Modeling Expectation Violation in Intuitive Physics with Coarse Probabilistic Object Representations

TitleModeling Expectation Violation in Intuitive Physics with Coarse Probabilistic Object Representations
Publication TypeConference Proceedings
Year of Publication2019
AuthorsSmith, KA, Mei, L, Yao, S, Wu, J, Spelke, ES, Tenenabum, JB, Ullman, TD
Conference Name33rd Conference on Neural Information Processing Systems (NeurIPS 2019)
Date Published11/2019
Conference LocationVancouver, Canada
Abstract

From infancy, humans have expectations about how objects will move and interact. Even young children expect objects not to move through one another, teleport, or disappear. They are surprised by mismatches between physical expectations and perceptual observations, even in unfamiliar scenes with completely novel objects. A model that exhibits human-like understanding of physics should be similarly surprised, and adjust its beliefs accordingly. We propose ADEPT, a model that uses a coarse (approximate geometry) object-centric representation for dynamic 3D scene understanding. Inference integrates deep recognition networks, extended probabilistic physical simulation, and particle filtering for forming predictions and expectations across occlusion. We also present a new test set for measuring violations of physical expectations, using a range of scenarios derived from de- velopmental psychology. We systematically compare ADEPT, baseline models, and human expectations on this test set. ADEPT outperforms standard network architectures in discriminating physically implausible scenes, and often performs this discrimination at the same level as people.

URLhttp: //physadept.csail.mit.edu/
Download:  PDF icon ADEPT_NeurIPS.pdf

Associated Module: 

CBMM Relationship: 

  • CBMM Funded