Genetic Algorithms Digest Wednesday, April 5 1991 Volume 5 : Issue 8 - Send submissions to GA-List@AIC.NRL.NAVY.MIL - Send administrative requests to GA-List-Request@AIC.NRL.NAVY.MIL Today's Topics: *** Special Issue *** ICGA-91 Status Report ****************************************************************************** CALENDAR OF GA-RELATED ACTIVITIES: (with GA-List issue reference) AISB 91, Leeds, UK (v5n5) Apr 16-19, 1991 4th Intl. Conference on Genetic Algorithms (v4n17) Jul 13-16, 1991 AAAI 91, National Conference on AI, Anaheim, CA Jul 14-19, 1991 EUROPEAN SUMMER SCHOOL on MACHINE LEARNING (v5n7) Jul 22-31, 1991 IJCAI 91, International Joint Conference on AI, Sydney, AU Aug 25-30, 1991 (Send announcements of other activities to GA-List@aic.nrl.navy.mil) ****************************************************************************** ------------------------------------------------------------------------------ From: booker@starbase.MITRE.ORG (Lashon Booker) Date: Thu, 4 Apr 91 16:13:00 EST Subject: ICGA-91 Status Report Dear Colleagues, The final decisions about ICGA-91 papers have been made, so this is a good time to give everyone a summary of where we stand. A total of 159 papers were submitted to ICGA-91 and we accepted 78 (~ 49%). This compares with the approximately 95 papers submitted to ICGA-89, of which 60 were accepted. The relatively lower acceptance rate this year can be attributed to several factors. The most important factor was that the committee was less inclined than before to accept papers describing very preliminary work or papers requiring major revisions. Once again, our conference will have significant international participation. Of the 78 papers accepted, 24 had a non-U.S. first author (~ 30%). For those who care about such things, here's a count of the number of papers selected based on the first author's nationality: Belgium 3 Canada 2 France 2 Germany 4 Israel 2 Italy 2 Japan 2 Mexico 1 U.K. 6 U.S. 54 It is especially pleasing that papers representing the Evolutionsstrategie point of view will be presented at ICGA-91. We hope this helps to promote a more extensive dialog between the GA and ES communities. The ICGA-91 conference format will be significantly different from, and we believe better than, the format of previous conferences. One of the frequent complaints many of us heard after ICGA-89 was that there was not enough opportunity for meaningful discussion and interaction. This problem threatened to become worse for ICGA-91 given a growth in attendance and an increase in the number of new ideas and results (especially in GA theory) that we all need to talk about. Consequently, we decided to deemphasize the "lecture mode" constraints of plenary sessions in favor of promoting more lively interactions in small groups. We hope to achieve this objective with a format emphasizing poster sessions and informal workshops. The default forum for presenting papers at ICGA-91 will be poster sessions. There will be a 90 minute poster session scheduled before each of three lunch breaks (see tentative schedule below). Since the lunch area will be adjacent to the poster area, there is the potential for up to two and a half hours of small group discussions/debates per poster session. This should promote considerable interaction between authors and those interested in their work. Moreover it promises to be a more relaxing and enjoyable way to assimilate a large amount of information. We are considering at least one afternoon poster session to help minimize the number of papers presented in the same session. The afternoon time slot will also allow discussion to continue beyond the allotted time. The informal workshops are another idea for increasing the amount of discussion at ICGA-91. Time has been set aside for special interest groups to discuss selected topics in depth, present preliminary results, exchange notes on work in progress, etc. The workshops also provide a forum where those whose papers were not accepted for ICGA-91 can present their ideas to the rest of the community. The workshops are intended to be informal but not completely spontaneous. Several workshop topics and chairpersons have already been identified. We hope that more will be organized soon. A separate announcement containing details about ICGA-91 workshops will be posted on GA-List. The committee selected 15 papers to be presented in plenary sessions. Papers were chosen for plenary presentation for any of several reasons: they were representative of papers on a given topic, they were thought to be of special interest to a large segment of the community, they were deemed especially original in some respect, etc. Our goal in making these selections was to provide an overview of current research activities that might help "prime the pump" for the poster session and workshop discussions. It is especially important to note that the plenary/poster distinction says nothing about the relative quality of the papers. Excellent work has deliberately been allocated to *both* the poster and plenary sessions. Here is a *TENTATIVE* layout of the conference program: Sat Sun Mon Tues Wed AM 1: Tutorials Plenary 1 Plenary 3 Plenary 5 Workshops AM 2: Tutorials Poster 1 Poster 2 Poster 3 (if needed) Lunch Lunch Lunch Lunch PM 1: Invited Talks Plenary 2 Plenary 4 Panel PM 2: Invited Talks Workshops [Posters or Bus. Mtg. Workshops] Evening: Social Mixer Workshops Banquet Workshops (if needed) Let me emphasize that this is *VERY PRELIMINARY AND LIKELY TO CHANGE*, but it makes the point about what we're trying to achieve. Note that on Saturday there will be tutorials and invited talks. Three tutorials will be offered this year: A 3 hr introduction to the field, given concurrently with two 90 minute tutorials - one on advanced GA theory and one on modern biology as it relates to GAs. This year's invited talks illustrate the breadth of issues related to GAs. John Holland will talk about Echo, an ecological model he devised to study the properties of distributed systems with adaptive components. Marcus Feldman, the renowned mathematical geneticist, will discuss GAs as they relate to *real* mathematical genetics. John Miller of the Santa Fe Institute will talk about exciting developments in economic theory and how GA/Classifier ideas can have a major impact on future directions. William Wimsatt, a philsopher of biology at the University of Chicago will be discussing "Developmental constraints on evolution." In summary, ICGA-91 promises to be an exciting event! I hope to see you there! Lashon P.S. Questions/comments about the ICGA-91 program can be directed to me (booker@mitre.org) or to Rik Belew (rik@cs.ucsd.edu) -------------------------------- End of Genetic Algorithms Digest ********************************