Masking out parts of the image and monitoring how each part changes the prediction. This lets you porduce and "importance map" about which parts of the image contribute more to the prediction.
4,368 karma · joined October 10, 2018
"In the year 2081, the 211th, 212th, and 213th amendments to the Constitution dictate that all Americans are fully equal and not allowed to be smarter, better-looking, or more physically able than anyone else. The Handicapper General's agents enforce the equality laws, forcing citizens to wear "handicaps": masks for those who are too beautiful, loud radios that disrupt thoughts inside the ears of intelligent people, and heavy weights for the strong or athletic. "
https://addons.mozilla.org/en-GB/firefox/addon/feedbroreader...
https://www.wolframalpha.com/input/?i=sin%28pi+%2F+6%29+%3D%...
https://hnrss.org/replies?id=xcambar
For more options see: https://hnrss.github.io/