Bulletin of the American Physical Society
APS March Meeting 2018
Volume 63, Number 1
Monday–Friday, March 5–9, 2018; Los Angeles, California
Session K32: Data Science as the Driving Force for Industrial PhysicsIndustry Invited Undergraduate
|
Hide Abstracts |
Sponsoring Units: FIAP FECS Chair: Jason Stewart Gardner, National Synchrotron Radiation Research Center Room: LACC 408A |
Wednesday, March 7, 2018 8:00AM - 8:36AM |
K32.00001: How Big Data Unlocks the New Many-body Physics of Online Threats Invited Speaker: Neil Johnson Physics is ultimately driven by empirical observations. Thanks to the Internet and its underlying infrastructure, a new world has opened up in which large amounts of data can be freely obtained by academics and industries for both research and commercial applications. Just like measurements of the physical world, however, the data that emerges can be big but messy. In this talk, I will show you that such Big Data not only provides a new source for supplementing existing studies, but it can also open up new Physics as well as a new understanding of pressing problems in the real world. My first example concerns large-scale, ultrafast cyber-physical (e.g. financial) systems in which there is no hope of real-time human intervention when things go wrong, since they are limited only by the speed of light. Though already in use 24/7, the science of such systems is not well understood — in particular, the extreme events or Black Swans which are like digital tsunamis. I will show how studying this particular Big Data can improve our understanding not only of threats in this important real-world all-electronic system, but it can also help our understanding of system failures in many others, including the human brain. As my second example, I will mention how Big Data has revealed a new Physics-centric understanding of arguably the most urgent threat facing society: human extremism and its rapid online growth. Though very applied in nature, the Big Data from this latter domain has unwrapped a wealth of interesting Physics related to many-body out-of-equilibrium systems, and complex dynamical networks. More details of the background research is available at http://www.physics.miami.edu/~njohnson/ |
Wednesday, March 7, 2018 8:36AM - 9:12AM |
K32.00002: Solving industrial materials problems by using machine learning across diverse computational and experimental data Invited Speaker: Bryce Meredig Many materials characteristics of broad industrial relevance, such as in-service degradation, lack satisfactory physics-based models. Nonetheless, we often have access to more fundamental simulations, mechanistic theories, and experiments that we might expect to correlate in some way with a highly applied property of interest. In these situations, we suggest that machine learning can serve as an effective integrator of physical signal from many different sources, ranging from first-principles calculations to analytical models to experimental observations, in service of predicting a complex property that has traditionally resisted accurate modeling. This data-driven approach is advantageous because it utilizes our existing understanding of materials physics to the greatest extent possible, while still enabling us to extend beyond well-understood regimes. Further, it allows us to partially offset the need for very expensive real-world tests with databases of simpler experiments, or plentiful computational data. |
Wednesday, March 7, 2018 9:12AM - 9:48AM |
K32.00003: What physics does and doesn't teach you about data science Invited Speaker: David Purdy Paraphrasing Leo Breiman: almost nobody can claim their childhood ambition was to be a data scientist. Other fields have “dedicated travelers” - people who can say that ever since childhood they wanted to be a physicist, or a doctor, or an engineer. Not so for data science in 2017, which is a convergence point for people from many backgrounds. |
Wednesday, March 7, 2018 9:48AM - 10:24AM |
K32.00004: Machine Learning Models vs Physics Models: The Battle for acceptance Invited Speaker: Sergey Yurgenson "All models are wrong but some are useful" (George Box) For many centuries, physics models helped us to understand the universe, visualize processes, grasp the main ideas in hypotheses and theories. They were simplified representations of reality, which, otherwise, could be too complex for human mind to comprehend. They were useful because we could understand them. Now we are dealing with different type of models: Machine Learning Models. They are not designed to make a picture simpler. They are not based on first principles or even hypotheses. They are mainly created with one specific goal – to make predictions. In some situations, they can be understood, but in many cases. they are just black boxes that accept some input and generate predictions as output. Are those models “useful” ? Can we accept predictions without completely understanding how these predictions are made? Are there ways to make those predictions more “transparent”? And if they are not transparent enough, is there still a place for them in our pursuit of knowledge? |
Wednesday, March 7, 2018 10:24AM - 11:00AM |
K32.00005: A hitchhiker’s guide to Data Science Invited Speaker: Sundeep Das This abstract not available. |
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