HSE AI Research Centre Simplifies Particle Physics Experiments
Scientists at the HSE AI Research Centre have developed a novel approach to determining robustness in deep learning models. Their method works eight times faster than an exhaustive model search and significantly reduces the need for manual verification. It can be applied to particle physics problems using neural networks of various architectures. The study has been published in IEEE Access.
Machine learning (ML) and artificial intelligence (AI) are increasingly used in particle physics to make the analysis of experimental data easier and faster. Neural networks, for example, help process instrument signals and reconstruct missing information about particle properties. Since such predictions influence subsequent analysis, it is essential to understand how robust they are. However, in practice, model accuracy is often the only metric evaluated, and little attention is paid to how much the results may vary across different training instances. This issue is particularly pronounced for deep neural networks: their behaviour is hard to interpret, and repeated training can produce noticeably different outcomes. As a result, despite their potential advantages, many physicists remain wary of using neural networks.
Scientists at the HSE AI Centre have proposed a novel solution. They developed a method that automatically compares dozens of neural network variants and identifies the most reliable and stable ones. The idea is as follows: if a model is trained repeatedly on slightly modified data with different initial weights, the distribution of errors reveals how robust the model is to small changes in conditions. A robust model will produce nearly identical results across these tests.
The researchers tested their method on a task in which, based on an image formed by the cells of an electromagnetic calorimeter, one must determine the energy of a particle and the point at which it struck the detector. An electromagnetic calorimeter is a device made up of many cells that measure the amount of energy deposited in each cell when a particle passes through.
Fedor Ratnikov
'For the analysis, we generated half a million virtual signals simulating the detector’s operation and repeatedly fed them into different models, each time changing the training and test samples. We then used this method to identify the most reliable models and examine their properties. In doing so, we determined the minimum amount of training data required for a model to become robust—that is, to perform consistently across different training runs,' comments Fedor Ratnikov, Leading Research Fellow at the Laboratory of Methods for Big Data Analysis of the HSE AI and Digital Science Institute.
A key element of the approach is a special selection algorithm. For each model variant, the researchers collected a set of errors accumulated over dozens of independent runs and used this distribution to estimate how predictably the model behaves. This makes it possible to automatically filter out models that performed well only by chance and to identify those that remain stable under any reasonable changes in conditions.
Alexey Boldyrev
'We repeatedly trained all the models on half a million calorimeter simulation events, each time splitting the data into new training and test sets and initialising the weights randomly. This allowed us not only to measure how often each model made mistakes but also to track how its learning behaviour changed from one run to the next,’ comments Alexey Boldyrev, Research Fellow at the Laboratory of Methods for Big Data Analysis of the AI and Digital Science Institute.
The study also showed that models supplied not only with raw signals but also with simple, pre-known physical values require less data and reach stable results more quickly. The authors estimated the minimum amount of data needed for such models to maintain consistent performance across runs and identified two architectures that were reliably accurate and robust.
Andrey Shevelev
'The new method allows for much faster selection of robust AI models for certain particle physics problems—achieving results eight times more quickly than the traditional exhaustive search of all model variants,' comments Andrey Shevelev, Research Assistant at the Laboratory of Methods for Big Data Analysis of the AI and Digital Science Institute.
The researchers emphasise that the algorithm is fully automated and does not require manual tuning. As a result, it can serve as the foundation for self-learning systems that operate robustly, regardless of fluctuations in the training data or inherent model limitations.
See also:
HSE Develops App for Assessing Phonological Processing in Children
Researchers at the HSE Centre for Language and Brain have developed a new digital tool for assessing children's phonological processing skills—the ZARYA (Sound Analysis of the Russian Language) test battery. It is the first standardised application in Russia designed to provide a fast and reliable assessment of children's ability to distinguish speech sounds, retain them in working memory, and perform phonemic analysis. The app runs on Android tablets and smartphones and is available for download from RuStore. Details of the test validation have been published in the Journal of Speech, Language, and Hearing Research.
Researchers Discover How Spelling Errors Slow Down Reading in Russian
Psycholinguists from the Centre for Language and Brain at HSE University–St Petersburg have shown that words that are frequently misspelled are processed more slowly by readers, even when presented with the correct spelling. The researchers confirmed this effect for the first time using Russian-language materials and found that response speed is most strongly linked to how confidently individuals can distinguish the correct spelling of a word from an incorrect one. The study has been published in The Mental Lexicon.
Scientists Discover Why Europium 'Misbehaves'
Europium is a rare-earth metal responsible for the pure red glow in displays and other luminescent materials. For a long time, however, it refused to emit light when surrounded by certain organic molecules known as acylpyrazolone ligands. Chemists have now uncovered the reason: in europium complexes with these ligands, a 'black window' appears—a charge-transfer state in which the energy absorbed by the ligand is dissipated as heat rather than emitted as light. Understanding this mechanism opens the way to designing more efficient red-emitting materials for displays, fluorescent thermometers, and chemical sensors. The results have been published in Dalton Transactions.
HSE Economists Reveal How the Wage Gap Emerges Among Vocational School Graduates
HSE researchers examined the careers of 600,000 graduates of Russian secondary vocational education programmes and found that at the start of their careers, the gender wage gap reaches 23%, doubling after three years. This disparity is largely due to male and female students choosing different occupations when enrolling in vocational schools. These were the findings made by Sergey Roshchin, Natalya Yemelina, and Ksenia Rozhkova from of the HSE Faculty of Economic Sciences. The article has been published in Educational Studies.
HSE Researchers Make Aldehydes Perform Dual Function
Chemists from HSE University have discovered a way to carry out a reductive addition reaction without using an external reducing agent. Instead, the required 'resource' is supplied by the aldehyde itself, one of the reaction participants. This approach helps prevent unwanted side reactions, reduces toxicity, and simplifies the production and synthesis of organic molecules, including those used in the manufacture of medicines. The study has been published in Journal of Catalysis.
HSE Scientists Explain Why Findings in Autism Research Differ
Researchers from the Cognitive Health and Intelligence Centre at HSE University conducted the first-ever systematic review of studies on the specifics of emotion-from-motion perception in autism. The review showed that differences found between autistic and non-autistic individuals are largely associated with the experimental design and the types of tasks given to study participants. The review findings have been published in Research in Autism.
Tremors: Scientists Develop Method for Real-Time Tracking of Hazardous Underground Vibrations
Researchers from HSE MIEM and IPKON RAS have developed a new mathematical monitoring model that can identify the source of hazardous underground vibrations in real time. The technology could help reduce the risk of damage to buildings, roads, and other infrastructure located near quarries and mining sites. The paper has been published in Russian Mining Industry.
HSE Researchers Determine Which Internet Users Are More Likely to Fact-Check
Researchers at HSE University examined the strategies employed by Russian internet users to verify unreliable information and the factors that motivate them to do so. The study found that more than half of users who encounter potentially false information online attempt to verify it by locating the original source. The likelihood of fact-checking is influenced by several factors, including age, place of residence, social status, information literacy skills, and the use of AI. The findings have been published in Monitoring of Public Opinion: Economic and Social Changes.
Tabular Data Anonymisation Solution for Safe Use in AI Systems Developed at HSE University
The AI and Digital Science Institute at the HSE Faculty of Computer Science has developed a tabular data anonymisation service designed to prepare corporate datasets for use in analytics and AI applications. The solution can identify personal data in structured datasets, apply consistent and reproducible anonymisation rules, and generate the artifacts required for quality control, auditing, and subsequent use of data in secure environments.
Population Lifespan Is Governed by Mathematical Laws
Researchers at HSE University and MSU have established a universal law governing the time to extinction of a population in a random environment. Their analysis of the evolution of branching processes—complex probabilistic systems—shows that, regardless of the initial population size, extinction follows strict mathematical laws. The results have been published in the Journal of Applied Probability.