SC²S Colloquium - March 9, 2016

From Sccswiki
Revision as of 12:05, 22 February 2016 by Bakhtiar (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
Date: March 9, 2016
Room: 02.07.023
Time: 3:00 pm, s.t.

Vinod Rajendran: Deep Learning for the Prediction of Molecular Properties

The accurate and efficient prediction of molecular properties throughout chemical compound space is a critical ingredient towards rational compound design in chemical and pharmaceutical industries. Since the recent past, Machine Learning (ML) techniques applied to ab initio calculations i.e. nuclear charges and Cartesian coordinates of all atoms, have been proposed as an efficient approach for predicting quantum-mechanical molecular properties directly from the raw molecular geometry. This dexterous approach results from a combination of multiple factors such as the choice of ML model and an appropriate representation of the physical properties and invariance structure of molecules. In this work, a number of well established deep learning methods have been evaluated for predicting the molecular properties. In addition, the influence of these methods on various molecular representations are examined. The results of this evaluation show that deep learning methods can be successfully applied for accelerating the chemical accuracy of molecular properties. The application of deep learning methods on quantum-chemical systems especially for the purpose of molecular properties prediction, is a general direction that is well worth for pursuing a future research.

Vladimir Daskalov: Integration of a Structure Optimization Method in a Fluid Optimization Engine

A subject of a great interest nowadays is how to improve the tools employed in the product design-process and thus improve the results delivered by this process. More and more industry projects are depending on multiple commercial software suites in order to achieve the project’s predefined goals. Some of these projects are combining optimiza- tion techniques and tools from several different domain fields. Although this approach might lead to great improvements the complexity and interconnectivity between sepa- rate software tools that it implies leads to compatibility, reliability and usability issues. Furthermore to conduct such a multi-domain optimization a broad optimization domain- specific know-how is required. Reflecting those factors such optimization scenarios are still fairly inaccessible to most of the small scale development groups. The purpose of the present master’s thesis is to examine the employed workflows of the commercial products SIMULIA TOSCA Structure and SIMULIA TOSCA Fluid and to propose an approach to conduct a simple structure-topology optimization using SIMULIA TOSCA Fluid as a base framework extended by the required components from SIMULIA TOSCA Structure. The proposed workflow can be than used as a base for further research on possible combined fluid-structure non-parametric optimization.