What do you mean with picking pixels optimally? For very close to all images, I expect there to exist six pixels such that the judge identifies the correct label, if they are revealed. That doesn’t seem like a meaningful metric, though.
What do you mean with picking pixels optimally? For very close to all images, I expect there to exist six pixels such that the judge identifies the correct label, if they are revealed. That doesn’t seem like a meaningful metric, though.