Enhancing deep neural networks via multiple kernel learning

作者：

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.

摘要

•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.

论文关键词：Deep neural networks,Deep learning,Multiple kernel learning,Ensemble learning

论文评审过程：Received 2 April 2019, Revised 11 December 2019, Accepted 1 January 2020, Available online 2 January 2020, Version of Record 7 January 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107194