This is nice. The idea is that a learning algorithm is a natural transformation between a functor which generates training data and a functor which generates the possible functions. Naturality says that if you alter your data you should get a nice way to change your functions.https://twitter.com/_julesh_/status/1125705332305670144 …
-
-
I was talking about the issue in terms of the wiring diagrams - but you're right. There's also been plenty of papers discussing what's the best distance measure to use in GANs. I wonder what kind of light CT will cast on this matter
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.