[Triumf-seminars] TRIUMF Colloquium today at 14:00

TRIUMF Seminars triumf-seminars at lists.triumf.ca
Thu Mar 10 05:00:01 PST 2016


Date/Time: Thu 2016-03-10 at 14:00

Location:  Auditorium          

Speaker:   Roman Krems (UBC)

Title:     Statistical learning for quantum scattering theory: applications of Gaussian Processes for improved predictions of molecular collision observables

Abstract: I will illustrate that statistical learning techniques based on kriging (Gaussian Process regression) have enormous potential for improving the predictions of classical and/or quantum scattering theory. I will discuss the following applications of Gaussian Process models: (i) efficient non-parametric fitting of multi-dimensional potential energy surfaces without the need to fit ab initio data with analytical functions; (ii) obtaining scattering observables as functions of individual PES parameters; (iii) using classical trajectories to interpolate quantum results; (iv) extrapolation of scattering observables from one molecule to another; (v) obtaining scattering observables with error bars reflecting the inherent inaccuracy of the underlying potential energy surfaces.

I will argue that the application of Gaussian Process models to quantum scattering calculations may potentially elevate the theoretical predictions to the same level of certainty as the experimental measurements and can be used to identify the role of individual atoms in determining the outcome of collisions of complex molecules.

References:

(1) Jie Cui and R. V. Krems, “Gaussian Process Model for Collision Dynamics of Complex Molecules”, Phys. Rev. Lett. 115, 073202 (2015).

(2) Jie Cui, Zhiying Li and R. V. Krems, “Gaussian Process Model for Extrapolation of Collisions Observables for Complex Molecules: from Benzene to Benzonitrile”, J. Chem. Phys. 143, 154101 (2015).

(3) Jie Cui and R. V. Krems, “Efficient non-Parametric Fitting of Potential Energy Surfaces for Polyatomic Molecules with Gaussian Processes”, arXiv: 1509.06473.



______________________________

General TRIUMF seminar information available at http://www.triumf.ca/home/upcoming-events/about-seminars-lectures



More information about the Triumf-seminars mailing list