Fig. 6From: Evaluation of maxout activations in deep learning across several big data domainsScaled exponential linear unitBack to article page