Date and Time: 24th January 2023 (Tuesday) at 11:00 am (Tea: 10:45 am)
Place: Nevill Mott Hall, JNCASR
Title: Machine Learning for Molecular Simulation
Abstract:
In this talk, I will begin by briefly explaining the rationale for utilizing machine learning in molecular simulation. I will then showcase the application of the state of the art machine learning techniques in two of our recent projects: (i) quantifying the dynamic acceleration [1] in coarse-grained simulations , and (ii) predicting the synthesis condition of the metal-organic frameworks. [2] While the machine learning model performed well in the first case, it unfortunately had poor performance in the second task. However, we were surprised to find that even this subpar machine learning model outperformed the human predictions, highlighting the potential utility of machine learning in this context.
[1] Bag, Saientan, Melissa K. Meinel, and Florian Müller-Plathe. "Toward a Mobility-Preserving Coarse-Grained Model: A Data-Driven Approach." Journal of Chemical Theory and Computation (2022), 18, 12, 7108–7120
[2] Luo, Yi, Saientan Bag, Orysia Zaremba, Adrian Cierpka, Jacopo Andreo, Stefan Wuttke, Pascal Friederich, and Manuel Tsotsalas. "MOF Synthesis Prediction Enabled by Automatic Data Mining and Machine Learning." Angewandte Chemie International Edition 61, no. 19 (2022): e202200242.
