Register
M
Professor profile

Matus Telgarsky

New York University · Computer Science

About
Regular biography

Matus Telgarsky is an Associate Professor in the Department of Computer Science at New York University. His research focuses on deep learning theory and machine learning theory. He has contributed to the understanding of deep networks, margin-based generalization, and gradient descent convergence. Telgarsky has taught courses on deep learning theory and machine learning, including at UIUC and NYU. His work includes studies on coordinate descent methods, inapproximability of deep networks, and the analysis of gradient descent for non-separable data. He maintains an active research profile and shares course materials and lecture notes online.


Scholar profile summary
Scholar-generated biography

Matus Telgarsky is a researcher at the Courant Institute of Mathematical Sciences, New York University, focusing on deep learning theory and machine learning theory. His work explores the theoretical foundations of neural networks, including their generalization capabilities, optimization properties, and representation power. Key areas of interest include margin bounds, implicit bias, and the role of depth in neural networks. His research also examines stochastic gradient methods, regularization paths, and the interplay between non-convex optimization and learning. Telgarsky's publications highlight the importance of spectral normalization, tensor decompositions, and the alignment of layers in deep learning. His contributions provide theoretical insights into the behavior of deep models and their performance in practical settings.

Source: google_scholar · 111 words
Related professors