Log-Nets: Logarithmic Feature-Product Layers Yield More Compact Networks

Philipp Grüning, Thomas Martinetz, Erhardt Barth

Abstract: We introduce Logarithm-Networks (Log-Nets), a novel bio-inspired type of network architecture based on logarithms of feature maps followed by convolutions. Log-Nets are capable of surpassing the performance of traditional convolutional neural networks (CNNs) while using fewer parameters. Performance is evaluated on the Cifar-10 and ImageNet benchmarks.