I am currently in the first year of my doctoral program at UChicago. My research interests include Probability Theory, Theoretical Statistics and Functional Analysis, as well as related disciplines like Machine Learning and Pattern Recognition. My Master's Thesis was on Marcinkiewicz Strong Law of Large Numbers for products of heavy-tailed and long-range dependent linear processes. Before that, I held student research positions in ISI Bangalore and TIFR-CAM. My hobbies include playing the violin, reading superhero stories, and cricket. Please scroll down to learn more about me.
Sounak Paul
Department of Statistics
G.H. Jones Laboratory
5747 South Ellis Avenue
Chicago, IL 60637
paulsounak96@gmail.com
Doctor of Philosophy (PhD) in Statistics• Chicago, USA • 2019 - Present
Incoming student.
Master of Science (MSc) in Mathematics• Edmonton, Canada • 2017 - 2019
Obtained a perfect 4.0 GPA.
Bachelor of Mathematics (Honours)• Bangalore, India • 2014 - 2017
Obtained First Division with Distinction, and secured Second Rank in my batch.
CBSE(Class 10) and AISSCE(Class 12)• Kolkata, India • 2008 - 2014
University of Alberta • 2018
University of Alberta • 2017
University of Alberta • 2017
Awarded to 20-odd masters students in the entire Faculty of Graduate Studies and Research.
Indian Academy of Sciences • Summer 2016
TIFR Mumbai and TIFR-CAM Bangalore • Summer 2016
Regional Mathematics Olympiad • 2013
I cracked the Regional Mathematics Olympiad, Bengal Zone 2013.
Kishore Vaigyanik Protsahan Yojana • 2013
I received the prestigious KVPY scholarship awarded by the Department of Science and Technology and the Indian Institute of Science Bangalore.
Bucharest, Romania • 2006
I Represented India, and was the youngest participant at this event organised by the Ministry of Education, Research and Youth, and Lumina institutii de invatamant, Bucharest.
Prof. Michael Kouritzin, University of Alberta • May 2018 - Aug 2019
Several big datasets we encounter, especially in network, financial and environmental series, is often not only big but also heavy-tailed and long-range dependent. This combination of big, heavy and long data can often be modeled well by multivariate linear processes but is lethal for most classical statistics. Long memory pretty much rules out any use of strong mixing; heavy tails makes use of moments impossible without techniques like truncation, and the two-sided nature of the processes means we need to worry about the future as well as the past. Recently, Marcinkiewicz strong laws of large numbers (MSLLN) were established for showing consistency and optimal polynomial rates of convergence for this kind of data for parameter estimation, adaptive machine learning, and econometrics. I generalized the MSLLN results of well known autocovariance case to multiple products of Heavy-Tailed and long range dependent data. I have found that a MSLLN for even products of linear processes with symmetric innovations converge at a faster rate than those for odd products.
Prof. BS Daya Sagar, ISI Bangalore • Summer 2017
This project was conducted in Systems Science and Informatics Unit (SSIU) of ISI, under the KVPY fellowship. I extended notions of dilations, erosions, opening, closing, and other Morphological operators to vertex and edge weighted graphs, and formulated a notion of medians in these graphs via morphological operators. I also formulated a method of interpolation using successive medians, following ideas from supervisor's previous papers, and simulated on various datasets in Python.
Prof. K Sandeep, TIFR-CAM • Summer 2016
I worked under Prof. K Sandeep for 2 months. It was primarily an advanced comprehensive reading project in areas related to Partial Differential Equations, where I reviewed basic concepts of Functional Analysis, Operator Theory and Geometric Analysis, then studied Frechet and Gateux derivatives. Then, I conducted a comprehensive study of stability theory and distribution theory, with emphasis on distribution solutions of general PDEs from analytic as well as geometric point of view.
Prof. Yogeshwaran Dhandapani, ISI Bangalore • Summer 2015
Graduate Colloquium, University of Alberta • Fall 2018
University of Chicago • September 2019
University of Alberta • September 2018
University of Alberta • September 2018
University of Alberta • August 2018
Chennai Mathematical Institute • Summer 2017
University of Alberta
University of Alberta
University of Alberta
University of Alberta
University of Alberta
In my due course of study, my formal training of Computer Science has been limited. But working in various academic and non-academic projects has enabled me to acquire the followig skill-set.