Department of Biostatistics Seminar/Workshop Series

Scaling Bayes: Probabilistic Modeling for Large Models and Datasets

Chris Fonnesbeck, PhD, Assistant Professor, Department of Biostatistics, Vanderbilt University Medical Center

Bayesian models are appealing to practitioners because they are flexible, they automatically account for uncertainty in inferences and predictions, and their outputs are straightforward to interpret by non-expert users. However, fitting such models typically requires computationally-intensive algorithms (e.g. Markov chain Monte Carlo) that have traditionally scaled poorly with either the size of the model or the quantity of data used to fit the model. In the past several years there have been important innovations in Bayesian approximation methods that make working with large models and datasets feasible. I will provide an overview of some of these methods, and their implementation in new software tools using open source computational backends, including Theano (via PyMC3) and TensorFlow (via Edward and GPflow).
Topic revision: r1 - 22 Sep 2017, AshleeBartley

This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback