“Preferred Stimuli” for DNNs


Not *entirely* sure what’s going on here (paper forthcoming) but I believe this is the equivalent of tickling a deep neural network: “understanding deep neural networks by synthetically generating the preferred stimuli for each of their neurons.” Via Anh Nguyen, Alexey Dosovitskiy, Jason Yosinski, Thomas Brox, and Jeff Clune at EvolvingAI.org.

Leave a Reply

Your email address will not be published. Required fields are marked *