A new classifier based on information theoretic learning with unlabeled data

Kyu-Hwa Jeong, Jian-Wu Xu, Deniz Erdogmus, Jose C. Principe
2005 Neural Networks  
Supervised learning is conventionally performed with pairwise input-output labeled data. After the training procedure, the adaptive system's weights are fixed while the testing procedure with unlabeled data is performed. Recently, in an attempt to improve classification performance unlabeled data has been exploited in the machine learning community. In this paper, we present an information theoretic learning (ITL) approach based on density divergence minimization to obtain an extended training
more » ... lgorithm using unlabeled data during the testing. The method uses a boosting-like algorithm with an ITL based cost function. Preliminary simulations suggest that the method has the potential to improve the performance of classifiers in the application phase. q
doi:10.1016/j.neunet.2005.06.018 pmid:16102941 fatcat:vgco272jrzdv5oxwvehd4n5ibi