{"DOI":"10.1007/s41095-020-0181-9","abstract":"Abstract\nThis paper proposes a kernel-blending connection approximated by a neural network (KBNN) for image classification. A kernel mapping connection structure, guaranteed by the function approximation theorem, is devised to blend feature extraction and feature classification through neural network learning. First, a feature extractor learns features from the raw images. Next, an automatically constructed kernel mapping connection maps the feature vectors into a feature space. Finally, a linear classifier is used as an output layer of the neural network to provide classification results. Furthermore, a novel loss function involving a cross-entropy loss and a hinge loss is proposed to improve the generalizability of the neural network. Experimental results on three well-known image datasets illustrate that the proposed method has good classification accuracy and generalizability.","author":[{"family":"Liu","given":"Xinxin"},{"family":"Zhang","given":"Yunfeng"},{"family":"Bao","given":"Fangxun"},{"family":"Shao","given":"Kai"},{"family":"Sun","given":"Ziyi"},{"family":"Zhang","given":"Caiming"}],"id":"unknown","issued":{"date-parts":[[2020,9,14]]},"language":"en","publisher":"Springer Science and Business Media LLC","title":"Kernel-blending connection approximated by a neural network for image classification","type":"article-journal"}