Live data from Hacker News

Viewing profile — aab0

aab0

HN member
Joined
Tue, Mar 08, 2016, 5:12 PM UTC
HN karma
961
Public activity
330 items

About aab0

No profile information was provided.

Recent public activity

  1. comment
    Comment #13009505

    Which can be changed at any time.

  2. comment
    Comment #12774086

    More realistically: just import it from the rest of the world.

  3. comment
    Comment #12770196

    Supply generates its own demand; by making captioning even cheaper, it can increase the demand for transcription services and people to check it over. There are a lot of podcasts a…

  4. comment
    Comment #12757449

    We have to hardwire architectures because we can't learn architectures yet, or to put it another way, backpropagation doesn't yet work on hyperparameters as well as on parameters. …

  5. comment
    Comment #12757431

    If you want fine-grained tagging/categorization, you would be even better off with the -boorus.

  6. comment
    Comment #12757426

    That was deliberate. I couldn't resist either.

  7. comment
    Comment #12756959

    It could also reflect what is most distinguishable. Which is easier for a NN to confidently distinguish: black pubic hair on black skin, or black pubic hair on white skin? Darker n…

  8. comment
    Comment #12748505

    > (He's at George Mason University, which is a right-wing think tank, and has to say stuff like that to get tenure.) Hanson was an extreme libertarian long before he got tenure or …

  9. comment
    Comment #12747475

    I remember the one you're talking about: http://fusion.net/story/244545/famous-and-broke-on-youtube-i... (first hit for 'Youtube star waitressing')

  10. comment
    Comment #12720263

    Everywhere else on the Internet?

  11. comment
    Comment #12705309

    What sort of performance can be expected compared to running in the terminal? How large NNs will this scale to in practice? I see a 50-layer resnet is mentioned; but not 1000-layer…

  12. comment
    Comment #12687508

    It's interesting to read about the debate between doing editing/screen-drawing on the physical terminal ('smart' hardware) and doing it all in software. The wheel of reincarnation …

  13. comment
    Comment #12674456

    How would you know if they have been? As the human population shoots up massively, car-miles have increased enormously over the past century, and rural areas become denser, the num…

  14. comment
  15. comment
    Comment #12668076

    The obsession with information theory here seems like a classic nail-hammer thing. The number of bits my tests convey is totally useless to think about and certainly not worth spen…

  16. comment
    Comment #12661182

    This is perhaps the stupidest and most hysterical thing I've ever read on Stross's blog. I don't think I need to explain why Brexit is not going to lead to the https://en.wikipedia…

  17. comment
  18. comment
    Comment #12658131

    You can definitely see where this is going. My first thought was the same: 'we need eyetracking for foveated rendering anyway, so we can get realistic eyes for free', and if you ca…

  19. comment
    Comment #12637350

    I wouldn't bet a bent penny on it: "Experts said it was likely that the NSA or FBI had approached other Internet companies with the same demand, since they evidently did not know w…

  20. comment
    Comment #12628653

    > If you think the Internet is as safe and controlled as a shopping mall, you probably should be reading Krebs on Security more. That's an amusing comparison, given how much of Kre…

  21. comment
    Comment #12620723

    You could probably apply the same code. The dataset ("acquire 75,471 sketches of 12,500 objects") sounds adequate, and if not, can be boosted by first training a CNN to do photo->s…

  22. comment
    Comment #12616300

    It doesn't seem to lead to any harms when done as part of preimplantation genetic diagnosis in general, so this specific case shouldn't be too bad.

  23. comment
    Comment #12615981

    You're not supposed to. Image generation and modeling scene dynamics is a hard task, and thumbnail scale is what we're at at the moment. Nevertheless, those and other papers do dem…

  24. comment
    Comment #12615974

    > All of those goals are achieved with its design and improving on this foundation is certainly possible. These were all hard problems to solve at the time, given the state of the …

  25. comment
    Comment #12609514

    Note the moving objects in those generated/predicted thumbnails.