3 ms·
"We extracted small patches from images, just like they do in neural networks research. Then we used neural networks and humans to identify the images patches w
by conjectures 11y ago
"We extracted small patches from images, just like they do in neural networks research. Then we used neural networks and humans to identify the images patches were drawn from. Because humans can do this and neural networks can too humans must work like neural networks."
Hmm. Did I miss something?
- gone35 11y agoA lot: the whole random projections thing.
- conjectures 11y agoThe random projections that happened to be blurred or overlaid image patches? My point isn't about whether RP makes for good ML algorithms, it's about whether human vision is patch-based.