a prototype system "A" , and then two improved systems "B" and "C". We improved the efficiency of these systems by adding new rules, and the results clearly showed that, if rule based program transformation is used as a design method, the system's modularity is increased and, thus, system can be easily improved.
View full abstract
-
Tomohiro YAMAGUCHI, Yuji NOMURA, Yasuhiro TANAKA, Masahiko YACHIDA
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
870-880
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
The advantage of emergence is that various solutions are emerged. However, it takes large computation cost to emerge them since it requires the numbers of iterations of simulation. So we try to reduces the computation cost without losing variety of solutions by introducing the abstraction technique in Artificial Intelligence. This paper presents Isomorphism Based Reinforcement Learning by Isomorphism of Actions that reduces the learning cost without losing variety of solutions. Isomorphism is one of the concepts in Enumerative Combinatorics of mathematics. First we explain Isomorphism of Actions, then explain Isomorphism of Behaviors. Isomorphic behaviors those perform the same task can be obtained by transforming the learning result of the task by "the appropriate permutation". However, a priori knowledge that represents "the appropriate permutation" is not always given, so this paper uses the generate & test method that first generates the isomorphic learning results by transforming the learning result of reinforcement learning for a task by the combinatorial permutations, then tests to select two kinds of the behaviors performing the following tasks ; (1) isomorphic behaviors those perform the same task ; (2) discovery of the behaviors those are converged to the new task state. Since the acquired learning results are isomorphic each other, the merits of our method are those the time cost for generating various learning results is small and also the space cost is small too because it needs only the original learning result and the set of permutations for it. For these reasons, this method is significant for realizing the learning various behaviors for the dynamic environment or multiagent.
View full abstract
-
Shigeo MATSUBARA, Makoto YKOO
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
881-890
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
We have formalized a finite iterated game with change. The formalization extends a traditional framework, e.g., the prisoner's dilemma, by incorporating an influence on the payoff matrix at some future point by executing an action at the present time. This enables us to explain why a rational agent behaves cooperatively, because profit through cooperative behavior finally exceeds that through noncooperative behavior. This situation is seen in human interactions, where We often behave cooperatively, even though from a myopic view, cooperative behavior may not seem profitable. Next, we propose a new method for selecting an action in such a framework. It overcomes the drawbacks of previous methods, thus yielding cooperative behavior and consuming little time. A simple model is used to investigate when our method is able to select an appropriate action. Finally, we compare previous methods and our method by evaluating some example problems. Through experimental results, we show that our method is both efficient and simple.
View full abstract
-
Megumi ISHII, Shigeo KANEDA
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
891-900
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
In this paper, we propose a method to learn guidance strategies for data-entry systems. A feature of our method is that guidance strategies are learned by generalizing user's operation logs based on constraints among data items. The proposed method enables a data-entry system to acquire the strategies through its operation stage. As a result, no programming is needed for guidance and the system provides guidance which is adapted to a user.
View full abstract
-
Ken SATOH
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
901-910
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
In legal reasoning, judging from precedent cases is very important since precedent cases complements incompleteness of legal rules. Therefore, CBR systems such as HYPO have been studied. A main characteristics of such CBR systems in legal reasoning is that similarity between cases are dynamically changed according to contexts. For example, if the system is on the plaintiff side, it will retrieve similar cases which are in favor of the plaintiff side. On the other hand, if the system is on the defendant side, it will retrieve entirely different cases which are in favor of the defendant side based on another similarity measure which is different from the measure used for the sake of the plaintiff. This kind of similarity is naturally implemented in abductive logic programming since we can regard the dynamic similarity as abducible predicates and change these predicates by context. In this paper, we define a relevance criteria for dynamic similarity and show a translation method to abductive logic programming. Moreover, by using abducibles, we show how to construct an explanation why the current case is similar to the cited case.
View full abstract
-
Hiroshi G. OKUNO, Katashi NAGAO, Yoshiyuki KOSEKI, Hiroshi YASUHARA, K ...
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
911-920
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
JSAI Tenth Anniversary Commemorative CD-ROM, hereafter, the TAC CD-ROM, contains image files of all the technical papers and articles published for the past 10 years. It also contains title, author and keyword indeces and category classification. The files are in Hyper Text Markup Language and WWW viewers are used to browse the TAC CD-ROM. The TAC CD-ROM is not only a stand-alone and portable system, but is open to the Internet. This paper discusses and reports the design and implementation of the TAC CD-ROM.
View full abstract
-
Yasuhiko WATANABE, Yoshihiro OKADA, Tatsuhiko TSUNODA, Makoto NAGAO
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
921-927
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
It is important to use pattern information (e.g. TV newscasts) and textual information (e.g. newspapers) together. For this purpose, we describe a method for aligning articles in TV newscasts and newspapers. For aligning articles, we use nouns which appear in the articles of TV newscasts and newspapers which correspond with each other. The recall and the precision of the alignment process are 97% and 89%, respectively.
View full abstract
-
Hiroshi MASUDA, Masayuki NUMAO
Article type: Technical paper
1997 Volume 12 Issue 6 Pages
928-935
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
This paper proposes methods and a technique for converting incorrect or ambiguous orthographic projections to solid models. First two methods are proposed for handling incorrect draftings that have redundant lines or missing lines, and the third one is a technique to reduce the number of candidates from ambiguous draftings. The first method is for detecting inconsistent lines, which must not appear in orthographic projections. For generating solid models, the system searches for 2D lines that cause inconsistency, and automatically removes them from the original draftings. Such inconsistent lines can be efficiently detected using nogood environments maintained by an ATMS. The second method is for the recovery of missing lines. If the proper wireframe model cannot be obtained from orthographic projections, it is impossible to search for solid shapes that match the draftings. In such a case, candidate missing lines are added in draftings to satisfy the condition that all 2D lines correspond to at least one edge in a 3D model. Boolean equations are re-calculated based on the modified draftings, and are solved using ATMS. Finally, to reduce the number of candidate solids that are generated from ambiguous orthographic projections, we use a technique to introduce additional constraints of cells. Our experiments show that our methods and technique are efficient enough for practical use. We observe that an ATMS mechanism is adequate to solving these equations, because sufficient constraints are derived by examining relationships among 3D models and projections, and described as Boolean equations.
View full abstract
-
Hidetaka NAMBO, Haruhiko KIMURA, Masakazu MIURA, Shigeki HIROBAYASHI, ...
Article type: Corner article
1997 Volume 12 Issue 6 Pages
936-943
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
Production systems are an established method for encoding knowledge in an expert system. In production systems, it's an important problem how to deal with expensive productions. We had proposed three methods for dealing with expensive productions using the expression method of direct-product instansiation. The evaluations of these methods, however, have reported on only simulations. In this paper, we evaluate three methods we have proposed using knowledge base, and classify a type of expensive productions suited for each expression method of direct-product instansiation.
View full abstract
-
[in Japanese]
Article type: Other
1997 Volume 12 Issue 6 Pages
944
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
[in Japanese], [in Japanese], [in Japanese]
Article type: Corner article
1997 Volume 12 Issue 6 Pages
945-950
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
[in Japanese]
Article type: Corner article
1997 Volume 12 Issue 6 Pages
951
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
[in Japanese]
Article type: Corner article
1997 Volume 12 Issue 6 Pages
952
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Akihiro KASHIHARA
Article type: Corner article
1997 Volume 12 Issue 6 Pages
953
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Activity report
1997 Volume 12 Issue 6 Pages
954-957
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Activity report
1997 Volume 12 Issue 6 Pages
958
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Activity report
1997 Volume 12 Issue 6 Pages
959-960
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Activity report
1997 Volume 12 Issue 6 Pages
961-966
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Activity report
1997 Volume 12 Issue 6 Pages
967-968
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Activity report
1997 Volume 12 Issue 6 Pages
b001-b011
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Cover page
1997 Volume 12 Issue 6 Pages
c006
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Cover page
1997 Volume 12 Issue 6 Pages
c006_2
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Table of contents
1997 Volume 12 Issue 6 Pages
i006
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS
-
Article type: Table of contents
1997 Volume 12 Issue 6 Pages
i006_2
Published: November 01, 1997
Released on J-STAGE: September 29, 2020
MAGAZINE
FREE ACCESS