10th Presentation | Coarse Grained Molecular Simulations Facilitated By Machine Learning | Eleonora Ricci
You can watch presentation’s rercording here!
NanoAI welcomes Eleonora Ricci who will present his work, titled “Coarse Grained Molecular Simulations Facilitated By Machine Learning”.
As always, an open discussion will follow.
When? November 3rd, 15:30 Greece
Description: Molecular simulations offer deep insight into the microscopic mechanisms underlying macroscopic behavior, supporting theoretical development and the discovery of structure-property relationships for a rational materials design. Coarse grained (CG) molecular simulations enable the investigation of molecular properties for larger systems and at longer timescales than the ones attainable at the atomistic resolution.
Machine learning (ML) techniques have been recently proposed to learn CG particle interactions, i.e. develop CG force fields. Graph representations of molecules and supervised training of a graph convolutional neural network architecture are used to learn the potential of mean force through a force matching scheme. A main advantage of ML models in the context of molecular simulations is that they are not constrained to a predefined mathematical function, therefore they are endowed with higher flexibility and expressive character compared to traditional CG models, allowing to potentially capture many-body interactions more accurately.
In this work, the force acting on each CG particle is correlated to a learned representation of its local environment that goes under the name of SchNet, constructed via continuous filter convolutions. We explore the application of SchNet models to obtain CG potentials for a liquid system, investigating the effect of model architecture and hyperparameters on the thermodynamic, dynamical, and structural properties of the simulated CG systems, reporting and discussing challenges encountered and future directions envisioned.
Eleonora Ricci obtained a BSc (2014) and a MSc (2016) in Chemical Engineering from the University of Bologna, Italy, and earned a PhD in 2020 from the same university, with a dissertation on ‘Thermodynamic and Molecular Simulation of Pure and Mixed Gas Sorption in Polymeric Membranes’. In 2021 she was a postdoctoral researcher and adjunct professor at the university of Bologna. Since November 2021 she is a Marie Curie Post-Doctoral Research Fellow at NCSR Demokritos, in the framework of the ML-MULTIMEM project, a multidisciplinary endeavour that aims at applying artificial intelligence tools in computer simulations of materials for carbon capture. The project is a synergy between the Molecular Thermodynamics and Modelling of Materials Laboratory (Dr. N. Vergadou) of the Institute of Nanoscience and Nanotechnology (INN) and the Software and Knowledge Engineering Lab (Dr. V. Karkaletsis, Dr. G. Giannakopoulos) of the Institute of Informatics & Telecommunications (IIT) of NCSR “Demokritos”.
Subscribe to our mailing list: https://lists.iit.demokritos.gr/mailman/listinfo/nanoai
You can find more info on this upcoming talk as well as on our previous talks at http://nano-ai.iit.demokritos.gr/