Operationally meaningful representations of physical systems in neural networks

HP Nautrup, T Metger, R Iten, S Jerbi… - Machine Learning …, 2022 - iopscience.iop.org
To make progress in science, we often build abstract representations of physical systems
that meaningfully encode information about the systems. Such representations ignore …

Predicting the physical dynamics of unseen 3d objects

D Rempe, S Sridhar, H Wang… - Proceedings of the …, 2020 - openaccess.thecvf.com
Abstract Machines that can predict the effect of physical interactions on the dynamics of
previously unseen object instances are important for creating better robots and interactive …

[PDF][PDF] Predicting the physical dynamics of unseen 3d objects

SS Davis Rempe - IEEE Winter Conference on Applications of Computer …, 2020 - par.nsf.gov
Abstract Machines that can predict the effect of physical interactions on the dynamics of
previously unseen object instances are important for creating better robots and interactive …

Combining learning and structure for robotic manipulation

A Kloss - 2021 - tobias-lib.ub.uni-tuebingen.de
Household robots have been a long standing promise of robotics. But despite of decades of
robotic research and progress in the field, we still do not encounter many robots in our …