Neuron Pruning for Compressing Deep Networks using Maxout Architectures
Authors:
Preprint
Abstract
This paper presents an efficient and robust approach for reducing the size of deep neural networks by pruning entire neurons. It exploits maxout units for combining neurons into more complex convex functions and it makes use of a local relevance measurement that ranks neurons according to their activation on the training set for pruning them. Additionally, a parameter reduction comparison between neuron and weight pruning is shown. It will be empirically shown that the proposed neuron pruning reduces the number of parameters dramatically. The evaluation is performed on two tasks, the MNIST handwritten digit recognition and the LFW face verification, using a LeNet-5 and a VGG16 network architecture. The network size is reduced by up to $74\%$ and $61\%$, respectively, without affecting the network's performance. The main advantage of neuron pruning is its direct influence on the size of the network architecture. Furthermore, it will be shown that neuron pruning can be combined with subsequent weight pruning, reducing the size of the LeNet-5 and VGG16 up to $92\%$ and $80\%$ respectively.
Keywords
All Related Versions
This publication has 5 references indexed in Scilit:
- Optimistic and pessimistic neural networks for object recognition2017
- Fast computation of the performance evaluation of biometric systems: Application to multibiometricsFuture Generation Computer Systems, 2013
- Permutation Tests for Studying Classifier Performance2009
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- An Ordination of the Upland Forest Communities of Southern WisconsinEcological Monographs, 1957