Tweetovi

Blokirali ste korisnika/cu @MuhammadPuter12

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @MuhammadPuter12

  1. proslijedio/la je Tweet
    16. sij

    How much information does a cubic centimeter of water hold? That's close to a gram of water. If you work it out, you'll see our data storage abilities are nowhere near the fundamental limits imposed by physics! Of course, you can't *store* information in liquid water.

    Poništi
  2. 12. sij
    Poništi
  3. proslijedio/la je Tweet
    10. sij

    I do not like my computation in the cloud, I not like remote GPUs I yell out loud. I did not sign up for dev ops, I just want more goddamned flops. I dont want docker compose, I dont want to write this prose. I dont want compute by the hour, I just want more computing power.

    Poništi
  4. 11. sij

    It always makes me happy to see one of my favorite writers gush about another of my favorites. And it's easy to see that David himself is also an absolute force of nature

    Poništi
  5. proslijedio/la je Tweet
    10. sij
    Odgovor korisniku/ci

    Japanese kids do the same with English words. They memorise the shape of English words the same way as they do with Chinese characters. Until they learn English phonics which allows them to sound words out

    Poništi
  6. proslijedio/la je Tweet
    10. sij
    Odgovor korisniku/ci

    If you want to get more of them, try asking people explicitly what their work taught them that outsiders don't know. I often do this.

    Poništi
  7. proslijedio/la je Tweet
    28. pro 2019.
    Odgovor korisniku/ci
    Poništi
  8. 9. sij

    Estimating theoretical bounds forced by physical laws is my kind of jam

    Poništi
  9. 29. pro 2019.
    Poništi
  10. proslijedio/la je Tweet
    24. pro 2019.

    Earth from Apollo 8 51 years ago

    Poništi
  11. proslijedio/la je Tweet
    19. pro 2019.

    Is there a set with 2.5 elements? No! But here's a "groupoid" with 2.5 elements. To get it, just take a set with 5 elements and fold it in half. The point in the middle gets folded over, and becomes half a point. Sounds wacky, but you can make this into rigorous math! (1/n)

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    24. pro 2019.

    Detecting AI generated DeepFakes using AI is a losing game. I give it 2 years tops Eventually it will be impossible for the same reason it’s impossible to humans. It’s an information bottleneck. Example: “010010” <- is this fake or was this created based on real world events?

    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet
    20. pro 2019.

    I often hear people say that X is broken. Congress is broken. Academia is broken. Physics is broken. Deep learning is broken. People like to stand in crowds and point fingers at the thing that is broken. I prefer the quiet folks who roll up their sleeves and get to work, fixing.

    Poništi
  14. proslijedio/la je Tweet
    4. pro 2019.

    Really excited that YouTube gave me the first 90 seconds of this year's Rewind to explain the volume formulas for higher dimensional spheres.

    Poništi
  15. proslijedio/la je Tweet

    .’s viral real-time AI art sensation just won a 2019 Magazine “Best of What’s New Award” in the engineering category. Developed by NVIDIA researchers in 2019, GauGAN is the first AI model that can produce complex images with only a few brushstrokes.

    Poništi
  16. proslijedio/la je Tweet

    Steve Ballmer and Bill Gates once made an Austin Powers parody video

    Poništi
  17. proslijedio/la je Tweet
    2. pro 2019.

    Saturn's moon Tethys (1,076 km wide) imaged by in front of Saturn on December 3, 2005. North is to the left in this view, the rings are seen on edge along the side.

    Prikaži ovu nit
    Poništi
  18. 3. pro 2019.

    Always happy to see papers in the microgenre typified by "Can a neuroscientist understand a microprocessor?"

    Poništi
  19. proslijedio/la je Tweet
    28. stu 2019.

    For anyone who has ever thought - "Can I learn the math needed for Deep Learning all in one place (& maybe skip the other stuff)?" - this is quite a nice resource! "The Matrix Calculus You Need For Deep Learning" (Table of Contents 👇)

    Poništi
  20. 29. stu 2019.

    By Jove, this is the greatest thread I've seen in a while

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·