Intelligent, autonomous parameter space exploration of self-assembly simulations

ORAL

Abstract

Researchers studying self-assembly are often plagued with a problem: the systems they study exhibit emergent behavior that could result from any combination of the many independent parameters they can tune. The typical solution to this problem is to restrict a study to the few most interesting variables and perform a screening experiment on a grid in this space. But what can be done when it is not clear which variables are most important? The scale of these studies quickly gets out of hand as analysis, visualization, and even selecting new experiments are multiplied by the dimensionality of the parameter space. Here we discuss approaches to incorporate machine learning methods into the experimental design and analysis loop of exploratory self-assembly simulations in order to optimize computational time spent simulating interesting and novel behaviors. By harnessing these methods, we can begin to probe the behavior of even more complex design spaces.

–

Presenters

  • Matthew Spellings

    Chemical Engineering, University of Michigan

Authors

  • Matthew Spellings

    Chemical Engineering, University of Michigan

  • Alexey Feofanov

    University of Innsbruck, University of Waterloo, Korea University, Okinawa Institute of Science and Technology, University of California - Los Angeles, The University of Manchester, University of Puerto Rico at Humacao, Department of Physics & Electronics, University of Puerto Rico at Cayey, Department of Mathematics-Physics, Oak Ridge National Lab, Max Planck Institute for Chemical Physics of Solids, Department of Physics, University of Puerto Rico, Electrical Engineering Department, University of Arkansas, Department of Physics, University of Arkansas, School of Basic Sciences at IIT Mandi, H.P., India, Computational Biology, Flatiron Institute, Physics, Hong Kong Univ of Sci & Tech, University of California, Los Angeles, Max Planck Inst, Institute for Theoretical Physics, University of Cologne, Department of Physics, Simon Fraser University, Deutsches Elektronen Synchrotron (DESY), Institut fur Theoretische Physik, Univerisitat zu Berlin, Institut fur Physik, Univerisitat zu Berlin, Plymouth State University, The Graduate Center, CUNY, Nordita, KTH Royal Institute of Technology and Stockholm University, Univ of Connecticut - Storrs, Univ Stuttgart, University of Chicago, University of Texas at El Paso, University of Tulsa, California Institute of Technology, Georgia Institute of Technology, Universite Paris Diderot, Laboratoire MPQ, Universita di Trento, BEC Center, ICTP Trieste, Universita di Pisa, Inst of Physics Academia Sinica, Batelle, Cal State Univ- San Bernardino, Chemical Engineering, University of Michigan, QCD Labs, Department of Applied Physics, Aalto University, Yale University, MIT, Harvard Univ, Chemical & Environmental Engineering, University of California, Riverside, University of Frankfurt, Germany, University of Hamburg, Germany, Naval Research Laboratory, Cornell Univ, National Institute for Material Science, U.S. Naval Research Laboratory, Washington DC, Materials Engineering, University of Santa Barbara, Institute of Physics, Chinese Academy of Sciences, Univ of Texas, Arlington, MIT Lincoln Laboratory, University of Sydney, Iowa State University, Purdue University, Kansas State University, University of Maryland, John Hopkins University, Universite de Sherbrooke, Physics, Konkuk University, Perimeter Institute, University of Waterloo, D-Wave, San Jose State University, Université de Sherbrooke, Institute of Physics, EPFL - Lausanne​

  • Sharon Glotzer

    Chemical Engineering, Univ of Michigan - Ann Arbor, Univ of Michigan - Ann Arbor, Department of Chemical Engineering, University of Michigan - Ann Arbor, Department of Chemical Engineering, University of Michigan, Chemical Engineering, University of Michigan, Department of Chemical Engineering, Univ of Michigan - Ann Arbor