Don't worry. It was a bit ambiguous 👍

Yeah. Sorry. I figured it was possible that you were using desktop or something and maybe you'd just not realized it wasn't visible on mobile

Wikipedia called these fencepost errors at one point (they now just say it is the specific type of off-by-one error in which you miscount the posts (vertices) or panels (arcs) in a fence (graph) by using one to count the other). I read this before my first programming class and then mentioned the term to my professor. She had no idea what I was talking about 😅

Yes. My (minimally informed from a single class) understanding is that it sort-of depends on the problem too. Like perhaps in looking at all the data on proteins, the neural network might notice a pattern in protein folding is applicable to the tweaked problem. Of course, there is no guarantee that such a generally applicable rule exists. And even if it does, it might not be discovered by the net before overtraining occurs.

Nekomusume ni naritai!