A new seminar in the Machine Learning Meets Materials Science series is coming up

We are thrilled to announce the next installment in the Machine Learning Meets Materials Science seminar series, led by WISE-affiliated researcher Maryna Pankratova from Uppsala University. This interesting initiative, part of a WISE-WASP collaboration, fosters interdisciplinary dialogue and synergy between the cutting-edge fields of materials science and machine learning.

Each seminar in the series provides a unique platform for experts from both domains to share insights and explore innovative ideas. However, the upcoming fifth seminar will feature a solo presentation by Igor Abrikosov, Professor from Linköping University and WISE-affiliated researcher. In his presentation, Prof. Abrikosov will discuss capabilities of state-of-the-art theoretical simulations combined with experiment, artificial intelligence (AI) and machine learning (ML) to disclose materials functionalities attractive for sustainable applications.

Don’t miss this opportunity to gain fresh perspectives, learn foundational AI concepts, and connect with leading minds in materials science and machine learning. Join us as we explore how these dynamic fields intersect to shape the future of science and technology!

Seminar Title: Combining theory, experiment and AI in materials design: new opportunities for sustainable world.
Date: May 8th, 2025
Time: 10:00 AM (CET)
Location: Online via Zoom https://uu-se.zoom.us/j/62054045821

Meeting ID: 620 54045821

Abstract:

 Breakthrough discoveries of novel materials with advanced functionalities enable materials science to participate in a transition towards a sustainable society and contribute to a solution of UN Sustainable Development Goals (SDG). Unfortunately, it still takes up to 20 years and longer to design materials. To reduce the time, we need to find non-conventional paths and develop approaches that go beyond state-of-the-art materials design paradigms. In this talk we demonstrate capabilities of state-of-the-art theoretical simulations combined with experiment, artificial intelligence (AI) and machine learning (ML) to disclose materials functionalities attractive for sustainable applications. We review recent advances in theoretical description of materials properties, strengthened with AI/ML tools [1] and advanced visual exploration [2]. Focusing on solid state spin qubits, we present results of systematic theoretical exploration of point defects in wide band gap semiconductors [3]. We identify materials systems with properties attractive for quantum computing and communications [4] and present a strategy of using SiC defects qubits to design quantum reservoir computing (QRC) systems, a new disruptive technology with a potential to achieve qualitative improvements in speed and reduction in power consumption – two or more orders of magnitude – compared to classical machine learning systems. Next, we demonstrate the capability of a collaboration between experiment and theory to discover materials with in high-pressure high-temperature (HPHT) synthesis at TPa compression and temperature above 2000 K [5] and show that theory guided decompression allows to quench phases with advanced functionalities to ambient conditions [6].

[1] B. Mukhamedov, F. Tasnádi and I. A. Abrikosov, Mater. Des. 253, 113865 (2025); H. Levämäki, et al., NPJ Comp. Mater. 8, 17 (2022).

[2] M. Bykov, et al., Phys. Rev. Lett. 126, 175501 (2021); D. Laniel, et al., Nature Chem. 15, 641 (2023).

[3] V. Ivády, et al., Npj Comp. Mater. 4,76 (2018); J. Davidsson, et al., Comput. Phys. Commun 269, 108091 (2021).

[4] O. Bulancea-Lindvall, et al., Phys. Rev. B 108, 224106 (2023); J. Davidsson, et al., Npj Comp. Mater. 10, 109 (2024).

[5] L. Dubrovinsky, et al., Nature 605, 274 (2022).

[6] D. Laniel, et al., Adv. Mater. (2023); Adv. Funct. Mater. 2416892 (2024).

Profesor Abrikosov’s research at WISE: https://wise-materials.org/project/theoretical-modelling-of-alloys-for-rare-earth-free-high-performance-permanent-magnets/

For more information on Maryna Pankratova’s research in WISE visit:

Explainable machine learning for magnetic interactions extraction