Bulletin of the American Physical Society
2023 APS March Meeting
Volume 68, Number 3
Las Vegas, Nevada (March 5-10)
Virtual (March 20-22); Time Zone: Pacific Time
Session T60: Emerging Trends in Molecular Dynamics Simulations and Machine Learning VI
11:30 AM–1:54 PM,
Thursday, March 9, 2023
Room: Room 419
Sponsoring
Unit:
DCOMP
Chair: Pankaj Rajak, Argonne National Laboratory
Abstract: T60.00007 : Physics-Guided Problem Decomposition for Scaling Deep Learning of Quantum Spin Models*
12:42 PM–12:54 PM
Presenter:
Wei-Cheng Lee
(Binghamton University)
Authors:
Wei-Cheng Lee
(Binghamton University)
Sangeeta Srivastava
(Ohio State University)
Samuel Olin
(Binghamton University)
Viktor A Podolskiy
(University of Massachusetts Lowell)
Anuj Karpatne
(Virginia Tech)
Anish Arora
(Ohio State University)
In this paper, we alleviate the compute bottleneck in the output layer by using physics knowledge to decompose the complex regression task of predicting the high-dimensional eigenvectors into multiple simpler sub-tasks, each of which are learned by a simple "expert" network. We call the resulting architecture of specialized experts Physics-Guided Mixture-of-Experts (PG-MoE). We demonstrate the efficacy of such physics-guided problem decomposition for the case of the Schrödinger's Equation in Quantum Mechanics. Our proposed PG-MoE model predicts the ground-state solution, i.e., the eigenvector that corresponds to the smallest possible eigenvalue. The model is 150x smaller than the network trained to learn the complex task while being competitive in generalization. To improve the generalization of the PG-MoE, we also employ a physics-guided loss function based on variational energy, which by quantum mechanics principles is minimized iff the output is the ground-state solution.
*This work was supported by the National Science Foundation via grants 2026702, 2026703, 2026704, and 2026710.
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700