Statistics and Probability seminar
Thursday, 16 March 2023, 3 pm
============================
Venue: Online, link will be sent later
Host: Debraj das
Speaker: Sandipan Roy
Affiliation: University of Bath
Title: Statistical Inference in Complex Data with Network Structure
Abstract: New technological advancements have allowed the collection of datasets of large volume and different levels of complexity. Many of these datasets have an underlying network structure. Networks are capable of capturing dependence relationships among a group of entities and hence analyzing these datasets unearth the underlying structural dependence among the individuals. Examples include gene regulatory networks, understanding stock markets, protein-protein interaction within the cell, online social networks etc. We present two important aspects of large high-dimensional data with network structure. The first one focuses on a data with a network structure that evolves over time. Examples of such data sets include time course gene expression data, voting records of legislative bodies etc. Traditionally, the estimation of Gaussian graphical models (GGM) is performed in an i.i.d setting. More recently, such models have been extended to allow for changes in the distribution, but primarily where changepoints are known a priori. In this work, we study the Group-Fused Graphical Lasso (GFGL) which penalizes partial correlations with an L1 penalty while simultaneously inducing block-wise smoothness over time to detect multiple changepoints. The other aspect that we examine is heterogeneity in a network structure and how we can use such heterogenous features in a predictive model. We use a linear latent variable model viz. PCA and its extensions to learn an underlying network structure from data varying over time. We then employ the learned network as a feature in a predictive model to perform the downstream task in the test data. Neuroimaging-driven prediction of brain age, defined as the predicted biological age of a subject using only brain imaging data, is an exciting avenue of research. In this work, we seek to build models of brain age based on functional connectivity while prioritizing model interpretability and understanding. This way, the models serve to both provide accurate estimates of brain age as well as allow us to investigate changes in functional connectivity that occur during the aging process.