Max Jaderberg

@maxjaderberg

Advance artificial intelligence. Research Scientist

London, England
Vrijeme pridruživanja: travanj 2009.

Tweetovi

Blokirali ste korisnika/cu @maxjaderberg

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

  1. Prikvačeni tweet
    17. lip 2019.

    Check out our work on human-level agents in Quake III Arena in ! This has been an immensely fun project, especially as we are scaling these agents to full tournament maps on multiple game modes

    Poništi
  2. proslijedio/la je Tweet
    29. sij

    Our paper 'A distributional code for value in dopamine-based reinforcement learning' on the cover of ! Read it here: Shout out to the amazing artists/designers at who make this possible, while we get to focus on the research.

    Poništi
  3. proslijedio/la je Tweet
    15. sij

    When neuroscience and AI researchers get to chatting, cool stuff happens! My first, and I hope not last, trip into neuroscience has been published in Nature. 1/

    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    15. sij

    We have 2 papers published in today! 🎉 One describes AlphaFold, which uses deep neural networks to predict protein structures with high accuracy. AlphaFold made the most accurate predictions at the 2018 scientific community assessment CASP13. 1/4

    Prikaži ovu nit
    Poništi
  5. proslijedio/la je Tweet
    25. pro 2019.

    Rephrasing with my own words: DL is a collection of tools to build complex modular differentiable functions. These tools are devoid of meaning, it is pointless to discuss what DL can or cannot do. What gives meaning to it is how it is trained and how the data is fed to it

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

    Have you heard of "RL as Inference"? ... you might be surprised that this framing completely ignores the role of uncertainty! (confusing, since it talks a lot about "posteriors") Our spotlight tries to make sense of this:

    Poništi
  7. 17. pro 2019.

    This builds upon the work of Skorokhodov and Burtsev, 2019

    Prikaži ovu nit
    Poništi
  8. 17. pro 2019.

    This theoretical model predicts the ability to easily find any smooth bounded low dimensional pattern in the local loss landscape of a big enough neural net. And you can even decorate your minima (red dot is minima)

    Prikaži ovu nit
    Poništi
  9. 17. pro 2019.

    Looking for the perfect gift this holiday season? Well look no further than your neural net's loss landscape where turns out you can find anything you want! New work from at

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet

    Another big year comes to a close at , from StarCraft to the Serengeti, protein folding to wind power. Feels like only yesterday we were starting out as a tiny start-up in an unfashionable area! 1/3

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    2. pro 2019.

    It's time for reinforcement learning researchers to take the domains they train on seriously. Training on a set of fixed scenarios/levels leads to brittle policies that don't generalize. Procedural content generation can help, argue and I here:

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet

    Anything that can be implemented in JAX *will* be implemented in JAX. Here's a differentiable path tracer (and a tutorial!) Blog Post: Code:

    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet
    25. stu 2019.

    For anyone in the Oxford area next week - I'll be leading a workshop for learning to code quantum computers using . Aimed at beginners to QM and quantum circuits with no prior experience required! Sign up opens tomorrow here

    Poništi
  14. proslijedio/la je Tweet
    21. stu 2019.

    This feels like a real breakthrough: Take the same basic algorithm as AlphaZero, but now *learning* its own simulator. Beautiful, elegant approach to model-based RL. ... AND ALSO STATE OF THE ART RESULTS! Well done to the team at

    Poništi
  15. proslijedio/la je Tweet
    12. stu 2019.

    1/ Excited to share our NeurIPS paper, Geometry-Aware Neural Rendering! . With , , and the rest of the robotics team

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    13. stu 2019.

    This week on the Nature cover: Game plan. AI program learns to play StarCraft II to Grandmaster level. Browse the issue here:

    Poništi
  17. proslijedio/la je Tweet
    6. stu 2019.

    Happy birthday Nature!!! 150 Years!!!! my last project with so many amazing people: Alex Gates, Alice Grishchenko Claire Welsh. Enjoy this gorgeous must-see video:

    Poništi
  18. 6. stu 2019.

    If you’re interested in some of the high level ideas of evolutionary methods in the deep learning age have a read of this!

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

    peer review in machine learning is broken

    Poništi
  20. proslijedio/la je Tweet

    AlphaStar vs Serral Hype! Current score 1-0 for

    Prikaži ovu nit
    Poništi
  21. proslijedio/la je Tweet
    1. stu 2019.
    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:

    ·