AAAI 2008 Fall Symposia Call for Participation
The Association for the Advancement of Artificial Intelligence is pleased to present the 2008 Fall Symposium Series, to be held Friday through Sunday, November 7-9, at the Westin Arlington Gateway in Arlington, Virginia. The titles of the seven symposia are as follows:
- Adaptive Agents in Cultural Contexts
- AI in Eldercare: New Solutions to Old Problems
- Automated Scientific Discovery
- Biologically Inspired Cognitive Architectures
- Education Informatics: Steps Toward the International Internet Classroom
- Multimedia Information Extraction
- Naturally Inspired AI
The highlights of each symposium will be presented at a special plenary session. Notes will be prepared and distributed to participants in each symposium, but will not otherwise be available unless published as an AAAI Technical Report or edited collection.
Each symposium will have limited attendance. Participants will be expected to attend a single symposium throughout the symposium series. In addition to participants selected by the program committee of the symposia, a limited number of other interested parties will be allowed to register in each symposium on a first-come, first-served basis. To register, please fill out the registration form, and send it along with payment to:
2008 Fall Symposium Series
AAAI
445 Burgess Drive
Menlo Park, CA 94025-3442
Telephone: (650) 328-3123*
Fax: (650) 321-4457*
Email: fss08@aaai.org*
*Credit card orders only, please. Please note that there are security issues involved with the transmittal of credit card information over the internet. AAAI will not be held liable for any misuse of your credit card information during its transmittal to AAAI.
Online registration is available.
Tentative Program Schedule
(subject to change)
Friday, November 7
9:00 AM – 5:30 PM: Symposia sessions
6:00 PM – 7:00 PM: Reception
Saturday, November 8
9:00 AM – 5:30 PM: Symposia sessions
6:00 PM – 7:00 PM: Plenary session
Sunday, November 9
9:00 AM – 12:30 PM: Symposia sessions
Registration will be located in the Ballroom Foyer on the Second Level.
To obtain registration information, see the on-line registration form, consult the PDF form (for mail-in registrations) or write to:
- 2008 Fall Symposium Series
AAAI
445 Burgess Drive
Menlo Park, CA 94025-3442
Telephone: 650-28-3123*
Fax: 650-321-4457*
E-mail: fss08@aaai.org
Adaptive Agents in Cultural Contexts
Computational human behavior models, in extending a conventional information-processing approach, face two complex problems: adaptation and evolution of behavior, and the sociocultural specificity of cognition. These fields are vast, variegated, informed by disparate theoretical and technical disciplines, and interrelated. This symposium seeks to focus research by examining their intersection. In addition to informing academic research, applications include simulations and training for international commercial enterprise, non-governmental organizations, and military, as well as commercial games.
We aim to bring together communities of artificial intelligence, social science, and cognitive science researchers, with developers of games and simulations within both commercial and governmental sectors. To this end, presented papers will include the following:
- Case studies of simulation or game development that contain adaptive and/or cultural aspects
- Theoretical work in modeling cultural behavior, adaptive behavior, or their interrelation
- Analysis of human behavior in relevant domains
Additionally, Marcus Griffin, Sr. Social Science Advisor to the Human Terrain System and professor of anthropology at Christopher Newport University has been invited to speak about his recent experiences as a cultural anthropologist with the U.S. Army in its effort to implement the Human Terrain System.
Organizing Committee
Alex Davis, cochair (Stottler Henke), Jeremy Ludwig, cochair (Stottler Henke), David W. Aha (Naval Research Laboratory), Harold Hawkins (Office of Naval Research), Lewis Johnson (Tactical Language Training), Helen Altman Klein (Wright State University), Glenn Taylor (Soar Technology, Inc.), Michael van Lent (University of Southern California/Institute for Creative Technologies), Abbas K. Zaidi (George Mason University)
For more information about the symposium, see the supplementary symposium web site.
AI in Eldercare: New Solutions to Old Problems
There is a wide range of problems facing older adults as they age. Many represent old challenges to health care providers, including chronic illnesses like heart disease, diabetes, and hypertension, as well as deterioration of physical function, high risk of falling, strokes, memory problems, cognitive decline, and loneliness. At the same time, the population of older adults is growing dramatically, giving concern as to how these people will get the care they need.
AI technology offers the potential for innovative solutions, spanning such areas as sensing and sensory perception, computer vision, planning, reasoning, smart homes, robotics and human-robot interaction. We invite an interdisciplinary group with joint interests in addressing aging-related challenges. In addition to AI researchers, gerontologists, geriatric nurses and psychiatrists, rehabilitation therapists, social workers, counselors, epidemiologists, and those from other related professions and disciplines are invited to attend. We will provide a forum to share ideas, foster new collaborations, and investigate funding opportunities.
The symposium will focus on a variety of topics that address the physical, cognitive, and emotional challenges of aging:
- Smart homes
- Reminder systems
- Fall detection
- Automatic gait analysis
- Passive sensing for monitoring physical and/or cognitive condition
- Wearable sensors
- Aging assessment tools
- Robotics for eldercare
- Stroke rehabilitation
- Systems to provide emotional support
- Aging assistance for people with disabilities
- Ethical considerations of eldercare systems
- Evaluating eldercare systems
A combination of presentation and discussion styles is planned: Focused panel discussions including one offering the perspective of gerontology experts: talks describing established work followed by discussion periods; short talks describing emerging work followed by discussion periods; and breakout sessions with small groups discussing challenges and possible solutions; reports to the large group
Organizing Committee
Marge Skubic, chair (University of Missouri), Michael Anderson (University of Hartford), Susan Anderson (University of Connecticut), Tim Bickmore (Northeastern University), Cynthia Breazeal (Massachusetts Institute of Technology), Jesse Hoey (University of Dundee, Scotland), Stephen Intille (Massachusetts Institute of Technology), Ben Krose (University of Amsterdam), Alex Mihailidis (University of Toronto), Federico Pecora (Orebro University, Sweden), Rich Simpson (University of Pittsburgh), Holly Yanco (University of Massachusetts Lowell), Howard Wactlar (Carnegie Mellon University)
For more information about the symposium see the supplementary symposium web site.
Automated Scientific Discovery
There is a long and fascinating history of humankind's endeavor to explain and, with the advent of AI, ultimately mechanize the overarching processes that lead to scientific discoveries. This quest dates back to Aristotle's account of human deductive reasoning (the theory of the syllogism, developed to model the discoveries of Euclid), and continues through modern AI, which, through impressive systems like LT, Bacon, GT, Eurisko, and Graffiti (and many theorem provers, model finders, and computational frameworks for machine-assisted reasoning), has placed some degree of such automation within reach. Over the past 60 years, starting with AI's inaugural conference, systems such as these have automated aspects of scientific discovery. Machines have generated novel and interesting conjectures (some of which have spawned new scientific research areas), and increasingly efficient techniques have been invented to prove or refute them.
Nevertheless, the sobering fact remains that such advances fall far short of approaching the creativity and innovation of even amateur scientists. We believe that AI is ripe for revolutionary progress in automated and semi-automated scientific discovery, in no small part because the field now has on hand systems that mark advances in various parts of discovery — parts that, when interconnected, may make for some very exciting new systems. We also believe that dialogue between researchers behind these systems could lead to a new generation of powerful AI discovery systems.
This symposium will survey the state of the art in systems that cover some aspects of the entire process of scientific discovery (including, for example, representation, exploration, conjecture generation, validation, and publishing/reporting). Of particular interest is how the current technologies can fit together to form an environment by which the human reasoner’s vision and reach can be augmented, and what goals should be set in order to move closer to the complete mechanization of general scientific discovery—or at least closer to a time when machines can truly operate as intelligent assistants in the search for new discoveries.
Organizing Committee
Andrew Shilliday, cochair (Rensselaer Polytechnic Institute), Selmer Bringsjord, cochair (Rensselaer Polytechnic Institute) (selmer@rpi.edu), Alan Bundy (University of Edinburgh), Simon Colton (Imperial College London), Doug Lenat (Cycorp)
For more information about the symposium see the supplementary symposium web site.
Biologically Inspired Cognitive Architectures
We invite you to participate in the first International Symposium focused on the challenge of creating a computational equivalent of the human mind in its higher cognitive abilities. This fundamental scientific task calls for the design and experimental study of biologically inspired cognitive architectures (BICA), including those capable of human-like cognitive growth. The primary objective of the symposium is to showcase recent modeling and rapid prototyping experience aimed at building architectures of cognitive agents that have been inspired by the human brain and, in a definite sense, operate like the human mind. At the same time, theoretical discussion of the underlying mechanisms is equally encouraged. Topics of the program include the following:
- Cognitive architectures describing the human brain-mind at a computational level
- Models of human-like learning, meta-learning, the self and self-awareness
- Educational practice of self-regulated learning as a source of inspirations for BICA
- Natural language acquisition and NLP-based learning: identifying the 'critical mass' of capabilities that enables cognitive growth
- Developing systems of values and human-level emotional intelligence in artifacts
- Bridging the gap between biological and computational systems in robustness, flexibility, integration and human-like learning abilities
- Vital biological constraints informed by neuroscience and their computational leverage in embodied cognition
- Large-scale computational BICA projects and their future real-world applications
Other key speakers include Michael Anderson, Michael Arbib, Giorgio Ascoli, Roger Azevedo, Susan Epstein, Paul Kogut, Chris Lebiere, Marge McShane, Dan Oblinger, Martha Palmer, Don Perlis, Dennis Perzanowski, Walt Schneider, Kristinn Thorrison, Richard Waldinger, Ralph Weischedel.
The format of the symposium is a one-track session with panel discussions and a poster session. A joint session with the parallel Symposium on Naturally Inspired AI is expected. An early notification of the intent to participate with name, affiliation, address, phone and fax sent by email to samsonovich@cox.net is strongly encouraged. A limited number of scholarships are available for students attending the symposium. Interested students should apply via email.
Organizing Committee
Alexei Samsonovich (chair) (George Mason University), Deepak Khosla (HRL Laboratories, LLC), Laurent Itti (University of Southern California), Murray Shanahan (Imperial College London), Antonio Chella (University of Palermo), Richard Granger (Dartmouth College), Shane Mueller (Klein Associates Division / ARA), Ben Goertzel (Novamente / AGIRI), David Noelle (UC Merced).
Program Committee
Samuel Adams (IBM Research), James Albus (NIST), Alan Bond (University of California, Los Angeles), Michael Cox (BBN Technologies), Son Dao (ISSL, HRL Laboratories, LLC), Stanley Franklin (University of Memphis), John Gero (George Mason University and Massachusetts Institute of Technology), Elkhonon Goldberg (New York University School of Medicine), Andreas Herzog (IESK), Eva Hudlicka (Psychometrix), Neil Jacobstein (Stanford University and Teknowledge Corp.), Evguenia Malaia (SLHS Department, Purdue University), Howard Shrobe (Massachusetts Institute of Technology CSAIL), Narayan Srinivasa (HRL Laboratories LLC), Brian Tsou (Air Force Research Laboratory), Pei Wang (Temple University), Juyang (John) Weng (Michigan State University).
For more information about the symposium see the supplementary symposium web site.
Education Informatics: Steps Toward the International Internet Classroom
Could an international group provide free access to primary and secondary school curricula, aligned with national, state and local standards, delivered by our best AI tutoring technologies, in several languages, over the Internet? The purpose of this symposium is to discuss the feasibility of an International Internet Classroom Project.
We welcome education technologists with interests in architectures for content-delivery systems, content development and markup, pedagogical strategies, educational data mining, assessment, and classroom deployment; as well as new delivery hardware and modalities (e.g., the XO machine, mobile education). In addition to AI researchers, we encourage participation from government, the private sector, foundations, and the defense establishment, all of which have vital interests in very large-scale education and training systems.
The goals of the symposium are to sketch the core requirements for such a project and the immediate needs to get the project started. One critical requirement is that the core should be extensible, so researchers can incorporate and test the efficacy of their new technologies within the infrastructure — and with the participant pool — of the International Internet Classroom; to review the available sources of content, tools, platforms, student modeling methods, educational data mining techniques, pedagogical strategies, modes of content delivery, and concrete experiences deploying technology-based education; to explore the feasibility of common standards for representing content and student data and to review previous standards efforts in academia, the private sector, and government; and to build connections among people who work on related projects and sketch a plan to build the International Internet Classroom.
The symposium will include some talks and panels and much discussion. We assume no single group can build the International Internet Classroom, so an overarching goal of the symposium will be to build a community, assess what's currently available and establish working groups on the major aspects of the system.
Organizing Committee
Paul Cohen, chair (University of Southern California Information Sciences Institute), Carole Beal (K12@USC, University of Southern California Information Sciences Institute), Niall Adams (Imperial College London)
Program Committee
Ivon Arroyo (University of Massachusetts, Amherst), Ryan Baker (Carnegie Mellon University), Avron Barr (Aldo Ventures), Joe Beck (Worcester Polytechnic University), Bert Bredeweg (University of Amsterdam), Paul Brna (University of Edinburgh), Susan Bull (University of Birmingham), Weiqin Chen (University of Bergen), Pierre Dillenbourg (École Polytechnique Fédérale de Lausanne), Judith Good (University of Sussex), Jim Greer (University of Saskatchewan), Gord McCalla (University of Saskatchewan), Bruce McLaren (German Research Institute for AI (DFKI) and Carnegie Mellon University), Ulrich Hoppe (University of Duisburg-Essen), James Lester (North Carolina State University), Rose Luckin (London Knowledge Laboratory), Erica Melis (German Research Institute for Artificial Intelligence (DFKI), Jean-François Nicaud (Grenoble 1 University), Helen Pain (University of Edinburgh), Ron Stevens (University of California, Los Angeles), Giasemi Vavoula (University of Leicester), Beverly Woolf (University of Massachusetts, Amherst)
For more information about the symposium see the supplementary symposium web site. Please contact Paul Cohen (cohen@isi.edu) with questions.
Multimedia Information Extraction
This fall symposium will bring together researchers and practitioners in multimedia information extraction algorithms and systems and their underlying theories. Drawing from the language, image/video and spatial/temporal reasoning communities, the symposium will include presentations and demonstrations that address the processing of multiple media (for example, text, speech, maps, imagery, video) and multiple human perceptual modalities (for example, audition, vision). Accepted submissions will be organized into interactive panels focused on cross cutting topics such as (cross) media data sets, multimedia machine learning algorithms, innovative architectures, transmedia applications, evaluation methods. To increase the lasting value of the symposium, the chairs will capture capabilities and challenges into a roadmap containing lanes (for example, multimedia data, methods, and applications) that will then be used to stimulate iterative brainstorming “roadmap” sessions. An edited collection to include extended versions of the best papers and roadmap is planned.
Issues
Topics of discussion include:
- Object, attribute, and relation extraction from media (for example, text, audio, maps, imagery, and video)
- Simple and complex event detection and extraction from text, audio, imagery, and video
- Integrated speech, language, and image processing methods for cross media information extraction (that is, transmedia information extraction)
- Emotion and sentiment detection and tracking from media
- Tailoring multimedia information extraction to particular users, tasks, and contexts
- Intra- and intermedia representation languages and cross media ontologies
- Architectures for multimedia information extraction
- Constraints and capabilities of IE components and their integration
- Psychoperceptual and cognitive issues in multimodal information extraction
- Multimedia browsing/visualization tools and cross-media query (for example, visual, linguistic, and auditory)
- Studies and analyses of multimedia corpora
- Multimedia annotation schemes and tools
- Evaluation methods and metrics
- Innovative machine learning approaches
Organizing Committee
Mark Maybury, chair, (MITRE) (maybury@.mitre.org), Sharon Walter, cochair, (AFRL) (Sharon.Walter@rl.af.mil), Kelcy Allwein (DIA), Elisabeth Andre (University of Augsburg), Thom Blum (Audible Magic), Shih-Fu Chang (Columbia University), Bruce Croft (University of Massachusetts), Alex Hauptmann (Carnegie Mellon University), Andy Merlino (Pixel Forensics), Ram Nevatia (University of Southern California), Prem Natarajan (BBN), Kirby Plessas (Open Source Works), (David Palmer (Virage), Mubarak Shah (University of Central Florida), Rohini K. Shrihari (State University of New York Buffalo), Oliviero Stock (Istituto per la Ricerca Scientifica e Tecnologica), John Smith (IBM T. J. Watson Research Center), Rick Steinheiser (DNI/Open Source Center)
Naturally Inspired AI
The divide between how biological and computational systems solve cognitive problems and adjust to novel circumstances is readily apparent. While animals display marked flexibility in adjusting to new situations, our current computational approaches excel in well-defined, formally structured domains.
We are interested in new approaches to bridging this gap. Our perspective is that studies of natural and artificial intelligences can and should be mutually informative. Even young animals solve historically difficult computational problems, and we believe understanding how they do this will enable the creation of more sophisticated artificial systems. Conversely, computational models provide structure and insight into understanding animal learning and cognition. By combining biological and computational perspectives, we expect to obtain new insights that further the classical goals of artificial intelligence.
This symposium is intended to bring together researchers working on models that pertain directly to both natural and machine cognition. Particular methodology, motivation, or implementation decisions do not constrain our interests—we expect that relevant work may touch on themes as diverse as human experiments, neural models, engineering of complex systems, mathematical analysis, programming language design, and high-level cognitive models, to name only a few possibilities. We are interested in any work that has a clearly described relationship between a line of investigation and the larger problem of producing computational models that illuminate the peculiar nature and capabilities of cognition.
Topics of discussion include:
- Approaches to attaining breadth and flexibility
- Systems or models incorporating multiple cognitive capabilities
- Applying models of natural intelligence to engineered systems, or vice versa
- Case histories of recent success or interesting failure in crossing between these fields
- Near-term tractable problems deserving of greater attention
- Experimental techniques and measurement strategies
The symposium will mix short talks from participants with extensive discussion on the challenges of doing research relevant to both natural and artificial systems.
Organizing Committee
Jacob Beal (Massachusetts Institute of Technology CSAIL), Paul Bello (Office of Naval Research), Nick Cassimatis (Rensselaer Polytechnic Institute), Michael Coen (University of Wisconsin-Madison), Patrick Winston (Massachusetts Institute of Technology CSAIL)
For more information about the symposium see the supplementary symposium web site.