Deep Convolutional Neural Networks (CNN) are widely used in many practical tasks in the field of computer vision . However, in order to ensure performance, neural networks are usually over-parameterized, so there will be a lot of redundant parameters. The recently proposed additive neural network [2] greatly reduces the multiplication operation in the neural network by replacing the distance measurement function in the convolution operation with the L1 distance, thereby reducing the power consumption and chip area required for network operation.

However, ANN still has a certain gap in accuracy compared with CNN of the same structure, which limits the replacement of CNN by ANN in practical applications to some extent. …










About

Nabil MADALI

I am actually student in Data Science at Ecole Polytechnique the leading French institution combining top-level research, academics, and innovation .

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store