Research

In recent past, my research has focused on the following themes, as reflected in my publications and grants:

Modeling high-dimensional time series: The topics of interest include connections to networks, multiple-subjects high-dimensional time series, discrete-valued high-dimensional time series, and others. The data for which the methods are explored and developed come from Economics and Finance, Psychology, Neuroscience, and other areas.

Extremes and uncertainty quantification in physical systems: The topics of interest include multi-fidelity (semi-supervised learning) methods, reduced-order models, use of neural networks, and others. This research direction has been pursued in collaboration with researchers in the US Navy, especially around ship motions and other processes.

Sampling and streaming algorithms in connection to “big data”: The topics of interest include sampling and sketching in the settings of complex networks and communication networks. Most recent project looks into these questions in the context of offline and online classification of Internet packet flow traffic.