This GitHub repository contains code accompanying my third-year project titled "A study on approximation techniques for Bayesian Logistic Regression" which covers Laplace Approximation, Markov Chain Monte Carlo (MCMC) and Variational Methods.
The work is based on
- The original paper by Jaakkola and Jordan's 1996 paper called "A variational approach to Bayesian logistic regression models and their extensions" and by the subsequent analysis done by Bishop (2006) in his "Pattern Recognition and Machine Learning" textbook.
- "MCMC in Practice" by David Spiegelhalter, Sylvia Richardson, and W. R. Gilks and many other good MCMC textbooks.
To see the most up-to-date version of my dissertation, you can navigate to here.