View Position
| Project Title | Grouping Seach Regions for Evolutionary Algorithm of Quantum Chemical Systems |
| Summary | The intern will use machine learning to group together intelligently created candidate structures into chemically reasonable groups. An evolutionary algorithm will then be applied to each of these groups where the result from each search will be used to determine the most favourable structure. It is expected that the partitioning of the search space around chemically relevant initial populations should lead to an increase in the rate at which the evolutionary algorithm locates the most favourable structure. |
| Job Description | A typical genetic algorithm runs in a loop where trial structures are selected from an exclusive population and are paired and/or mutated. Molecular configurations that are better will replace weaker candidates in the population. We have created an intelligent molecule creator which uses standard atomic orbital hybridization geometries to make randomly generated systems which are chemically relevant. It is now important to create a systematic search scheme which leverages these improved starting conditions. The student will use existing protocols in Python to make groups of intelligently created candidate structures in order to search within those groups for the global minimum. The successful intern will introduce the Covariance Matrix Adaptation Evolution Strategy in order to help find the local minimum within each group. A reasonable condition will be determined in order to conclude that the local minimum has been found so that the region can be eliminated from further examination. This leads to the evolutionary algorithm continually focusing on novel regions. During the test phase, where we will use a stoichiometry with a known global minimum, this systematic search scheme will move through each group until the global minimum is found. It is expected that this procedure should increase the rate of determining the global minimum structure over traditional evolutionary algorithm searches since candidate structures from the same group will pair better than pairings from different groups leading to faster evolution in each group. Additionally, eliminating groups from the search should limit examining duplicate trial structures. This new strategy will then be used on a variety of trial systems in order to determine a generic procedure which would work on a broad set of chemical systems. |
| Use of Blue Waters | To test the effectiveness of global optimisation search schemes, hundreds of runs comprised of thousands of trial structures must be calculated to get the statistics necessary to quantify improvements. Local resources here at CSUF will be used to prototype initial population generation protocols. The resources at Blue Waters will be used for promising candidates to parallelize all the necessary runs to get relevant statistical accuracy to benchmark improvements. Also the parallelization available at Blue Waters will enhance the chemical accuracy of the quantum mechanics based calculations as well as enhance the clustering algorithm's ability to categorize structures. This will be useful when the improved code is applied to new, catalytically relevant systems to determine important chemical properties. |
| Conditions/Qualifications | Must be an undergraduate at CSUF Must have Python programming experience |
| Start Date | 05/31/2018 |
| End Date | 05/31/2019 |
| Location | Groves Research Group Department of Chemistry and Biochemistry California State University, Fullerton Fullerton, CA |
| Interns | |