Enhancing deep neural networks via multiple kernel learning. (May 2020)

• Record Type: Journal Article
• Authors: Lauriola, Ivano; Gallicchio, Claudio; Aiolli, Fabio
• Is Part Of: Pattern recognition. Volume 101(2020:May)
• Publication Date: 2020-05
• Subjects: Deep neural networks -- Deep learning -- Multiple kernel learning -- Ensemble learning; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2020.107194
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We introduce KerNET, which combines Deep Neural Networks and Multiple Kernel learning; The framework optimally aggregates the hidden representations of a Deep Neural Network. KerNET enables to boost the predictive accuracy of individual Deep Neural Networks. KerNET allows to relieve the overfitting problem in deeper architectures. Abstract: Deep neural networks and Multiple Kernel Learning are representation learning methodologies of widespread use and increasing success. While the former aims at learning representations through a hierarchy of features of increasing complexity, the latter provides a principled approach for the combination of base representations. In this paper, we introduce a general framework in which the internal representations computed by a deep neural network are optimally combined by means of Multiple Kernel Learning. The resulting ensemble methodology is instantiated for Multi-layer Perceptrons architectures (both fully trained and with random-weights), and for Convolutional Neural Networks. Experimental results on several benchmark datasets concretely show the advantages and potentialities of the proposed approach.