An ensemble approach to robust classifier fusion

In this thesis I present an ensemble approach that is a robust tool for classifier fusion. The proposed technique is a multiple view generalization of AdaBoost in the sense that weak learners from various information sources are selected in each iteration based on lowest weighted error rate. Weak learners trained on individual views in each iteration rectify the bias introduced by learners in preceding iterations resulting in a self regularizing behavior. I compare the classification performance of proposed technique with recent classifier fusion strategies in various domains such as face detection, gender classification and texture classification. In addition to that, I provide theoretical guarantees that the approach will always provide better results compared to the situation when no classifier fusion is used.