Models of Core Knowledge (Physics, Really)
Cognition, Cognitive Psychology, Computer Vision and seminar by Prof Tomer Ullman, Harvard University
Hosted by CompCogSci Darmstadt
Wednesday 14:00–15:10 London (GMT+1)
Ended
Abstract
Even young children seem to have an early understanding of the world around them, and the people in it. Before children can reliably say "ball", "wall", or "Saul", they expect balls to not go through walls, and for Saul to go right for a ball (if there's no wall). What is the formal conceptual structure underlying this commonsense reasoning about objects and agents? I will raise several possibilities for models underlying core intuitive physics as a way of talking about models of core knowledge and intuitive theories more generally. In particular, I will present some recent ML work trying to capture early expectations about object solidly, cohesion, and permanence, that relies on a rough-derendering approach.
Topics
Related seminars
- Video · 36 min · Feb 11, 2021Dr Tommi Kokkonen / Prof Lennart Schalk, University of Helsinki / University of Education Schwyz
Dr Lindsay reads from "Models of the Mind : How Physics, Engineering and Mathematics Shaped Our Understanding of the Brain" 📖
Video · 1 h 7 min · May 10, 2021Grace Lindsay, Gatsby Unit for Computational NeuroscienceUnderstanding machine learning via exactly solvable statistical physics models
Video · 1 h 13 min · Jun 24, 2020Lenka Zdeborová, CNRS & CEA Saclay
More from CompCogSci Darmstadt
All talksFlexible codes and loci of visual working memory
Jul 13, 2022Rosanne RademakerSuccesses and failures of current AI as a model of visual cognition
Jul 6, 2022Gabriel KreimanFrom Computation to Large-scale Neural Circuitry in Human Belief Updating
Jun 29, 2022Tobias DonnerFeedforward and feedback processes in visual recognition
Jun 22, 2022Thomas Serre