Reject Option Paradigm for the Reduction of Support Vectors
Ricardo Sousa, Ajalmar R Da, Rocha Neto, Guilherme A. Barreto, Jaime S. Cardoso, Miguel Tavares Coimbra · 2014
Abstract. In this paper we introduce a new conceptualization for the reduction of the number of support vectors (SVs) for an efficient design of support vector machines. The techniques here presented provide a good balance between SVs reduction and generalization capability. Our pro-posal explores concepts from classification with reject option. These meth-ods output a third class (the rejected instances) for a binary problem when a prediction cannot be given with sufficient confidence. Rejected instances along with misclassified ones are discarded from the original data to give rise to a classification problem that can be linearly solved. Our experi-mental study on two benchmark datasets show significant gains in terms of SVs reduction with competitive performances. 1