Lab Home | Phone | Search
	Center for Nonlinear Studies
	Home  ▶ People   Current   Executive Committee   Postdocs   Visitors   Students   Research   Publications  ▶ Conferences   Workshops   Sponsorship  ▶ Talks   Seminars   Postdoc Seminars Archive   Quantum Lunch   Quantum Lunch Archive   P/T Colloquia   Archive   Ulam Scholar     Postdoc Nominations  ▶ Student Requests   Student Program  ▶ Visitor Requests   Description   Past Visitors  ▶ Services   General     History of CNLS     Maps, Directions   CNLS Office   T-Division   LANL				Thursday, November 07, 201909:30 AM - 11:00 AMCNLS Conference Room (TA-3, Bldg 1690) Colloquium Can machine learning make rare events more frequent? Gerhard HummerMax Planck Institute of Biophysics Computational molecular sciences face three major challenges: model quality (i.e., force fields), data explosion, and nonetheless inadequate sampling of the most interesting events. Machine learning approaches hold much promise in all three areas. Our work concerns the latter two: how to extract useful information from large amounts of trajectory data, and how to steer the sampling to maximize the amount of useful information. We will report on our progress towards a fully autonomous, data-driven production and interpretation of molecular dynamics trajectories by combining machine learning and enhanced sampling methods. Host: Angel E. Garcia