oh nice I wish it had been mehttps://twitter.com/ITBeHa/status/1048315485601914881 …
-
-
I'm very vulgar in my understanding of knowing: You can make good predictions or build stuff that don't fall apart very often, you know about a thing. One can think about different kinds of knowledge for people and other entites, but I probably don't believe in them.
-
I read the study. I think that I have an unusual and relatively bad understanding/definition of knowing in my head. When they do something relatively better than others, it means that they really know more about the thing, or something like that.
-
A man, his wife, and a statistician working for Facebook walk into a bar. man: I think I'll get a cocktail. wife: An Olympic, maybe? (I know he likes brandy and orange.) statistician: Probably a martini. (He's usually ordered a martini at bars.) Who understands him better?
-
Past statistics about something are one heuristic, and they work well within a narrow domain. Model-based understanding attempts to be *able* to make a prediction about *anything* about a system.
-
It's not about statistics vs models. More about model selection. People never act like a model-free bot though.
-
I can say that I better understand model, but I can't say that a q-learning bot that performs better on all tasks understands the problem less than a model-based bot with a weaker perfomance.
-
It's clearly all relative. The same with people: if your model works worse (early medicine?), you really know less.
- 2 more replies
New conversation -
-
-
I don't like Hayek's essays for some reason.
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.