( events)   Timezone: »  
Mon Dec 13 05:45 AM -- 02:30 PM (PST)
Metacognition in the Age of AI: Challenges and Opportunities
Ingmar Posner · Francesca Rossi · Lior Horesh · Steve Fleming · Oiwi Parker Jones · Rohan Paul · Biplav Srivastava · Andrea Loreggia · Marianna Ganapini

Workshop Home Page

Recent progress in artificial intelligence has transformed the way we live, work, and interact. Machines are mastering complex games and are learning increasingly challenging manipulation skills. Yet where are the robot agents that work for, with, and alongside us? These recent successes rely heavily on the ability to learn at scale, often within the confines of a virtual environment. This presents significant challenges for embodied systems acting and interacting in the real world. In contrast, we require our robots and algorithms to operate robustly in real-time, to learn from a limited amount of data, to take mission and sometimes safety-critical decisions, and increasingly even to display a knack for creative problem solving. Achieving this goal will require artificial agents to be able to assess - or introspect - their own competencies and their understanding of the world. Faced with similar complexity,  there are a number of cognitive mechanisms which allow humans to act and interact successfully in the real world. Our ability to assess the quality of our own thinking - that is, our capacity for metacognition - plays a central role in this. We posit that recent advances in machine learning have, for the first time, enabled the effective implementation and exploitation of similar processes in artificial intelligence. This workshop brings together experts from psychology and cognitive science with cutting-edge research in machine learning, robotics, representation learning and related disciplines, with the ambitious aim of re-assessing how models of intelligence and metacognition can be leveraged in artificial agents given the potency of the toolset now available.

Introduction to the Workshop on Metacognition in the Age of AI: Challenges and Opportunities (Live introduction from organizers)
How does a brain compute confidence? (Invited Talk)
Credit Assignment & Meta-Learning in a Single Lifelong Trial (Invited Talk)
Panel Discussion 1 (Panel Discussion/Q&A)
Coffee Break (Break)
Freespace Supports Metacognition for Navigation (Invited Talk)
Desiderata and ML Research Programme for Higher-Level Cognition (Invited Talk)
Panel Discussion 2 (Panel Discussion/Q&A)
Lunch (Break)
Poster Session
Performance-Optimized Neural Networks as an Explanatory Framework for Decision Confidence (Invited Talk)
Causal World Models (Invited Talk)
Panel Discussion 3 (Panel Discussion/Q&A)
Closing Remarks (Capstone to the day, and thanks, from the organizers)
An Algorithmic Theory of Metacognition in Minds and Machines (Poster)
Non-Robust Feature Mapping in Deep Reinforcement Learning (Poster)
Thinking Fast and Slow in AI: The Role of Metacognition (Poster)
Meta Dynamic Programming (Poster)
Measuring and Modeling Confidence in Human Causal Judgment (Poster)
Have I done enough planning or should I plan more? (Poster)
Promoting Metacognitive Learning through Systematic Reflection (Poster)