Inverse Design of Simple Liquids using Machine Learning and the Ornstein-Zernike Equation

Add to your list(s) Download to your calendar using vCal

• Rhys Goodall
• Tuesday 14 April 2020, 16:30-17:00
• virtual ZOOM meeting ID: 263 591 6003, https://zoom.us/j/2635916003.

If you have a question about this talk, please contact Bingqing Cheng .

The Ornstein-Zernike framework provides an elegant route for solving the inverse problem of determining a pairwise interaction potential for a simple liquid given its structure. However, in order to realise the potential of the formalism superior closure relationships are required. Current approximate closure relationships have been shown to have restricted universality and give rise to thermodynamic inconsistencies. In this work rather than attempting to analytically derive a new closure relationship we return to the point of the approximation and investigate whether machine learning can be used to infer a universal closure for the framework directly from simulation data. We show that this is a fruitful approach that allows for improved inversion performance.

This talk is part of the Machine learning in Physics, Chemistry and Materials discussion group (MLDG) series.

This talk is included in these lists:

• Hanchen DaDaDash
• Lennard-Jones Centre external
• Machine learning in Physics, Chemistry and Materials discussion group (MLDG)
• virtual ZOOM meeting ID: 263 591 6003, https://zoom.us/j/2635916003

Note that ex-directory lists are not shown.