Ben Lansdell

@benlansdell

Neuroscience, applied mathematics, deep learning, causality

Vrijeme pridruživanja: travanj 2009.

Tweetovi

Blokirali ste korisnika/cu @benlansdell

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

  1. proslijedio/la je Tweet
    20. sij

    Given the smoothness of videos, can we learn models more efficiently than with ? We present Sideways - a step towards a high-throughput, approximate backprop that considers the one-way direction of time and pipelines forward and backward passes.

    Poništi
  2. proslijedio/la je Tweet
    12. sij

    What makes a good lab? Are group meetings really the best way? In the we recently reexamined how we organize our weeks, and then redesigned everything in a systematic way (100% democratically!). I blogged about our design process and takeaways:

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet

    Can a neuroscientist understand a virtual rodent? Authors took a realistic 3D animal, trained deep RL to control it during tasks, and then used neuroscience techniques to peak inside this animal... [Josh Merel, Diego Aldorando, Greg Wayne, ]

    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    18. stu 2019.
    Prikaži ovu nit
    Poništi
  5. proslijedio/la je Tweet
    18. stu 2019.

    Trends in Cognitive Sciences Theories of Error Back-Propagation in the Brain

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

    “I’m going to work on artificial general intelligence.” – John Carmack

    Prikaži ovu nit
    Poništi
  7. 5. stu 2019.
    Poništi
  8. proslijedio/la je Tweet
    4. stu 2019.

    Happy to share this work, with Claudia Clopath, Rob Froemke & Co. We study how cortical synapses can rely on limited error information to decide whether to potentiate/depress. To approach theoretical capacity, plasticity must have certain properties. 1/2

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet

    Our opinion piece that deep learning should guide neuroscience research just came out:

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

    1/ New tweeprint from my lab! This one is work done by the amazing , and was inspired by the work of and (who was also a collaborator in this project).

    Prikaži ovu nit
    Poništi
  11. 11. lis 2019.

    Such a nice piece on behavior. At least the first two points, today's neglect of an organism's umwelt and the focus on stim -> behavior arcs instead of loops, seem, ironically, to be relics of last century's behaviorism.

    Poništi
  12. proslijedio/la je Tweet
    9. lis 2019.

    Formatted version of FB discussion instrumental convergence with , me and others.

    Poništi
  13. proslijedio/la je Tweet

    Three biological principles for understanding behavior: materiality, agency, and historicity. (a followup (?) to Neuroscience Needs Behavior, from , Asif Ghazanfar)

    Prikaži ovu nit
    Poništi
  14. 9. lis 2019.

    Interesting articles on current issues in AI... including our own on learning-to-learn.

    Poništi
  15. proslijedio/la je Tweet
    6. lis 2019.

    New work: we built a meta-learning algorithm for an agent to discover the causal and effect relations from its visual observations and to use such causal knowledge to perform goal-directed tasks. Paper: Joint work w/

    Poništi
  16. proslijedio/la je Tweet
    2. lis 2019.

    In new work with , we reconstruct the architecture, weights, and biases of unknown deep ReLU networks by observing their output. ReLU nets are piecewise linear functions - a lot of info is in the boundaries between linear regions. Full paper:

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

    New preprint! Does V4’s tuning to hue on artificial stimuli tell us about its tuning to hue on natural images? Not really. (Twitterified in the thread below) with , , Matthew Smith, and

    Prikaži ovu nit
    Poništi
  18. 23. ruj 2019.
    Tweet je nedostupan.
    Poništi
  19. proslijedio/la je Tweet
    13. ruj 2019.

    Like hundreds of other neuroscientists, I just flew to Berlin for . What’s the carbon cost of all this flight? I’m thankful to have received the list of origins for the 565 registrants, so let's see! We’re from around the world:

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    12. ruj 2019.

    It's that time of year - the call for workshops is out! Here's what you need to know: Cosyne 2020 Workshops will be held on 02-03 March 2020 in Breckenridge, CO. PRE-PROPOSAL DEADLINE: 06 October 2019 FULL PROPOSAL DEADLINE: 10 November 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:

    ·