About me

I am an optimization and multi-agent systems researcher working at the intersection of distributed intelligence, networked control, and adaptive learning.

My research develops mathematically grounded algorithms for distributed optimization, coordination, estimation, and decision-making in systems that operate under uncertainty, nonstationarity, communication constraints, noisy measurements, and limited feedback. This includes stochastic approximation, zeroth-order optimization, SPSA-type methods, graph-based coordination, and control algorithms for cyber-physical and multi-robot systems.

I received my PhD in Mathematical Cybernetics from Saint Petersburg State University, where my dissertation focused on multi-agent observer control in sensor networks. Since then, my work has developed along three connected directions: optimization under uncertainty, distributed coordination in networked systems, and scalable control of multi-agent and robotic systems.

I have served as Principal Investigator and research lead on funded projects related to zeroth-order optimization, distributed optimization, accelerated algorithms, and networked control. My current research agenda focuses on safe and communication-efficient zeroth-order optimization for control and learning in networked systems.

VERA — Victoria Erofeeva Research — is my research umbrella for work on adaptive networked intelligence: optimization, learning, and coordination methods for intelligent systems composed of many interacting agents.

My publications include work in control, optimization, networked systems, and multi-agent coordination, with papers in venues such as IEEE Transactions on Automatic Control, Automatica, Information Sciences, Scientific Reports, IEEE Access, and major systems-and-control conferences.

I also supervise and mentor early-career researchers. My current PhD supervision focuses on meso-scale control of robot teams, including networked coordination algorithms and experimental validation on robotic testbeds. Previous mentorship has involved graduate and PhD students working on stochastic approximation, distributed optimization, multi-agent systems, and networked control.

Research focus

Adaptive learning and efficient optimization
Stochastic approximation, zeroth-order optimization, SPSA-type methods, nonstationary optimization, and learning under limited, noisy, or delayed feedback.
Distributed intelligence and multi-agent systems
Consensus algorithms, distributed optimization, graph-based coordination, decentralized clustering, task allocation, and decision-making in large-scale networked systems.
Networked control and multi-robot autonomy
Meso-scale control, networked coordination, robotic teams, formation control, cyber-physical systems, and experimental validation on robotic testbeds.

Current interests

I am especially interested in how rigorous optimization and control methods can support adaptive distributed intelligence in systems where gradients are unavailable or expensive, communication is constrained, and decisions must be made online from imperfect feedback.

This includes applications to robotic teams, networked AI systems, cyber-physical infrastructures, LLM-agent coordination, and efficient model adaptation.

Contact

For research collaboration, invited talks, supervision, or academic and R&D opportunities, please contact me by email or through the links below.