Machine Learning Like a Physicist
Event details
| Date | 08.11.2018 |
| Hour | 16:00 › 17:00 |
| Speaker |
Prof. Michele Ceriotti Laboratory for Computational Science and Modeling, EPFL Lausanne |
| Location | |
| Category | Conferences - Seminars |
ChE-605 - Highlights in Energy Research seminar series
Statistical regression techniques have become very fashionable as a tool to predict the properties of systems at the atomic scale, sidestepping much of the computational cost of first-principles simulations and making it possible to perform simulations that require thorough statistical sampling without compromising on the accuracy of the electronic structure model.
In this talk I will argue how data-driven modelling can be rooted in a mathematically rigorous and physically-motivated framework, and how this is beneficial to the accuracy and the transferability of the model. I will also highlight how machine learning - despite amounting essentially at data interpolation - can provide important physical insights on the behavior of complex systems, on the synthesizability and on the structure-property relations of materials.
I will give examples concerning all sorts of atomistic systems, from semiconductors to molecular crystals [1], and properties as diverse as drug-protein interactions [2], dielectric response of aqueous systems[3] and NMR chemical shielding in the solid state [4].
[1] F. Musil, S. De, J. Yang, J. E. J. E. Campbell, G. M. G. M. Day, and M. Ceriotti, Chem. Sci. 9 (2018) 1289
[2] A. P. A. P. Bartók, S. De, C. Poelking, N. Bernstein, J. R. J. R. Kermode, G. Csányi, and M. Ceriotti, Sci. Adv. 3, (2017) e1701816
[3] A. Grisafi, D. M. Wilkins, G. Csányi, and M. Ceriotti, Phys. Rev. Lett. 120 (2018) 36002
[4] http://shiftml.
Statistical regression techniques have become very fashionable as a tool to predict the properties of systems at the atomic scale, sidestepping much of the computational cost of first-principles simulations and making it possible to perform simulations that require thorough statistical sampling without compromising on the accuracy of the electronic structure model.
In this talk I will argue how data-driven modelling can be rooted in a mathematically rigorous and physically-motivated framework, and how this is beneficial to the accuracy and the transferability of the model. I will also highlight how machine learning - despite amounting essentially at data interpolation - can provide important physical insights on the behavior of complex systems, on the synthesizability and on the structure-property relations of materials.
I will give examples concerning all sorts of atomistic systems, from semiconductors to molecular crystals [1], and properties as diverse as drug-protein interactions [2], dielectric response of aqueous systems[3] and NMR chemical shielding in the solid state [4].
[1] F. Musil, S. De, J. Yang, J. E. J. E. Campbell, G. M. G. M. Day, and M. Ceriotti, Chem. Sci. 9 (2018) 1289
[2] A. P. A. P. Bartók, S. De, C. Poelking, N. Bernstein, J. R. J. R. Kermode, G. Csányi, and M. Ceriotti, Sci. Adv. 3, (2017) e1701816
[3] A. Grisafi, D. M. Wilkins, G. Csányi, and M. Ceriotti, Phys. Rev. Lett. 120 (2018) 36002
[4] http://shiftml.
Practical information
- General public
- Free