Understanding Many-Particle Systems with Machine Learning

September 12 - December 16, 2016


MPS2016 ImageInteractions between many constituent particles (bodies) generally give rise to collective or emergent phenomena in matter. Even when the interactions between the particles are well defined and the governing equations of the system are understood (for example the Coulomb interaction between protons and electrons and the Dirac/Schroedinger equation in quantum mechanics), the collective behavior of the system as a whole does not trivially emerge from these equations. Examples of collective behavior are abundant in nature, manifesting themselves at all scales of matter, ranging from atoms to galaxies. Machine learning methods have been used extensively in a wide variety of fields ranging from, for example, the neurosciences, genetics, multimedia search to drug discovery. Machine learning models can be thought of as universal approximators that learn a (possibly very complex) nonlinear mapping between input data (descriptor) and an output signal (observation). It is the goal of this IPAM long program to bring together experts in many particle problems in condensed-matter physics, materials, chemistry, and protein folding, together with experts in mathematics and computer science, to synergetically address the problem of emergent behavior and understand the underlying collective variables in many particle systems.

Organizing Committee

Alán Aspuru-Guzik (Harvard University)
Gabor Csanyi (University of Cambridge)
Mauro Maggioni (Duke University, Mathematics and Computer Science)
Stéphane Mallat (École Normale Supérieure)
Marina Meila (University of Washington, Statistics)
Klaus-Robert Müller (Technische Universität Berlin)
Alexandre Tkatchenko (University of Luxembourg, Theory)