We propose a novel discriminative leing approach for Bayesian patt classification,called ’constrained maximum margin(CMM)’.We define the margin between two classes as the difference between the minimum decision value for positive samples and the maximum decision value for negative samples.The leing problem is to maximize the margin under the con-straint that each training patt is classified correctly.This nonlinear programming problem is solved using the sequential un-constrained minimization technique.We applied the proposed CMM approach to le Bayesian classifiers based on Gaussian mixture models,and conducted the experiments on 10 UCI datasets.The performance of our approach was compared with those of the expectation-maximization algorithm,the support vector machine,and other state-of-the-art approaches.The experimental results demonstrated the effectiveness of our approach.