Ensemble-Average Representation of Pt clusters in Conditions of Catalysis, accessed through GPU Accelerated Deep Neural Network Fitting Global Optimization Anastassia Alexandrova University of California, Los Angeles (UCLA) We first report a global optimization approach based on GPU accelerated Deep Neural Network (DNN) fitting. The seven-layer multi-dimensional and locally connected DNN is combined with limited-step Density Functional Theory (DFT) geometry optimization to reduce the time cost of full DFT local optimization, which is considered to be the most time-consuming step in global optimization. An algorithm based on bond length distribution analysis is used to efficiently sample the configuration space and generate random initial structures. A structure similarity measurement method based on depth-first search is used to identify duplicates. The performance of the new approach is examined by the application to the global minimum searching for Pt9 and Pt13. The averaged statistical mechanical ensemble properties of these clusters at elevated T of catalysis are then evaluated, and shown to be significantly different from the properties of just the global minimum on the potential energy surface.