SLAI 8th Seminar will be discussing the topic on "Synergizing Physics and Data: AI with Structual Constraint", from 3pm to 5pm, November 4th (Tuesday) at B301 Classroom.

Speaker Biography: 

Denis Derkach graduated from the Saint-Petersburg State University in 2007 and later obtained a PhD in particle physics from the University of Paris 11. After postdocs at Istituto Nazionale di Fisica Nucleare and University of Oxford, he joined HSE, Moscow and currently is an assistant professor there. He is the head of the Laboratory of Methods for Big Data Analysis (LAMBDA), Faculty of Computer Science, HSE University. His main research interest concentrates around developing, adopting, and applying data science methods in the field of physics and related industry.

Abstract:

The rapid ascent of data science has revolutionized analytics across numerous domains. However, purely data-driven approaches often struggle with extrapolation, physical inconsistency, and a reliance on massive, high-quality datasets. Conversely, classical physics-based models provide a robust structural understanding but can be computationally expensive and sensitive to initial conditions and parameterization. This talk convers a synergistic merger of these paradigms in several fields, demonstrating that the integration of mechanistic models with modern machine learning leads to more accurate, efficient, and interpretable solutions for complex scientific and industrial problems. I will cover several examples ranging from weather prediction to industrial data anomaly detection. The possible applications are also discussed.