Noor Sajid

@nsajidt

I am interested in understanding neuronal adaptation post system-level shocks and active exploration. || PhD Student 🧠

Vrijeme pridruživanja: listopad 2013.

Tweetovi

Blokirali ste korisnika/cu @nsajidt

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

  1. proslijedio/la je Tweet
    2. velj

    The 1986 classic 'Parallel Distributed Processing' uses the term 'threshold function' instead of 'rectified linear unit'. I prefer the 1986 version :)

    Poništi
  2. proslijedio/la je Tweet
    30. sij

    Check out my recent work with and on how active inference is, like, totally better than reinforcement learning:

    Poništi
  3. proslijedio/la je Tweet
    22. sij

    Active inference on discrete state-spaces: a synthesis.

    Poništi
  4. proslijedio/la je Tweet
    24. ruj 2019.

    First-order methods are great but we never know which one to use. In our paper (w/ John Duchi) , we tell you which one to use and when, depending on the quadratic convexity of the constraints. Also, adaptivity matters. To appear at as an oral.

    Poništi
  5. proslijedio/la je Tweet
    5. pro 2019.

    1/ Why do wide, random neural networks form Gaussian processes, *regardless of architecture*? Let me give an overview in case you are too lazy to check out the paper or the code . The proof has two parts…

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

    We introduce LOGAN, a game-theory motivated algorithm, which improves the state-of-the-art in GAN image generation by over 30% measured in FID: Here are samples showing higher diversity:

    Poništi
  7. 28. stu 2019.

    So it's entirely possible to get better statistical and behavioural accuracy with dual lesions, relative to singular lesions (Sprague: 1966). Fascinated by compensatory augmentation in my simulated agent - paradoxical af.

    Poništi
  8. proslijedio/la je Tweet
    23. stu 2019.
    Poništi
  9. proslijedio/la je Tweet
    31. lis 2019.

    New paper! Using a new VR-based grasping task and fMRI-DCM, we show that a task-induced attentional set to vision *or* proprioception is reflected by diametrical changes in neuronal gain in visual (V1) vs somatosensory (S2) areas.

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    Poništi
  11. proslijedio/la je Tweet
    8. lis 2019.

    New preprint: we show that regulating the impact of visual vs proprioceptive prediction errors helps with goal-directed hand movements under intersensory conflicts.

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet

    One of the saddest things about the field of AI is how ahistorical and siloed it is. Some of the key debates of our day (e.g. innate priors vs "blank slate" minds) have been going on for many decades, across multiple fields, and deep learners have *zero* knowledge of this context

    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet

    here's a fun one

    Poništi
  14. proslijedio/la je Tweet
    1. lis 2019.

    Do you formally know Monte-Carlo and TD learning, but don't intuitively understand the difference? This is for you. (with )

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    27. ruj 2019.

    My excellent PhD student had the last paper of his PhD published today: - work with and Thomas Parr... He devised a cool model of selective attention within a Bayesian framework (active inference) 1/

    Prikaži ovu nit
    Poništi
  16. 25. ruj 2019.

    Our preprint, with Karl Friston, and myself, on 'Demystifying active inference' is now out on ariXv: . It is a simplified perspective on active inference using a T-maze simulation!

    Poništi
  17. 17. ruj 2019.

    Do we exist in multiple worlds? Induced by measurements that are any interactions that cause a quantum system to become entangled with the environment, creating a branching into separate worlds, and (agents) we are any system that brings about such an interaction.

    Poništi
  18. 5. ruj 2019.

    I want Planck constant level of uncertainty, all-day and everyday.

    Poništi
  19. proslijedio/la je Tweet

    A straight line may be the shortest 𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 between two points, but a "Brachistochrone" curve is the path of least time

    Prikaži ovu nit
    Poništi
  20. 9. kol 2019.

    How can the brain modify its own structure?

    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:

    ·