Date of Original Version
Proceedings of 26th International Conference on Machine Learning, (ICML) pp. 481-488, 2009
Abstract or Table of Contents
Given multiple possible models b1; b2; : : : bn for a protein structure, a common sub-task in in-silico Protein Structure Prediction is ranking these models according to their qual- ity. Extant approaches use MLE estimates of parameters ri to obtain point estimates of the Model Quality. We describe a Bayesian alternative to assessing the quality of these models that builds an MRF over the parame- ters of each model and performs approximate inference to integrate over them. Hyper- parameters w are learnt by optimizing a list- wise loss function over training data. Our results indicate that our Bayesian approach can signicantly outperform MLE estimates and that optimizing the hyper-parameters can further improve results.