Comment on answer = sum(n) / len(n)

Socsa@sh.itjust.works ⁨1⁩ ⁨year⁩ ago

• TopRamenBinLaden@sh.itjust.works ⁨1⁩ ⁨year⁩ ago

  I like your funny words, magic man.

  source
• lseif@sopuli.xyz ⁨1⁩ ⁨year⁩ ago

  erm, in english, please !

  source

  • hotsox@lemmy.blahaj.zone ⁨1⁩ ⁨year⁩ ago

    Universal function approximation - neural networks. Auto-differentiation - algorithmic calculation of partial derivatives (aka gradients) Backpropagation - when using a neural network (or most ML algorithms actually), you find the difference between model prediction and original labels. And the difference is sent back as gradients (of the loss function) Parameter dimensionality - the “neurons” in the neural network, ie, the weight matrices.

    If thats your argument, its worse than Statistics imo. Atleast statistics have solid theorems and proofs (albeit in very controlled distributions). All DL has right now is a bunch of papers published most often by large tech companies which may/may not work for the problem you’re working on.

    Universal function approximation theorem is pretty dope tho. Im not saying ML isn’t interesting, some part of it is but most of it is meh. It’s fine.

    source
• kibiz0r@midwest.social ⁨1⁩ ⁨year⁩ ago

  A monad is just a monoid in the category of endofunctors, after all.

  source

  • AnarchistArtificer@slrpnk.net ⁨1⁩ ⁨year⁩ ago

    No, no, everyone knows that a monad is like a burrito.

    (Joke is referencing this: blog.plover.com/prog/burritos.html )

    source

    • Contravariant@lemmy.world ⁨1⁩ ⁨year⁩ ago

      So what you’re saying is that if you put a burrito inside a burrito it’s still a burrito?

      source

      • AnarchistArtificer@slrpnk.net ⁨1⁩ ⁨year⁩ ago

        youtu.be/2EWWL3niBWY

        source
• feedum_sneedson@lemmy.world ⁨1⁩ ⁨year⁩ ago

  pee pee poo poo wee wee

  source
• embed_me@programming.dev ⁨1⁩ ⁨year⁩ ago

  Any practical universal function approximation will go against entropy.

  source
• hotsox@lemmy.blahaj.zone ⁨1⁩ ⁨year⁩ ago

  Universal function approximation - neural networks. Auto-differentiation - algorithmic calculation of partial derivatives (aka gradients) Backpropagation - when using a neural network (or most ML algorithms actually), you find the difference between model prediction and original labels. And the difference is sent back as gradients (of the loss function) Parameter dimensionality - the “neurons” in the neural network, ie, the weight matrices.

  If thats your argument, its worse than Statistics imo. Atleast statistics have solid theorems and proofs (albeit in very controlled distributions). All DL has right now is a bunch of papers published most often by large tech companies which may/may not work for the problem you’re working on.

  Universal function approximation theorem is pretty dope tho. Im not saying ML isn’t interesting, some part of it is but most of it is meh. It’s fine.

  source