The HSE Laboratory of methods for Big Data Analysis won the Mirror Laboratories project competition. In collaboration with the University of Innopolis, the National University of Singapore, we are working on the project "Machine Intelligence Methods Development and Application for Effective Search for New Materials".
Project manager Andrey Ustyuzhanin. 

Project description

Modern materials are mainly determined by their properties, which are independent of time and environment. However, adaptive materials with memory functions and properties that change under the influence of environmental feedback and control signals are very useful for new technologies. Such materials can monitor external conditions and adapt their functionality to a new set of conditions through built-in feedback. A large subclass of such functional materials are materials out of equilibrium, which can change their conformation, producing useful work, feeding on the energy of the environment.
Such materials are called functional materials (FM). They are absolutely essential for building von Neumann (neuromorphic) computers, machine-to-human interfaces, artificial organs and tissues, smart membranes, smart batteries and catalysts, to name just a few. The development and industrial production of FM require accurate modeling of the functional and non-equilibrium properties of material systems. Thus, such models will allow the synthesis of intelligent and adaptive structures with built-in memory and control capabilities.

Targets and goals
Development of machine intelligence models to accelerate the search for new materials with desired properties.

• Creation of a structured database (DB) for searching experimental / simulation records for different versions of material structures;

• Development and implementation of algorithms and tools for dynamic machine intelligence that combine experimental database data with computer simulation to solve direct and inverse synthesis problems;
• Using real-world experiments and simulations using machine intelligence-based approaches and dynamic surrogate models;
• Development and implementation of optimization algorithms for the synthesis process, taking into account additional experimental data and simulation results;

• Application of the developed algorithms for the production of new materials with specified characteristics.