Bulletin of the American Physical Society
APS March Meeting 2022
Volume 67, Number 3
Monday–Friday, March 14–18, 2022; Chicago
Session T14: Learning dynamical models across physical systemsInvited Live Streamed
|
Hide Abstracts |
Sponsoring Units: DBIO Chair: Joshua Shaevitz, Princeton University Room: McCormick Place W-183B |
Thursday, March 17, 2022 11:30AM - 12:06PM |
T14.00001: Learning dominant physical processes with data-driven balance models Invited Speaker: Bing Brunton
|
Thursday, March 17, 2022 12:06PM - 12:42PM |
T14.00002: Using Knowledge-based Neural Ordinary Differential Equations to Learn Complex Dynamics and Chaos Invited Speaker: M. Ani Hsieh Robots and autonomous systems give us unprecedented access to landscapes and habitats both big and small. They provide in-situ monitoring of the environments they are immersed in and adapt their strategies to respond to various external stimuli. These systems enable us to more richly and extensively interact with the world we live in, better our understanding of the complexities of the world, and assist in the discovery of new processes and phenomena. Nevertheless, the ability to robustly operate in natural unstructured environments often requires robots to have or acquire a model or estimate the environment, often with limited sensing and communication resources. In this talk, I will present some of our recent efforts in developing knowledge embedded machine learning strategies for modeling and predicting complex spatiotemporal phenomena. |
Thursday, March 17, 2022 12:42PM - 1:18PM |
T14.00003: DBIO Dissertation Award (2021): Physics of Behavior Across Scales Invited Speaker: Antonio Carlos Costa Animal movement exhibits multiple time scales: from fine scale posture movements, to changes in navigation strategies driven by internal states, all the way up to aging. These disparate scales are often studied separately: long time behaviors are typically described through body centroid kinematics, while finer-scale behaviors are studied at the level of posture. Here we introduce a unifying formalism that bridges between these scales. Starting from posture measurements, we reconstruct a state space by concatenating measurements in time, building a maximum entropy partition of the resulting sequences, and choosing the sequence length to maximize predictive information. Trading non-linear trajectories for linear, ensemble evolution, we analyze reconstructed dynamics through transfer operators. The eigenvalues of the inferred transfer operators reveal a hierarchy of time scales, while the respective eigenvectors capture metastable states that correspond to long-lived behavioral states. Leveraging this analysis, we uncover slowly changing navigation strategiesĀ and predict large scale diffusive properties of centroid trajectories directly fromĀ fine-scale posture measurements. Additionally, we introduce slow non-ergodic drives that reflect changing internal states, resulting in behavioral dynamics that evolve in potential landscapes that fluctuate in time. We explore the consequences of such a picture and recover a number of statistical properties observed across behaving organisms, such as long-range correlations, heavy-tailed dwell-time distributions and non-Markovianity. |
Thursday, March 17, 2022 1:18PM - 1:54PM |
T14.00004: Machine learning approaches to biomechanics Invited Speaker: Vincenzo Vitelli
|
Thursday, March 17, 2022 1:54PM - 2:30PM |
T14.00005: Physics-informed machine learning: climate modeling and COVID-19 forecasting Invited Speaker: Rose Yu
|
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