Mihai received his PhD in Applied and Computational Mathematics (PACM) at Princeton University in 2012, supervised by Amit Singer. His thesis was on the low-rank matrix completion problem and several distance geometry problems with applications to sensor network localization and three-dimensional structuring of molecules. During 2013-2016 he was a CAM Assistant Adjunct Professor in Computational Applied Mathematics at UCLA. During Fall 2014, he was a Research Fellow at the Simons Institute for Theory of Computing at UC Berkeley, in the program Algorithmic Spectral Graph Theory.
Mihai's research interests concern the development and mathematical analysis of algorithms for large networks, certain inverse problems on graphs, and big data analysis, with applications to various problems in engineering, machine learning, finance, and biology. Particular areas of interest are spectral and SDP-relaxation algorithms and applications, the group synchronisation problem, ranking from noisy pairwise comparisons, lead-lag relationships in multivariate time series, clustering, core-periphery structure in networks, multiplex networks, dimensionality reduction and diffusion maps (with an eye towards heterogeneous data and nonlinear time series), spectral algorithms for analysis of signed graphs and correlation networks. The above problems share an important feature: they can all be solved by exploiting the spectrum of their corresponding graph Laplacian.