Alexander Korotin

Alexander Korotin is an Assistant Professor at the Skolkovo Institute of Science and Technology (Skoltech) where he heads the Generative AI research group. He is also a senior research scientist at the Artificial Intelligence Research Institute (AIRI), leading the "Foundations of Generative AI" group. His academic journey includes a PhD in Math & Physics from Skoltech (2023), an MSc in Computer Science from the Higher School of Economics (HSE), and a BSc in Mathematics also from HSE.

Dr. Korotin's research focuses on generative modeling, with particular emphasis on developing novel algorithms based on Optimal Transport and Schrodinger Bridges. His work bridges theoretical mathematics with practical machine learning applications, contributing significantly to the field of generative artificial intelligence. He has pioneered approaches to make Schrodinger Bridge solvers more efficient and practical, most notably with his "Light Schrödinger Bridge" framework that simplifies complex computational procedures while maintaining theoretical rigor.

His publication record shows a clear progression toward making advanced generative modeling techniques more accessible and computationally efficient. Recent work demonstrates increasing sophistication in handling complex distribution matching problems through physics-inspired approaches (like electrostatic field matching) and novel distillation techniques that accelerate inference. The research spans from theoretical foundations to practical applications in image processing, semi-supervised learning, and reinforcement learning.

Dr. Korotin has received recognition for his contributions to neural optimal transport and Schrodinger Bridges, with his papers frequently appearing in premier machine learning venues. His work on efficient computational methods for optimal transport has established him as a rising expert in these specialized areas of machine learning.

As an academic leader, Dr. Korotin advises research students and collaborates extensively with colleagues across institutions, contributing to the advancement of generative AI through both theoretical developments and practical implementations. His work continues to push the boundaries of what's possible in generative modeling, with recent publications focusing on making advanced mathematical approaches more computationally efficient for real-world AI applications.