GSOFT Short Course: Machine Learning and Data Science in Soft Matter
Sunday, March 4
8:30 a.m. - 6:00 p.m.
408B Los Angeles Convention Center
Los Angeles, CA
Organizers.

Andrew Ferguson, University of Illinois at Urbana-Champaign (alf@illinois.edu)
Eric Jankowski, Boise State University (ericjankowski@boisestate.edu‎)
Invited Speakers.

Yannis Kevrekidis - Johns Hopkins University
Ekin Dogus Cubuk - Google Brain
Frank Noé - Freie Universität Berlin
Juan de Pablo - University of Chicago
Bryce Meredig - Citrine
Johannes Hachmann - University at Buffalo
Overview.

Data-driven modeling approaches and machine learning have opened new paradigms in the understanding, engineering, and design of soft and biological materials. The advent of high-throughput experimental synthesis and characterization platforms, and the increasing prevalence of high-performance and multicore computer hardware have led to a deluge of data in soft matter. Analysis of these voluminous and multidimensional data sets requires soft matter researchers to implement and adapt tools from machine learning and data science. This one-day workshop will provide emerging and established soft matter researchers with exposure and training in machine learning and data science tools through a series of tutorials from some of the leading experts in the field. Topics to be covered include nonlinear manifold learning, enhanced sampling, materials informatics, and inverse soft materials design. Attendees will leave with both an appreciation for the state-of-the-art applications of data science in soft matter research, and a working knowledge of user-friendly Python libraries to implement these approaches in their own work.
Who Should Attend?

This workshop is appropriate for all soft matter physicists who wish to integrate machine-learning tools into their domain-specific expertise. The course is expected to be particularly well-suited to those who have not received formal training in data science tools, but recognize the value of these approaches in advancing their own research endeavors. The workshop is designed to accommodate all levels of attendees from students and post-docs to established faculty members. Computational and experimental researchers are equally welcome.
Pre-Reqs for Hands-On Python Workshop.

The “Machine Learning with Python” session is a hands-on workshop. Attendees will bring their own laptop with a working Anaconda Python 2.7 installation equipped with the scikit-learn and pandas libraries.

Anaconda provides free and easy-to-install Python releases equipped with scikit-learn, pandas, and numerous other scientific libraries. Instructions are available at http://scikit-learn.org/stable/install.html and https://pandas.pydata.org/pandas-docs/stable/install.html

For those attendees who have preferred integrated development environments, information about setting up Anaconda in various IDEs is available at https://docs.anaconda.com/anaconda/user-guide/getting-started#links-to-ide-documentation

Materials for the workshop can be downloaded from https://bitbucket.org/hachmanngroup/ml_tutorial/src
Registration.

Member: $150
Students/Postdocs: $100
Nonmembers: $300

Please complete your application at https://www.aps.org/meetings/march/shortcourses.cfm#gsoft
Workshop Schedule.
Stacks Image 1387