dc.contributor.advisor | Pati, Debdeep | |
dc.contributor.advisor | Bhattacharya, Anirban | |
dc.creator | Guha, Biraj Subhra | |
dc.date.accessioned | 2022-01-24T22:19:50Z | |
dc.date.available | 2022-01-24T22:19:50Z | |
dc.date.created | 2021-08 | |
dc.date.issued | 2021-07-16 | |
dc.date.submitted | August 2021 | |
dc.identifier.uri | https://hdl.handle.net/1969.1/195138 | |
dc.description.abstract | I provide statistical guarantees for Bayesian variational boosting by
proposing a novel small bandwidth Gaussian mixture variational family. We employ a functional
version of Frank-Wolfe optimization as our variational algorithm and study frequentist properties of the iterative boosting updates. Comparisons are drawn to the recent literature on boosting, describing
how the choice of the variational family and the discrepancy measure affect both convergence and
finite-sample statistical properties of the optimization routine. Specifically, we first demonstrate
stochastic boundedness of the boosting iterates with respect to the data generating distribution. We
next integrate this within our algorithm to provide an explicit convergence rate, ending with a result
on the required number of boosting updates. Next, I develop a framework to study posterior contraction rates in sparse high dimensional generalized linear models (GLM). We introduce a new family of GLMs, denoted by clipped GLM, which subsumes many standard GLMs and makes minor modification of the rest. With a sparsity inducing prior on the regression coefficients, we delineate sufficient conditions
on true data generating density that leads to minimax optimal rates of posterior contraction of the coefficients in l_1 norm. Our key contribution is to develop sufficient conditions commensurate with the geometry of the clipped GLM family, propose prior distributions which do not require any knowledge of the true parameters and avoid any assumption on the growth rate of the true coefficient vector | en |
dc.format.mimetype | application/pdf | |
dc.language.iso | en | |
dc.subject | Approximate Bayesian Inference | en |
dc.subject | Variational Boosting | en |
dc.subject | Frank--Wolfe Algorithm | en |
dc.subject | Convergence Rate | en |
dc.subject | Kullback-Leibler Divergence | en |
dc.subject | Gaussian Mixtures | en |
dc.subject | High-dimension | en |
dc.subject | Sparse Regression | en |
dc.subject | Generalized Linear Models | en |
dc.subject | Posterior Convergence | en |
dc.subject | Model Selection | en |
dc.subject | Adaptive Estimation | en |
dc.subject | Spike-and-slab Prior | en |
dc.subject | Minimax Rate | en |
dc.title | Theoretical Guarantees for Bayesian Generalized Linear Regression And Variational Boosting | en |
dc.type | Thesis | en |
thesis.degree.department | Statistics | en |
thesis.degree.discipline | Statistics | en |
thesis.degree.grantor | Texas A&M University | en |
thesis.degree.name | Doctor of Philosophy | en |
thesis.degree.level | Doctoral | en |
dc.contributor.committeeMember | Cline, Daren | |
dc.contributor.committeeMember | Carroll, Raymond | |
dc.contributor.committeeMember | Narayanan, Krishna | |
dc.type.material | text | en |
dc.date.updated | 2022-01-24T22:19:50Z | |
local.etdauthor.orcid | 0000-0002-1806-936X | |