Hey Rajbanul,

I highly recommend you to check out work from Michele Ceriotti's group. He's from EPFL and quite productive in the field of ML in atomistic simulations.

Here are a few example publications to check out:

Article Unsupervised machine learning in atomistic simulations, betw...

Article Machine Learning Unifies the Modelling of Materials and Molecules

Article Transferable Machine-Learning Model of the Electron Density

Article Machine Learning for the Structure-Energy-Property Landscape...

Cheers

Tim

Thanks, Tim.

Article Deep learning: Extrapolation tool for ab initio nuclear theory

Rajbanul Akhond hope this will help you

Check this out:

https://www.ctcms.nist.gov/jarvisml/

Hi Rajbanul,

For understanding the context and methods of data-driven approaches in materials science, I recommend our recent review article Article From DFT to Machine Learning: recent approaches to Materials...

The discussion presents:

• A general introduction presenting why machine learning and data-driven approaches are being increasingly used
• The fundamentals (how) of both density functional theory (DFT) –including the recent trend of high-throughput calculations for databases creation– and machine learning methods
• Examples (what) of machine learning applications in materials science

From there you will be able to further explore the literature for specific topics of your interest, such as atomistic machine learning potentials (force fields), property predictions, deep learning of Hamiltonians, and others.

I hope you find this article useful!