3d Physics Informed Machine Learning
January 13-17, 2020
Inn and Spa at Loretto, Santa Fe, New Mexico

Workshop website: cnls.lanl.gov/piml2020

Previous editions of the workshop are available at cnls.lanl.gov/piml2018 and cnls.lanl.gov/piml2016.

Topics:

• Theory and Methods (Monday and Tuesday morning)
  Deep learning, graphical models, optimization, tensor and message-passing methods, statistical learning
  
• Algorithms and Applications (Tuesday afternoon and Wednesday)
  Physics, computer science, biology, dynamical systems, novel computational paradigms (quantum, memristors), etc.
  
• Special Topic: Spatio-Temporal Flows (Thursday and Friday morning)
  Dynamical systems, fluid mechanics and turbulence

Important Dates:

• Financial support requests: November 15, 2019
• Abstract submissions: November 15, 2019
• Travel grant awards decisions: December 6, 2019
• Workshop dates: January 13-17, 2020

Travel Grants for Students and Postdocs:

A limited number of travel grants are available for selected graduate students and postdocs to help offset the cost of attending the conference. Those interested should submit an application, including poster abstract and CV, before the deadline of November 15, 2019. We will make funding decisions by December 6, 2020. Please submit your application to: piml2020@lanl.gov.

Posters:

Participants are encouraged to submit a poster, which will be displayed throughout workshop. We will also schedule a slot for poster discussions, tentatively scheduled for Monday, the first day of the workshop. Depending on the schedule, some posters will be selected for longer contributed presentations throughout the week. If you are applying for a travel grant, then you are required to submit your poster abstract by Friday, November 15, 2019.

Summary:

This workshop continues discussions and explorations started in January 2016 and January 2018 at the first and second editions of the workshop. A revolution in statistics and machine learning (ML) is underway. Modern algorithms can now learn high level abstractions via hierarchical models, leading to break-through accuracy in benchmarks for computer vision, language, etc. Underlying these advances is a strong and deep connection to various aspects of applied mathematics and statistical physics. For example, proper choice of statistical force allows to screen interaction and learn graphical models governing multi-dimensional distributions efficiently, gauge transformations from physics guide incorporation of symmetries in the neural network design, dynamical system interpretation helps to understand and improve performance of most efficient deep learning schemes, etc.

This workshop seeks perspectives on leveraging the deep connection between ML and physics, but now with the goal to better understand and model physical systems, static and dynamic. We invite experts both in machine learning techniques as well as domain science applications. In this third edition of PIML we will focus on applications of interest to our LANL and DOE sponsors, specifically on improving scale-reduced, large eddy modeling and simulations of turbulence that arise in various mechanical-engineering, aerospace and climate applications; building reduced models for infrastructures (energy systems, cyber-physical systems, etc.); guiding development of inverse/design approaches in quantum physics, e.g. related to tensor networks; designing new computational paradigms (such as related to quantum and neuromorphic computing).

The workshop discussions are aimed towards approaches and methods for physical modeling applications where a big-data, black-box approach to ML is only a starting point. We seek participants who may suggest innovative approaches that extend application agnostic ML techniques by incorporating complex constraints imposed by physical principles (e.g. conservation laws, causality, symmetries, entropy principles and related). The workshop format assumes 7 talks a day, late afternoon discussions and poster session. We encourage application of young researchers for fellowships (see website for details). We plan for active participation of LANL researchers and program managers across directorates and divisions interested in the physics informed learning. This emerging area of research has many aspects of computational co-design, and draws on LANL's strengths in statistical physics, modern applied mathematics, theoretical and applied computer science, infrastructure modeling and simulations, fluids and materials modeling, and high performance computing. Looking forward, we view physics informed learning as a viable path for LANL and DOE toward truly predictive multi-scale modeling, optimization, inference and leading for plethora of relevant applications in sciences and engineering.

Organizing Committee:

Michael Chertkov (University of Arizona)
Andrey Lokhov (Los Alamos National Laboratory)
Arvind Mohan (Los Alamos National Laboratory)