Data science and artificial intelligence are revolutionizing our world, and materials science is no exception. In the past, the ability to perform extensive quantum chemical simulations has accelerated the pace of materials design[1]. In the near future, we may overcome the need for resource-intensive computations by leveraging existing data. By combining data mining with intelligently crafted machine learning (ML) algorithms, we can design new materials with tailored properties more efficiently than ever before [2].
This talk will explore data-driven approaches to designing and investigating materials. It will be structured in two parts: first, a general overview of big data and machine learning techniques in materials science; and second, a showcase of our platform for semiconductor materials discovery.
The first part will start with a pedagogical introduction on the general concept of ML. This will be followed by a discussion on materials databases and on the main applications of ML in materials science.
The second part of the talk will focus on the computational platform for data-driven semiconductors design developed by our group. Our platform: i) integrates the major materials databases into a unified interface, and ii) leverages various ML methods to predict material properties based on chemical composition. While containing features tailored for semiconductors, the platform is adaptable to any type of material. It aims to accelerate materials discovery through efficient screening.

    1 Curtarolo, S., Hart, G. L., Nardelli, M. B., Mingo, N., Sanvito, S., & Levy, O. (2013). The high-throughput highway to computational materials design. Nature materials, 12(3), 191-201.  
    Keith, J. A., Vassilev-Galindo, V., Cheng, B., Chmiela, S., Gastegger, M., Muller, K. R., & Tkatchenko, A. (2021). Combining machine learning and computational chemistry for predictive insights into chemical systems. Chemical reviews, 121(16), 9816-9872.

Gabriele Saleh, ricercatore presso l’Istituto Italiano di Tecnologia, ha conseguito il dottorato in Scienze Chimiche presso l’Università degli Studi di Milano nel 2014, a seguito dei suoi studi sulla cristallografia e il legame chimico.  
Successivamente, la sua attività di ricerca si è concentrata sulla scienza dei materiali computazionale, toccando una vasta gamma di argomenti come la chimica planetaria, la corrosione e, più recentemente, la nanoscienza.  
Ha inoltre maturato un'esperienza diversificata lavorando in numerosi istituti di ricerca in Europa e Russia (Università di Milano, Università di Aarhus, Istituto di Fisica e Tecnologia di Mosca, Trinity College di Dublino, Istituto Italiano di Tecnologia), e collaborando a progetti con laboratori industriali come Nokia Bell Labs e Siemens.  
È autore di articoli scientifici pubblicati su riviste prestigiose quali Nature Chemistry, Advanced Materials e Angewandte Chemie.