Il Memming Park

@memming

Computational neuroscientist. I eat spike trains. How do population of neurons compute? Assistant professor at Stony Brook University.

Stony Brook, NY, USA
Vrijeme pridruživanja: kolovoz 2008.

Tweetovi

Blokirali ste korisnika/cu @memming

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

  1. Prikvačeni tweet
    10. sij

    Fresh on . Latent trajectory analysis of MT ensemble spike trains reveals mis-aligned subspaces of stimulus and decision codes. Collaboration with Huk lab.

    Prikaži ovu nit
    Poništi
  2. prije 5 minuta
    Poništi
  3. proslijedio/la je Tweet
    31. sij

    We have an amazing lineup of speakers (2 days worth!): Day 1:, , , Krinstin Branson, , , , ,

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

    Cosyne schedules are up! Be sure to come to our workshop (organized w/ and Matt Whiteway) "Interpretable computational neuroscience: What are we modeling and what does it have to do with the brain?"

    Prikaži ovu nit
    Poništi
  5. 27. sij

    If they do live-stream (starting this morning), it'll be on

    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet
    26. sij

    Metastable attractors explain the variable timing of stable behavioral action sequences

    Poništi
  7. proslijedio/la je Tweet

    The professor who does not answer your email probably feels hopelessly overwhelmed. I was surprised that the average feeling is so strong.

    Poništi
  8. proslijedio/la je Tweet
    Poništi
  9. proslijedio/la je Tweet
    18. sij

    My paper is out on ! We explored movement signals in visual cortex and found a lot of surprising things.

    Prikaži ovu nit
    Poništi
  10. 18. sij

    I believe they plan to record most of the talks and might live steam. I'll keep y'all posted.

    Prikaži ovu nit
    Poništi
  11. 17. sij

    Computational neuroscience workshop with excellent speakers including Vijay Balasubramanian, Nicolas Brunel, Jim DiCarlo, Stefano Fusi, , Barry Richmond, Bruno Averbeck, , , and many more. Happening on campus.

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    16. sij

    New paper with and , which unifies a panoply of decision-making models (1D accumulation-to-bound, multi-D race models, discrete stepping, hybrid ramp+step models, collapsing bounds, etc) into single recurrent-state-space modeling framework!

    Poništi
  13. 16. sij
    Poništi
  14. proslijedio/la je Tweet
    13. sij

    Insights* from the interface of theory and experiment: -Hardest thing for a mathematician to learn: just because the math is right doesn't mean the model is. -Hardest thing for a biologist to learn: just because the model is wrong doesn't mean it's not useful.

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

    For years, I simulated perceptual/cognitive model predictions by running a large but fixed number of simulations (whether 250 or 10,000). Until Bas and Luigi opened my eyes to the errors of my way. Check out their preprint for a powerful combo of enlightenment and code.

    Poništi
  16. proslijedio/la je Tweet
    16. tra 2019.

    "Inverted Encoding Models Reconstruct an Arbitrary Model Response, Not the Stimulus"; sounds very interesting...

    Poništi
  17. 11. sij

    Interpretable AI workshop, April 3 (Friday), 2020 in NYC. Free event but requires registration. Space is limited!

    Poništi
  18. 10. sij

    Methods: all analysis of latent space was done using vLGP . MT ensemble was well described by 4-D latent space. (also 's brilliant work)

    Prikaži ovu nit
    Poništi
  19. 10. sij

    Furthermore, the choice-correlated variability in the null-space appears much later during the trial, suggesting that this is feedback information from the decision-making process! This information could be useful for learning and adaptation.

    Prikaži ovu nit
    Poništi
  20. 10. sij

    Our analysis suggests that a large portion of choice-correlated variability in MT is in a subspace NOT aligned with the sensory encoding and lives in the *null-space*!

    Prikaži ovu nit
    Poništi
  21. 10. sij

    But it could also be due to feedback, i.e., noisy temporal integration of sensory evidence can be contributing to the apparent variability in MT. Depending on the alignment between the sensory-coding subspace and the choice-correlation, there are several possibilities:

    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:

    ·