Sam Leroux, Steven Bohez, Cedric De Boom, Elias De Coninck, Tim Verbelen, Bert Vankeirsbilck, Pieter Simoens, Bart Dhoedt
Published 2016-05-27Version 1
In this paper we propose a technique which avoids the evaluation of certain convolutional filters in a deep neural network. This allows to trade-off the accuracy of a deep neural network with the computational and memory requirements. This is especially important on a constrained device unable to hold all the weights of the network in memory.
Predicting psychological attributions from face photographs with a deep neural network
Explaining How a Deep Neural Network Trained with End-to-End Learning Steers a Car
Collaborative Layer-wise Discriminative Learning in Deep Neural Networks
![QR code for arXiv:1605.08543 [cs.CV]](http://oss.arxitics.com/images/favicon.svg)