Kernel-based Translations of Convolutional Networks

2019/03/19 by Jones, Corinne, Roulet, Vincent, Harchaoui, Zaid
#FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Optimization and Control (math.OC)

paper · doi:10.48550/arxiv.1903.08131

Abstract

Convolutional Neural Networks, as most artificial neural networks, are commonly viewed as methods different in essence from kernel-based methods. We provide a systematic translation of Convolutional Neural Networks (ConvNets) into their kernel-based counterparts, Convolutional Kernel Networks (CKNs), and demonstrate that this perception is unfounded both formally and empirically. We show that, given a Convolutional Neural Network, we can design a corresponding Convolutional Kernel Network, easily trainable using a new stochastic gradient algorithm based on an accurate gradient computation, that performs on par with its Convolutional Neural Network counterpart. We present experimental results supporting our claims on landmark ConvNet architectures comparing each ConvNet to its CKN counterpart over several parameter settings.

Related

• Convolutional Deep Kernel Machines
• Adaptive Convolution Kernel for Artificial Neural Networks
• Kernel function impact on convolutional neural networks
• Toward Understanding Convolutional Neural Networks from Volterra Convolution Perspective
• Convolutional Neural Networks In Convolution