Konstantin Gubaev

Ph.D. thesis: “Machine-learning Interatomic Potentials for Multicomponent Alloys”

Supervisor: Alexander Shapeev 

IDC committee members: Alexander Shapeev , Artem Oganov​, Andriy Zhugayevych

Konstantin graduated from Moscow Institute of Physics and Technology in 2013 (BSc, devoted to nonlinear elasticity of liquid crystals) and 2015 (MSc, devoted to numerical simulation of hydrodynamics and of heat and charge transfer during hybrid laser-arc spot welding process). His research at Skoltech is focused on developing the machine-learning interatomic potentials (MLIPs) for multicomponent systems. These types of potentials are surrogate models interpolating the quantum mechanical data (energies, forces and stresses for atomic systems). MLIPs developed by Konstantin allow performing molecular dynamics or structure relaxations orders of magnitude faster than with ab initio methods, with an accuracy comparable to the accuracy of reference data provided. The MLIPs were verified on the tasks of predicting the properties of organic molecules and were also used for calculation of elastic constants, phase diagrams and convex hulls for metallic alloys. His research interests include: phase stability in high-entropy alloys, finite-temperature properties of alloys, atomistic simulations.

List of papers:

1. Gubaev, K., Podryabinkin, E. V., Hart, G. L., & Shapeev, A. V. (2019). Accelerating high-throughput searches for new alloys with active learning of interatomic potentials. Computational Materials Science, 156, 148-156.​
2. Gubaev, K., Podryabinkin, E. V., Hart, G. L., & Shapeev, A. V. (2019). Accelerating high-throughput searches for new alloys with active learning of interatomic potentials. Computational Materials Science, 156, 148-156.

List of conferences:

1. “Machine learning interatomic potentials for multicomponent systems” at Foundationsof Atomistic Multiscale Modeling and Simulation (August 2017, Abkhazia)
2. “Machine learning interatomic potentials for multicomponent systems” at 2nd PhysicsInformed Machine Learning Conference (February 2018, USA)
3. “Machine learning interatomic potentials for multicomponent systems” at Solid State Chemistry and Functional Materials (May 2018, Russia, St. Petersburg)
4. “Машинное обучение и многомасштабное моделирование” – II Всероссийская научно-практическая конференция молодых ученых (October 2017, Russia, Moscow)
5. “Accelerating high-throughput searches for new alloys with active learning of interatomic potentials”  – DPG Spring Meeting (April 2019, Germany)