Flexible neural representation for physics prediction
Name
8096-flexible-neural-representation-for-physics-prediction.pdf
Description
Published version
Size
2.77 MB
Format
Adobe PDF
Checksum (MD5)
417014241673d293c509e2c8937c9598
Author(s)
Mrowca, Damian
Zhuang, Chengxu
Wang, Elias
Haber, Nick
Li, Fei-Fei
Tenenbaum, Joshua B
Yamins, Daniel L. K.
Date Issued
December 2018
Journal
Advances in Neural Imaging Processing Systems
Publisher
Neural Information Processing Systems Foundation/Curran Associates
Citation
Mrowca, Damian et al. "Flexible neural representation for physics prediction." Advances in Neural Imaging Processing Systems, December 2018, Montreal, Canada, Neural Information Processing Systems Foundation/Curran Associates, December 2018 © 2018 Curran Associates
Version
Final published version
Abstract
Humans have a remarkable capacity to understand the physical dynamics of objects in their environment, flexibly capturing complex structures and interactions at multiple levels of detail. Inspired by this ability, we propose a hierarchical particle-based object representation that covers a wide variety of types of three-dimensional objects, including both arbitrary rigid geometrical shapes and deformable materials. We then describe the Hierarchical Relation Network (HRN), an end-to-end differentiable neural network based on hierarchical graph convolution, that learns to predict physical dynamics in this representation. Compared to other neural network baselines, the HRN accurately handles complex collisions and nonrigid deformations, generating plausible dynamics predictions at long time scales in novel settings, and scaling to large scene configurations. These results demonstrate an architecture with the potential to form the basis of next-generation physics predictors for use in computer vision, robotics, and quantitative cognitive science.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://papers.nips.cc/paper/8096-flexible-neural-representation-for-physics-prediction
