• 0 Posts
  • 47 Comments
Joined 1 year ago
cake
Cake day: July 2nd, 2023

help-circle
  • Well, if that commenter had more than just a vague idea of caching and/or swapping, they would know that the right algorithm can make or break performance.

    That paper is not “we invented caching”, but “this is how we make some certain models work well despite constraints imposed by RAM and flash storage.”

    It’s a worthy job for an engineer or researcher. Not quite as innovative as the invention of the wheel, but still enough to write a paper on (and read it, if you can manage to understand it).




















  • lemmylommy@lemmy.worldtoAsklemmy@lemmy.ml*Permanently Deleted*
    link
    fedilink
    arrow-up
    32
    arrow-down
    2
    ·
    8 months ago

    Don’t rely on the algorithms. They will inevitably steer you towards clickbait, mindless attention whoring and controversy baiting. After all, they are designed for keeping you engaged at the cost of wearing down your attention span.

    Instead consider what you are interested in and try to find likeminded people online. From there, look for connections to other places. Those can be links or something someone mentioned. Even in times of Facebook, Reddit etc. there are still some great forums on some topics around. And there are real gems buried under all the crap on YouTube, you just need to find them yourself.

    If you just want to browse for interesting videos without a specific topic in mind, try the channels of public broadcasting. PBS, dw, ARTE etc. have great documentaries on all kinds of topics.

    Also consider listening to audiobooks. Audible is not that expensive, otherwise some are on YouTube or in some places you can get access to online libraries with a offline library card.