July 11, 1993 Washington, DC Participants: Pat Hayes, Danny Bobrow, Randy Davis, Barbara Grosz, Norm Nielsen, Joe Bates, Paul Cohen, Tom Dean, Johan de Kleer, Bob Engelmore, Ed Feigenbaum, Richard Fikes, Ken Ford, Mark Fox, Peter Friedland, Barbara Hayes-Roth, Jim Hendler, Elaine Kant, Phil Klahr, Benjamin Kuipers, Ramesh Patil, Candy Sidner, Bill Swartout, Katia Sycara, Beverly Woolf, Carol Hamilton Pat Hayes called the meeting to order with the introduction of the newly elected officer and councilors of AAAI. Randy Davis has been elected to a two-year term as President-Elect. Tom Dean, Bob Engelmore, Peter Friedland, and Ramesh Patil have all been elected to three-year terms as AAAI councilors. Hayes gave a special thanks to retiring councilors Tom Dietterich, Richard Fikes, Mark Fox, and Barbara Hayes-Roth for their generous donations of time and energy over the past three years. Hayes also presented Danny Bobrow with a special plaque, noting his many years of service to AAAI.

President Pat Hayes opened the meeting by welcoming the four newly elected members of the Executive Council who are Johan de Kleer, Benjamin Kuipers, Paul Rosenbloom, and Beverly Woolf. He also thanked the four retiring councilors-Ken Forbus, Howie Shrobe, Bill Swartout, and Marty Tenenbaum-for their three years of service on the Council and their continuing service to AAAI. Standing Cqmmittee Reports Reports were presented by the finance, workshop grants, fellows, conference, publications, symposium, and scholarship committees. Finance Committee: Secretary-Treasurer Bruce Buchanan presented the financial report. Due to the decline in revenue for the association and poor interest rates, the financial outlook is not as positive as in past years.

Hojjat Adeli, coauthor of "A Novel Approach to Expert Systems for the Design of Large Structures, " is currently a professor of civil engineering at The Ohio State University, 470 Hitchcock Hall, 2070 Neil Avenue, Columbus, Ohio 43210. He received his Ph.D. from Stanford University in 1976 and is editor-inchief of the International Journal of Microcomputers in Civil Engineering. A contributor to 20 journals, he is the author or editor of over 160 publications in various fields of computer-aided engineering and is the editor of the forthcoming book series Knowledge Engineering, to be published by McGraw-Hill The first two volumes are scheduled for publication in mid-1989. Dean Allemang, coauthor of "Connectionism and Information Processing Abstractions: The Message Still Counts More Than the Medium," is a graduate research fellow at the Laboratory for Artificial Intelligence Research in the Department of Computer and Information Science at The Ohio State University, Columbus, Ohio 43210. He is currently writing his Ph.D. dissertation on a devicebased understanding of software.

James Peters, coauthor of "A Knowledge-Based Model of Audit Risk," is an assistant professor in the Department of Accounting, College of Business Administration, University of Oregon. Hans Berliner, author of the Hitech Computer Chess report, is a senior scientist in the Department of Computer Science, Carnegie-Mellon University, Pittsburgh, Pennsylvania 15213. R. Peter Bonasso, author of "An Assessment of What AI Can Do for Battle Management--A Report of the First AAAI Workshop on AI Applications to Battle Management" is the department head of the Artificial Intelligence Technical Center in The MITRE Corporation, Washington 01 Operations division, 7525 Colshire Drive, Mclean, VA 22102. His research interests include commonsense reasoning and qualitative processes with a view toward applications to military systems. Vasant Dhar, coauthor of "A Knowledge-Based Model of Audit Risk," is an associate professor in the Department of Information Systems, New York University.

At AAAI-91, AAAI conducted an experiment, running short (5 to 15 minute) videotapes of research projects and lectures, in the conference registration area. Conference attendees were pleased with the experiment, so AAAI would like to expand the number of tapes it has available for showing at AAAI-92 in San Jose, California (12-17 July). The conference is an excellent opportunity for your lab's research efforts to be shown to a large portion of the AI community. AAAI is looking for short tapes that can be run in parallel on several screens. Please do not send tapes of a particular project or lecture but, rather, tapes that present broad descriptions of different programs, projects, and systems that are used within your lab.

AAAI announces the continuation of its scholarship travel program for students who want to attend the National Conference on Artificial Intelligence in San Jose, California, 12-17 July 1992. Undergraduate or graduate students enrolled in a full-time degree program at any college or university are eligible to serve as student volunteers during AAAI-92, to be held at the San Jose Convention Center in San Jose, California, 12-17 July. In exchange for assisting AAAI staff members during your volunteer shift, you will receive complimentary conference registration, a copy of the AAAI-92 proceedings, and a special AAAI-92 T-shirt. If you are interested in assisting AAAI at the national conference, please contact AAAI at volunteer @aaai.org. All inquiries should include your name, address, telephone, advisor's name, and email address.

In this video course, you will learn to tap some of the powerful abilities of R. R is one of the leading packages in the world with a vast number of active users and, as a result, has a massive number of state-of-the-art libraries. You will master the basics and get comfortable with R, so you can then use its libraries to do the heavy-lifting. You'll begin by looking at high-performance computing in the classic, computationally intensive scenario: finding prime numbers.Then you'll learn how to use R, before moving on to using C, which is far faster. Next you will use the power of parallel, though that varies from problem to problem since some are more suitable for parallelization. Then you will look at some powerful options available on R where you don't just produce a static result but instead respond to user selections.

Column n The Educational Advances in Artificial Intelligence column discusses and shares innovative educational approaches that teach or leverage AI and its many subfields at all levels of education (K-12, undergraduate, and graduate levels). Have you ever been surprised by poor class performance on a midterm question, and wondered why you were met with silence each time you asked "Any questions?" during the lecture on that topic? Do your students sometimes feel like they understood everything that was said in lecture, only to go home, start the homework, and immediately get stuck? Do you find that you only really learn something when you have to explain it to others? Peer instruction is an active learning pedagogy that addresses these challenges and opportunities.

A common problem in machine learning is to rank a set of n items based on pairwise comparisons. Here ranking refers to partitioning the items into sets of pre-specified sizes according to their scores, which includes identification of the top-k items as the most prominent special case. The score of a given item is defined as the probability that it beats a randomly chosen other item. Finding an exact ranking typically requires a prohibitively large number of comparisons, but in practice, approximate rankings are often adequate. Accordingly, we study the problem of finding approximate rankings from pairwise comparisons. We analyze an active ranking algorithm that counts the number of comparisons won, and decides whether to stop or which pair of items to compare next, based on confidence intervals computed from the data collected in previous steps. We show that this algorithm succeeds in recovering approximate rankings using a number of comparisons that is close to optimal up to logarithmic factors. We also present numerical results, showing that in practice, approximation can drastically reduce the number of comparisons required to estimate a ranking.

Over the past two decades, the notion of implicit bias has come to serve as an important component in our understanding of discrimination in activities such as hiring, promotion, and school admissions. Research on implicit bias posits that when people evaluate others -- for example, in a hiring context -- their unconscious biases about membership in particular groups can have an effect on their decision-making, even when they have no deliberate intention to discriminate against members of these groups. A growing body of experimental work has pointed to the effect that implicit bias can have in producing adverse outcomes. Here we propose a theoretical model for studying the effects of implicit bias on selection decisions, and a way of analyzing possible procedural remedies for implicit bias within this model. A canonical situation represented by our model is a hiring setting: a recruiting committee is trying to choose a set of finalists to interview among the applicants for a job, evaluating these applicants based on their future potential, but their estimates of potential are skewed by implicit bias against members of one group. In this model, we show that measures such as the Rooney Rule, a requirement that at least one of the finalists be chosen from the affected group, can not only improve the representation of this affected group, but also lead to higher payoffs in absolute terms for the organization performing the recruiting. However, identifying the conditions under which such measures can lead to improved payoffs involves subtle trade-offs between the extent of the bias and the underlying distribution of applicant characteristics, leading to novel theoretical questions about order statistics in the presence of probabilistic side information.