The campus remains closed with all summer courses being conducted online and staff working remotely. Classes will resume on August 31, 2020 for the fall semester. For updates on campus operations, both virtual and in person, for the fall semester, please see the Fall 2020 Info Hub page.

Prof. Robert Topper co-authors Paper in the Journal of Physical Chemistry B

POSTED ON: April 16, 2020

Imgae: Hardcover of the 2020 Journal of Physical Chemistry B.

Imgae: Hardcover of the 2020 Journal of Physical Chemistry B.

Professor of Chemistry, Robert Q. Topper, co-authored a paper titled “Comparison of the Performance of Machine Learning Models in Representing High-Dimensional Free Energy Surfaces and Generating Observables” published in the Journal of Physical Chemistry B.

ABSTRACT:

Free energy surfaces of chemical and physical systems are often generated using a popular class of enhanced sampling methods that target a set of collective variables (CVs) chosen to distinguish the characteristic features of these surfaces. While some of these approaches are typically limited to low ($\sim$1-3)-dimensional CV subspaces, methods such as driven adiabatic free-energy dynamics/ temperature-accelerated molecular dynamics have been shown to be capable of generating free energy surfaces of quite high dimension by sampling the associated marginal probability distribution via full sweeps over the CV landscape. These approaches repeatedly visit conformational basins, producing a small scattering of points within the basins on each visit. Consequently, they are particularly amenable to synergistic combination with regression machine learning methods for filling in the surfaces between the sampled points and for providing a compact and continuous (or semi-continuous) representation of the surfaces that can be easily stored and used for further computation of observable properties. Given the central role of machine learning techniques in this combined approach, it is timely to provide a detailed comparison of the performance of different machine learning strategies and models, including neural networks, kernel ridge regression, support vector machines, and weighted neighbor schemes, for their ability to learn these high-dimensional surfaces as a function of the amount of sampled training data and, once trained, to subsequently generate accurate ensemble averages corresponding to observable properties of the systems. In this article, we perform such a comparison on a set of oligopeptides, in both gas and aqueous phases, corresponding to CV spaces of 2-10 dimensions and assess their ability to provide a global representation of the free energy surfaces and to generate accurate ensemble averages.

Read more here.

  • Founded by inventor, industrialist and philanthropist Peter Cooper in 1859, The Cooper Union for the Advancement of Science and Art offers education in art, architecture and engineering, as well as courses in the humanities and social sciences.

  • “My feelings, my desires, my hopes, embrace humanity throughout the world,” Peter Cooper proclaimed in a speech in 1853. He looked forward to a time when, “knowledge shall cover the earth as waters cover the great deep.”

  • From its beginnings, Cooper Union was a unique institution, dedicated to founder Peter Cooper's proposition that education is the key not only to personal prosperity but to civic virtue and harmony.

  • Peter Cooper wanted his graduates to acquire the technical mastery and entrepreneurial skills, enrich their intellects and spark their creativity, and develop a sense of social justice that would translate into action.