COGSAT 2020
The AAAI-20 Fall Symposium on
Cognitive Systems for Anticipatory Thinking
#COGSAT20
#COGSAT20
Anticipatory thinking, the deliberate and divergent exploration of relevant possible futures, is a key concept in several contexts. From formal definitions of intelligence analysis to the exploration of relevant possible futures in the presence of exogenous events of our everyday lives, we rely on anticipatory thinking (AT) to evaluate the current and possible future states of the world to prepare ourselves, avoid erroneous expectations, and mitigate risk.
This symposium focuses on understanding, quantifying, and improving anticipatory thinking capabilities across humans and machines. For more details on the Scope and Recommended Readings of our community, please visit the Home page.
Building off the success of our first symposium, we are now interested in looking at the intersection of cognitive systems, artificial intelligence, and prospective cognition communities with specific investigations into four AT challenge problems.
Progress, refinements and solutions, on challenge problems will be the primary focus of COGSAT 2020. Designing challenge problems is no small task and we have included reasoning behind them as well their concise statements in the finalized versions posted on July 15th, 2020 August 7, 2020. Concisely, they are:
Problem 1: Develop the capability to generate difficult autonomous vehicle scenarios that require learning without experience.
Problem 2: Extend Plan, Activity, and Intent Recognition (PAIR) agents to act to reduce prediction ambiguity
Problem 3: Formalize extensions to goal-reasoning agents that enables risk management for tactical and strategic planning.
Problem 4: Design a cognitive model that represents the effect counterfactual reasoning has on an individual’s ability to identify and forecast important events.
Acceptance to the symposium will depend on the strength of the contribution, but will also heavily depend on the clearly-articulated relevance to the above challenge problems. We expect a single type of submission:
Position papers (4-8 pages, plus 1 for references) describing interesting, novel extensions to existing work that addresses a challenge problem.
To clarify what the symposium would like to showcase, this is the review criteria for accepting submissions. Paper submissions should follow AAAI style guidelines.
Papers will be uploaded through EasyChair (link: https://easychair.org/conferences/?conf=fss20) and will be guaranteed at least two reviews. Further, papers should be format for blind review with all identifying author information removed. You are welcome to upload your paper on your own personal website or ArXiv with identifying information. As with last year, we plan to archive accepted submissions in a volume of CEUR-WS.
All deadlines are specified as Anywhere on Earth time unless otherwise noted.
Submissions due: September 25th, 2020
Notifications to authors: October 23th, 2020
Final papers due to organizers: Nov 6th, 2020
Registration deadline: TBD
TBD
Dr. Adam Amos-Binks (chair), Applied Research Associates, Inc.
Dr. Dustin Dannenhauer, Navatek LLC
Dr. Rogelio E. Cardona-Rivera, University of Utah
Dr. Gene Brewer, Arizona State University
Dr. Leilani H. Gilpin, MIT