AI in Russia Will Help Protect People From High-Power Laser Exposure
A new Russian algorithm is reshaping how scientists design materials capable of protecting human vision and sensitive equipment from high‑power laser radiation
Turning Limited Data Into Powerful Predictions
Russian researchers have developed an algorithm called CORRELATO that can rapidly and accurately predict the properties of new optical‑protective materials, even when only small datasets are available. The work was carried out by scientists from the Federal Research Center for Chemical Physics and Medicinal Chemistry of the Russian Academy of Sciences and MIET, with support from the Russian Science Foundation.
The key advantage of CORRELATO is its ability to function effectively with minimal data. While traditional machine‑learning methods require thousands of samples, CORRELATO generates interpretable mathematical relationships using only a few dozen examples.
In the study, the algorithm analyzed 24 specially synthesized phthalocyanine dyes used as optical limiters. CORRELATO predicted their performance with more than 95 percent accuracy and identified the parameters that determine how these materials function: polarizability, dipole moment, and charge‑transfer characteristics.
From Molecules to Nanomaterials
The algorithm was also successfully applied to 41 samples of complex nanomaterials—nanotubes, dyes, and hybrid structures. CORRELATO classified the materials into three effectiveness categories and formulated precise mathematical criteria that allow scientists to estimate a material’s potential before synthesis.
The materials created using these predictions can be used in protective filters for the eyes of pilots, surgeons, and specialists working with high‑power lasers, as well as for shielding expensive optical equipment. Researchers plan to expand CORRELATO’s applications into other areas of materials science, including organic electronics, photovoltaics, sensing, and catalysis.
Toward Self‑Improving Material‑Design Systems
According to Tolbin, the long‑term goal is to integrate CORRELATO into a full development cycle to create a self‑learning system capable of proposing new molecular structures, evaluating their properties, and updating models as new experimental data arrives.
The materials developed with the help of the algorithm will form the basis of next‑generation protection systems against high‑power laser beams. The universality of the method also opens the door to its adoption across a wide range of domains—from organic electronics to catalysis and sensing.
