I’d be interested to see what it said about differentiating a function (and I’m assuming you mean in the context of programming). I consider that one of the dir

I’d be interested to see what it said about differentiating a function (and I’m assuming you mean in the context of programming). I consider that one of the dirtier secrets of gradient-based ML.