Soumya D. Mohanty
Professor 
Ph.D., Inter-University Center for Astronomy & Astrophysics, Pune, India, 1997
Office: BINAB 2.130 (Brownsville)
Phone: 956 882 6680
Email: soumya.mohanty@utrgv.edu
Curriculum Vitae
Teaching
General Physics I
Introduction to Astronomy I and II
(Graduate and Undergraduate) Classical Mechanics
Quantum II (Quantum Field Theory)
(Graduate) Statistical Methods for Modern Astronomy
Research
My research is focused on solving some important data analysis challenges in Gravitational Wave (GW) astronomy across all observational frequency bands. The following are some of my notable contributions. (a) Coherent network analysis of data for burst GW signals [in these papers 1,2]. The relevance of this work to the first direct detection of GWs can be seen in this paper. (b) Population study method to detect an association between GWs and Gamma Ray Bursts (and other astrophysical triggers). This is now a standard way to analyze GW data. (c) Introduction of Particle Swarm Optimization (PSO) in GW data analysis, where it is being used more frequently now. For aficionados of Statistics and Big Data, these challenges straddle a wide-range of areas such as semi-parametric regression of weak signals in noisy data, high-dimensional non-linear parametric regression, time series classification, and analysis of data from large heterogeneous sensor arrays. My work on GW data analysis has been funded by grants from the Research Corporation, the U.S. National Science Foundation, and NASA. In the area of pedagogy, I developed an innovative course that uses the medium of video games to teach Physics. View an early version of the course on game-based-learning.
My recent work includes a realistic estimate of the GW sensitivity of a future large-scale Pulsar Timing Array (PTA) reported in this PRL paper and a demonstration, reported in ApJL, that PTAs are capable of detecting much higher signal frequencies than assumed so far. PSO has now been shown to be a promising tool for accelerating the network analysis of binary inspiral GW signals. (With a recently awarded NSF MRI grant, the PSO-based network analysis search will be implemented on a GPU cluster along with a search for unmodeled chirps.) In a departure from GW data analysis, I have contributed a non-parametric regression method that uses PSO for adaptive fitting of splines to data. The code for this method, called SHAPES, is available from GitHub. A pedagogical introduction to using PSO in statistical regression problems is given in my book Swarm intelligence methods for statistical regression, published by CRC Press.
I try to maintain a more up-to-date account of research at my ResearchGate page.
Graduate and undergraduate students who wish to engage in research involving large-scale data science applications and high-performance computing are encouraged to contact me by email.