Tweetovi

Blokirali ste korisnika/cu @dyamins

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

  1. 29. lis 2019.

    I would perhaps go further. It's precisely because AI has decided to use design principles similar to those exploited by evolution and development that it has started to work. The principles are properly understood as coming from biology, not traditional computer science.

    Prikaži ovu nit
    Poništi
  2. 29. lis 2019.

    I agree that the important point is not AI as such, but rather the design principles. I was fine with how we phrased it in the paper but I actually think the real source of the "three principles" is evolution and development itself.

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    28. lis 2019.

    Congratulations to my colleagues Daniel Yamins , Leo Guibas and students for their nominations to Best Paper Award!

    Poništi
  4. proslijedio/la je Tweet

    Our new address on Jane Stanford Way, increasing vegetable consumption, using particle physics to improve neuroscience, and more in this edition of the Stanford Psychology Department Newsletter! Check it out and subscribe to future editions here:

    Poništi
  5. proslijedio/la je Tweet
    17. ruj 2019.

    Come join us at Stanford - we have a faculty opening in the developmental group! Apply at:

    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet
    15. ruj 2019.
    Prikaži ovu nit
    Poništi
  7. 2. kol 2019.

    Very excited to see 's powerful new results on unsupervised learning accepted for an oral presentation at ! paper here: code here:

    Poništi
  8. proslijedio/la je Tweet
    24. srp 2019.

    Honored to be included alongside pioneers like Maneesh Sahani & others!

    Prikaži ovu nit
    Poništi
  9. 30. svi 2019.

    More deep thoughts from et. al: semi-supervised visual learning and unsupervised video learning . unsupervised methods really coming on to their own. @_alexandonian

    Poništi
  10. proslijedio/la je Tweet
    24. svi 2019.

    Really nice work! Wonder how it relates to deep embedding approach. actually we got > alexnet a couple mos. ago with much shallower network (resnet50). curious how we’re going to get close to supervised numbers without extra depth.

    Poništi
  11. proslijedio/la je Tweet
    18. svi 2019.

    V impressive talk from @defrietass &co using eyetracking, robots playing peekaboo, worlds & curiosity-driven deep learning to show that our tendency to attend to animate things cd arise if learning objective=maximise progress in ability to predict world!

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    10. svi 2019.
    Poništi
  13. proslijedio/la je Tweet
    2. svi 2019.

    New paper out in . Evolving Images for Visual Neurons Using a Deep Generative Network Reveals Coding Principles and Neuronal Preferences. with , , , , and .

    Poništi
  14. proslijedio/la je Tweet
    2. svi 2019.

    “Essentially all models are wrong, but some are useful” — here we demonstrate this pretty well.. I am excited about the future!

    Poništi
  15. 10. tra 2019.

    Neat new results on unsupervised visual learning from my student

    Poništi
  16. proslijedio/la je Tweet
    31. svi 2018.

    A blurb about and my recent Neuron paper was written up in !

    Poništi
  17. proslijedio/la je Tweet
    22. svi 2018.

    Can we define success (and key performance indicators) for cognitive computational neuroscience?

    Poništi
  18. proslijedio/la je Tweet
    2. svi 2018.

    7/ Across 100s of networks, a network’s task performance correlated with how well it predicted cortical responses

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    2. svi 2018.

    1/ Our paper using deep neural networks as a model of auditory cortex is out in today's issue of ! Here are a few highlights:

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    2. svi 2018.

    8/ Here’s the paper: (no paywall: ); net’s weights & biases released here: (thanks to !)

    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:

    ·