Tweetovi

Blokirali ste korisnika/cu @jdrugowitsch

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

  1. Prikvačeni tweet
    12. sij

    New on the distribution of information across neurons of large neural populations, with MohammadMehdi Kafashan, , , @RNugeiraNeuro, Iñigo Arandia-Romero, Chris Harvey, and :

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    prije 20 sati

    Happy to announce CoSMo 2020! Aug 2-16 @ London, ON We have revamped our summer school on Computational Sensory-Motor Neuroscience to integrate traditional modeling with modern ML! With

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    29. sij

    Our article is out today ! Great collaboration with Nicole Peltier, Akiyuki Anzai, Greg DeAngelis, and . See below for a short description 1/N

    Prikaži ovu nit
    Poništi
  4. 25. sij

    (credit: New Yorker)

    Poništi
  5. proslijedio/la je Tweet
    15. sij

    Distributional reinforcement learning (RL) algorithms predict the entire ‘distribution’ of future rewards not just the average and improve the AI performance. This work presents evidence that the brain’s dopamine system uses distributional RL.

    Prikaži ovu nit
    Poništi
  6. 15. sij

    A quick reminder to never cite predatory journals: one of them just published an almost verbatim copy of @AKutschireiter's (lab member) previous article

    Poništi
  7. proslijedio/la je Tweet
    13. sij

    PhD and Post-Doc positions are available in my lab. Do you love math, respect philosophy and have a passion for comp neuro? Then you can be a good fit. Talented candidates, please send me a CV to ruben.moreno@upf.edu. Funding from HHMI

    Poništi
  8. 12. sij

    Furthermore, until the DOI link works, you can access the preprint at

    Prikaži ovu nit
    Poništi
  9. 12. sij

    ... and here is Ramon's correct Twitter handle:

    Prikaži ovu nit
    Poništi
  10. 12. sij

    Overall, these findings suggest that the brain uses a widely distributed, but nonetheless redundant code that supports recovering most information from smaller subpopulations.

    Prikaži ovu nit
    Poništi
  11. 12. sij

    Furthermore, we found that most neurons in visual cortex contain relevant information about the stimulus. Notably, for the purpose of decoding, we found that a random set of neurons in visual cortex is as good as an "optimized" set of neurons.

    Prikaži ovu nit
    Poništi
  12. 12. sij

    This predicts that tens of thousands of neurons are required to encode 95% of the information about visual stimulus direction, a number much smaller than the number of neurons in V1.

    Prikaži ovu nit
    Poništi
  13. 12. sij

    Using the theory of information-limiting correlations, we compartmentalized noise correlations into information-limiting and non-limiting components, and then extrapolated to predict how information grows when neural populations are even larger.

    Prikaži ovu nit
    Poništi
  14. 12. sij

    We found that information scales sublinearly with population size, due to the presence of correlated noise in these populations.

    Prikaži ovu nit
    Poništi
  15. 12. sij

    To distinguish these possibilities we recorded from primary visual cortex of mice observing moving visual stimuli, and asked how information about the motion direction is distributed across hundreds of simultaneously recorded neurons.

    Prikaži ovu nit
    Poništi
  16. 12. sij

    We asked how information is distributed across such populations within a given brain area. If distributed roughly evenly, total information scales linearly with the number of recorded neurons. If highly redundant, total information saturates.

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    9. sij

    We're hiring! Join our lab for a postdoc in computational neuroscience in New York

    Poništi
  18. 27. pro 2019.

    For some holiday fun: adaptive worst-case Minesweeper

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

    The Williamson Lab is still looking to hire postdocs! Anyone interested in using cool technologies (2P, optogenetics, ephys, and machine learning) to study how auditory cortical organization leads to purposeful behavior, just get in touch! Please retweet!

    Poništi
  20. 5. pro 2019.

    ... and for those interested in the nitty-gritty details, here is the code: 8/

    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:

    ·