Anonymous ID: 4ad66c Nov. 14, 2018, 4:15 a.m. No.3897717   🗄️.is 🔗kun   >>7724

>>3897342

 

The natives greeted them kindly and invited them to dine

On yams and clams and human hams and vintage coconut wine

The taste of which was filthy, but the after-effects divine

Anonymous ID: 4ad66c Nov. 14, 2018, 4:33 a.m. No.3897766   🗄️.is 🔗kun

Look at this meme, Jim.

You'll never see another one like it.

(or recognize it if you do)

 

>We present a method to create universal, robust, targeted adversarial image patches

in the real world. The patches are universal because they can be used to attack

any scene, robust because they work under a wide variety of transformations,

and targeted because they can cause a classifier to output any target class. These

adversarial patches can be printed, added to any scene, photographed, and presented

to image classifiers; even when the patches are small, they cause the classifiers to

ignore the other items in the scene and report a chosen target class.

 

Adversarial examples have been shown

to generalize to the real world