Unke, Oliver Thorsten. Potential energy surfaces: from force fields to neural networks. 2019, Doctoral Thesis, University of Basel, Faculty of Science.

PDF
15Mb 
Official URL: http://edoc.unibas.ch/diss/DissB_13227
Downloads: Statistics Overview
Abstract
Almost a century ago, Paul A. M. Dirac remarked that the Schrödinger equation (SE) contains all that is necessary to describe chemical phenomena. Unfortunately, solving the SE, even approximately, remains a timeintensive task and is possible only for systems containing a few atoms. For this reason, potential energy surfaces (PESs) are used to circumvent the solution of the SE altogether: They estimate the energy of a chemical system by evaluating an analytical function. For example, socalled force fields (FFs) model chemical bonds as "springs", i.e. with harmonic potentials. While this is computationally efficient, it limits the accuracy of FFs. A promising alternative are machine learning (ML) methods, such as kernel ridge regression (KRR) and artificial neural networks (NNs), which allow the construction of a PES without assuming a functional form.
In the first part of this thesis, FFs are described in more detail and the minimal distributed charge model (MDCM) is introduced as a way to increase their accuracy. Applications to several challenging molecules are used to demonstrate the utility of this method. In the second part, KRR is reviewed and ways to improve its computational efficiency for PES construction are discussed. Further, the reproducing kernel Hilbert space (RKHS) toolkit is introduced, which largely automates the construction of efficient and accurate PESs for small systems. In the last part, a brief overview of NNs and their historic development is given. The use of NNs to construct PESs is explored and two alternatives are described in more detail. Both variants are applied to various benchmark datasets in order to demonstrate their versatility and accuracy.
In the first part of this thesis, FFs are described in more detail and the minimal distributed charge model (MDCM) is introduced as a way to increase their accuracy. Applications to several challenging molecules are used to demonstrate the utility of this method. In the second part, KRR is reviewed and ways to improve its computational efficiency for PES construction are discussed. Further, the reproducing kernel Hilbert space (RKHS) toolkit is introduced, which largely automates the construction of efficient and accurate PESs for small systems. In the last part, a brief overview of NNs and their historic development is given. The use of NNs to construct PESs is explored and two alternatives are described in more detail. Both variants are applied to various benchmark datasets in order to demonstrate their versatility and accuracy.
Advisors:  Meuwly, Markus and Müller, KlausRobert 

Faculties and Departments:  05 Faculty of Science > Departement Chemie > Chemie > Physikalische Chemie (Meuwly) 
UniBasel Contributors:  Meuwly, Markus 
Item Type:  Thesis 
Thesis Subtype:  Doctoral Thesis 
Thesis no:  13227 
Thesis status:  Complete 
Number of Pages:  1 OnlineRessource (xvii, 150 Seiten) 
Language:  English 
Identification Number: 

edoc DOI:  
Last Modified:  21 Aug 2019 04:30 
Deposited On:  20 Aug 2019 11:18 
Repository Staff Only: item control page