A Comparison of Bayesian Inference Techniques for Sparse Factor Models

Sparse factor models are used for discovering latent variables hidden in high-dimensional data, where the relations between latent variables and observed quantities is known to be sparse. The aim of this project is to compare the performance of two Bayesian inference techniques for sparse factor models: MCMC and Variational Inference. Specifically, the tradeoff between accuracy and computational efficiency is to be explored. The comparison will require an implementation of Variational Inference for sparse factor models. The evaluation is to be carried out in the context of inference of gene regulatory networks using RNA-seq data.

Yong See Foo

The University of Melbourne

Yong See is a third-year Bachelor of Science student at the University of Melbourne, undertaking a major in Statistics, with diplomas in informatics and music. The field of modern statistics interests him as it combines mathematical theory and the utilisation of computational techniques for understanding data. His previous research topics include anomaly detection in time series, and genome assembly of highly-rearranged chromosomes. Aside from studies, Yong See enjoys playing and writing music. He is regularly involved in a student orchestra, playing the violin.

You may be interested in

Liuhao Yu

Liuhao Yu

The Dirichlet Problem in the Light of Capacity Theory
Matthew Hanna

Matthew Hanna

Optimisation and Theoretical Implications of Probability Inequalities
Michael Carr

Michael Carr

Advanced Bayesian Statistical Inference Methods for Simulation-Based Models in Cell Biology
Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.

Not readable? Change text.