My background consists of 2 PhDs, the first in applied mathematics (“Methods of mathematical and imitational modelling of local interaction processes in transportation systems.”) and the second in informatics
(“Machine Learning for the distributed and dynamic management of a fleet of autonomous taxis and shuttles”). My research interests lie in optimisation of complex systems, game theory, as well as reinforcement learning.
I prefer real-world applications of my knowledge, ranging from games to big transportation networks.
Traffic flows are the easiest application of game theory. My appreciation for optimisation in transportation systems began with the Braess’s paradox, which is linked to the concept of user equilibrium in road networks. After applying methods of mathematical and numerical optimisation during my Master’s thesis and first PhD thesis, I moved to machine learning methods of optimisation in big systems with autonomous vehicles. It was an attempt to develop artificial intelligence in this system, and at the same time use a game theory approach (the attempt to move towards system optimum, or to approach the Pareto front), as well as combinatorial optimisation, multiagent reinforcement learning (with different levels of granularity), etc.
(“Machine Learning for the distributed and dynamic management of a fleet of autonomous taxis and shuttles”). My research interests lie in optimisation of complex systems, game theory, as well as reinforcement learning.
I prefer real-world applications of my knowledge, ranging from games to big transportation networks.
Traffic flows are the easiest application of game theory. My appreciation for optimisation in transportation systems began with the Braess’s paradox, which is linked to the concept of user equilibrium in road networks. After applying methods of mathematical and numerical optimisation during my Master’s thesis and first PhD thesis, I moved to machine learning methods of optimisation in big systems with autonomous vehicles. It was an attempt to develop artificial intelligence in this system, and at the same time use a game theory approach (the attempt to move towards system optimum, or to approach the Pareto front), as well as combinatorial optimisation, multiagent reinforcement learning (with different levels of granularity), etc.
After defending my first thesis in 2016, I was offered a job from Institut VEDECOM in France. As part of my work, I developed and implemented optimisation algorithms for a fleet of autonomous taxis (redistribution of empty vehicles, ride-sharing). From 2021 to 2025 I've worked at RATP Smart Systems as a Specialist in Operational Research, where my main goal was to develop new methods of multimodal trajectory searches in public transport networks.
My love for studying and sharing knowledge with others also led me to writing several books about Olympiad mathematics and some scientific articles listed in my CV. I supervised 2 Master’s students for their thesis in the game theory field (“Methods for resolving the Braess’s paradox in the presence of autonomous vehicles” and “ Anti-coordination games with relations structure”) as well as several Bachelor’s and Master’s theses in the fields of optimisation and machine learning.
My love for studying and sharing knowledge with others also led me to writing several books about Olympiad mathematics and some scientific articles listed in my CV. I supervised 2 Master’s students for their thesis in the game theory field (“Methods for resolving the Braess’s paradox in the presence of autonomous vehicles” and “ Anti-coordination games with relations structure”) as well as several Bachelor’s and Master’s theses in the fields of optimisation and machine learning.
