Andrew Nguyen L&S Math & Physical Sciences

Multidimensional Bayesian Density Deconvolution

In experimental sciences, effectively extracting “signals” hidden in noisy data is a constant challenge. The true signal may be highly structured. For example, think of true data lying on a circle within plane that is observed in the unit cube. How can we recover the circle effectively, in arbitrarily large dimension, and provide some measurement of uncertainty in our estimations to guide downstream tasks?

Currently, a well-established method called Non-Parametric Maximum Likelihood Estimation (NPMLE) is capable of isolating vertices of the circle, but it struggles in recovering its smooth shape, scales poorly in high-dimension, and does not provide uncertainty quantification (UQ).

My research addresses this gap by working to develop methodologies to achieve all three goals: recovering a smooth circle, scalability to arbitrary dimension, and providing UQ. This project aims to first provide a Bayesian counterpart to the NPMLE in a Dirichlet Gaussian Mixture Model, where both models are constrained in dimensionality and recovery of a smooth circle but the latter provides UQ, then develop methods to improve upon the Dirichlet.

Message To Sponsor

Thank you for your support! I have grown very interested in Bayesian Statistics and deeply appreciate the opportunity to deeply involve myself in research this summer. I particularly like the theoretical portions of this project, as I intend to pursue doctoral studies in this field. Thank you again for the opportunity to spend my summer deeply immersed in research; I can't wait to share what I find.
Headshot of Andrew Nguyen
Major: Applied Mathematics, Statistics
Mentor: Aditya Guntuboyina
Sponsor: Zara
Back to Listings
Back to Donor Reports