Vixtor Namaste

@_rdm_8

science enthusiast, random guy- daft punk, gorilla z.. Aleph Zero

Thailand
Vrijeme pridruživanja: svibanj 2014.

Tweetovi

Blokirali ste korisnika/cu @_rdm_8

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @_rdm_8

  1. proslijedio/la je Tweet
    23. sij

    I joined ten years (and a couple of days) ago. These are some of my favorite tweets. What’s your favorite tweet from the last ten years?

    Poništi
  2. proslijedio/la je Tweet
    19. sij

    El momento del bañista se transfiere a la pelota a través del agua y… • vía 💦🏐

    Poništi
  3. 7. sij
    Poništi
  4. proslijedio/la je Tweet
    2. sij

    To count an animal population if you can't catch them all: capture some, mark them, release, and capture again. The ratio of new vs already seen tells you something about the total number. Shown here is iterated mark-and-recapture with a Bayesian updates to belief about pop size

    Prikaži ovu nit
    Poništi
  5. proslijedio/la je Tweet

    Contrast is one hell of a drug

    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet
    4. pro 2019.

    Our new paper, Deep Learning for Symbolic Mathematics, is now on arXiv We added *a lot* of new results compared to the original submission. With (1/7)

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    4. pro 2019.

    BertViz now supports ALBERT, CamemBERT, CTRL, and other transformers through a simplified API: Can you guess which model below is BERT and which is ALBERT? (hint: cross-layer parameter sharing)

    Poništi
  8. proslijedio/la je Tweet
    21. lis 2019.

    A triangle with three right angles

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    19. ruj 2019.

    These tubes bring natural light indoors and help reduce energy costs

    Poništi
  10. proslijedio/la je Tweet

    An object in motion will stay in motion, unless acted upon by an external force

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    19. ruj 2019.

    Ever wondered, like me, if you can can turn hierarchical VAEs into a fully parallelizable lossless compression scheme? Wonder no more: new blogpost by Friso on "Bit-Swap", his recursive algorithm for doing just that: Disclaimer: he's my sibling 🤓

    Prikaži ovu nit
    Poništi
  12. 19. ruj 2019.
    Poništi
  13. proslijedio/la je Tweet
    17. ruj 2019.

    We've observed AIs discovering complex tool use while competing in a simple game of hide-and-seek. They develop a series of six distinct strategies and counterstrategies, ultimately using tools in the environment to break our simulated physics:

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    12. kol 2019.

    Buses appear to be reddish, though the pixels are not.

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet

    Professor Peter Bartlett of shares research on overfitting in deep networks, progress on interpolating prediction rules and how statistics give insight into deep networks. Hear how deep networks can give zero training error for regression loss:

    Poništi
  16. proslijedio/la je Tweet

    Surprising how simple ideas can yield such a good generative model! -Mean Squared Error loss on pixels -Non-autoregressive image decoder -Discrete latents w/ straight through estimator w/ & VQ-VAE-2: Code:

    Poništi
  17. 8. ruj 2019.

    200,000 Volt - Van de Graaff generator experiment!

    Poništi
  18. 7. ruj 2019.

    Max Greenfield - one of the greatest satire on tv

    Poništi
  19. 6. ruj 2019.

    'A Horse Loose In A Hospital' Lol ..

    Poništi
  20. proslijedio/la je Tweet
    6. ruj 2019.

    | hugs and consoles ISRO Chief K Sivan as he broke down. Communication from to the ground stations was lost earlier today.

    Prikaži ovu nit
    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:

    ·