Elina Robeva

Elina Robeva is an Assistant Professor in the Department of Mathematics at the University of British Columbia, Faculty of Science. She is an active researcher in algebraic statistics, tensor decomposition, and applied algebraic geometry, with a focus on structured tensor models and their statistical and computational properties.

Her research lies at the intersection of algebra, statistics, and data science. She investigates mathematical structures in high-dimensional data, particularly through the lens of tensor decompositions that generalize matrix factorizations. Her work includes studying orthogonally decomposable tensors, symmetric tensor eigenvalues, and connections to algebraic varieties and optimization methods such as the tensor power method. She also explores extensions to incoherent and frame-based tensor models with applications in machine learning and signal processing.

The available article from her 2021 seminar talk indicates a strong trend in theoretical and computational aspects of tensor analysis, emphasizing algebraic structure, spectral properties, and algorithmic approaches to decomposition. Her work bridges abstract algebra with practical data modeling challenges.

While no specific awards or students are listed in the provided text, her invited seminar presentation reflects recognition in the mathematical sciences community. She is involved in advancing the theoretical foundations of statistical learning through algebraic methods.

Elina Robeva contributes to the academic mission through research and scholarly dissemination, collaborating across mathematical disciplines to develop new tools for understanding complex data structures.