An improved error-correcting output coding framework with kernel-based decoding

Abstract

This paper presents a novel framework of error-correcting output coding (ECOC) addressing the problem of multi-class classification. By weighting the output space of each base classifier which is trained independently, the distance function of decoding is adapted so that the samples are more discriminative. A criterion generated over the Extended Pair Samples (EPS) is proposed to train the weights of output space. We first conduct empirical studies on UCI datasets to verify the presented framework with four frequently used coding matrixes and then apply it in RoboCup domain to enhance the performance of agent control. Color-based image segmentation is also tested in this paper. Experimental results show that our supervised learned decoding scheme improves the accuracy of classification significantly and betters the ball control of agents in a soccer game after learning from experience. Some properties still hold in the new framework: any classifier, as well as distance function, is still applicable. Besides, it is allowed in the new framework to implement nonlinear decoding processes using kernel tricks.