Feng Liang

Overview

Feng Liang is a statistician (Professor, Department of Statistics, University of Illinois at Urbana-Champaign) working on Bayesian model selection, nonparametric Bayes, and statistical machine learning.

Key Contributions / Features

• Mixtures of g-priors (Liang et al. 2008): lead author of the hyper-g / hyper-g/n prior families that resolve the g-prior's Bartlett and information paradoxes with closed-form marginal likelihoods (Zellner's g-Prior).

Related

• Zellner's g-Prior
• Variable Selection
• Liang et al. (2008) Mixtures of g-Priors for Bayesian Variable Selection
• Merlise Clyde
• James O. Berger