Alexander Lavin

@theAlexLavin

I code, I run, I yoga. Rocket scientist ➡️ AI engineer; . WIP:

Lavender Town
Vrijeme pridruživanja: listopad 2012.

Tweetovi

Blokirali ste korisnika/cu @theAlexLavin

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

  1. Prikvačeni tweet
    6. sij 2016.

    Thank you ! It's an honor to be listed amongst such accomplished and promising company 😄

    Poništi
  2. proslijedio/la je Tweet
    24. sij

    👋🏻 I'm looking for a Head of Product, experienced with AI/data platforms for b2b, must be NYC 🚀 Email me if you have a rec! lavin@latentsci.com

    Poništi
  3. proslijedio/la je Tweet
    31. sij

    On Tuesday, in my class, we have learnt that all a neural net does is stretching / contracting the space fabric. For example this 3-layer net (1 hidden layer of 100 positive neurons) gets its 5D logits (2D projections) linearly separable by the classifier hyperplanes (lines).

    Poništi
  4. 24. sij

    👋🏻 I'm looking for a Head of Product, experienced with AI/data platforms for b2b, must be NYC 🚀 Email me if you have a rec! lavin@latentsci.com

    Poništi
  5. proslijedio/la je Tweet
    10. sij

    New preprint from the lab: "Individual differences among deep neural network models." Work with , , and Courtney Spoerer. below. 1/7

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

    Working on a startup feels like

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    30. stu 2019.

    I really like blog posts which try to teach the reader old ideas. This one is about ways to visualise concepts in information theory, mentioned (cautiously) in solution 8.8 in Mackay's book (remember, positive areas can be negative quantities!) By

    Poništi
  8. proslijedio/la je Tweet
    26. pro 2019.

    Bayesian methods are *especially* compelling for deep neural networks. The key distinguishing property of a Bayesian approach is marginalization instead of optimization, not the prior, or Bayes rule. This difference will be greatest for underspecified models like DNNs. 1/18

    Prikaži ovu nit
    Poništi
  9. 3. sij

    😆 here are a few for the AI folks: - Meta - Eager - Autodiff - Bert - Elmo - GPT-2 - Jax - Pyro - YOLO - Reviewer 2 - Jurgen Schmidhuber

    Poništi
  10. proslijedio/la je Tweet
    23. pro 2019.

    Working for Microsoft cost me $200M I was 23 and I had a zero-risk opportunity to join a startup as employee #1. That startup is now worth $20 billion. Anyone who makes great software will run into situations like this— it's a measure of how valuable software is in the world

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    18. pro 2019.
    Odgovor korisnicima

    The recording of the will be available immediately after the livestream finishes

    Poništi
  12. proslijedio/la je Tweet
    20. pro 2019.

    AAAI is delighted to announce the election of ten new Fellows for their lasting contributions to the breadth of , including Planning, CSP, Mult-Agent Systems, Semantic Web, NLP, Vision, Deep Learning and HRI. These new fellows will be inducted at in New York.

    , , i još njih 2
    Poništi
  13. proslijedio/la je Tweet
    19. pro 2019.

    Oversimplify your science. I'll start: I rub dirty water on round jello.

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    19. pro 2019.

    Quantum picturalsm as pioneered by Bob Coecke and others in our Quantum Group featured on the blackboard of the latest episode. Just like in Nicola Pinzani’s, Stefano Gogioso’s and Bob Coecke’s 2019’s LiCS paper, it was used by the snakes to study time travel.

    Poništi
  15. proslijedio/la je Tweet
    14. pro 2019.

    Slides for my 2nd NeurIPS talk. "Materials Matter: How biologically inspired alternatives to conventional neural networks improve meta-learning and continual learning" Covers Differentiable Plasticity & ANMML

    Poništi
  16. 15. pro 2019.

    do {🧠++} while🧘‍♂️; "...behavioral interventions like yoga may hold promise to mitigate age-related and neurodegenerative declines"

    Poništi
  17. 13. pro 2019.

    . on 🐭 models: We create artificial diseases and put them in artificial animals to create artificial cures. This is a fundamental problem. 👆10x so in neurodegen!!

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

    Yoshua Bengio: "Another topic that is much on the minds of people in deep learning. Systematic generalization ... Today’s machine learning doesn’t know how to do that." I completely agree. It's a foundational problem that we need to be studying.

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

    In case you were wondering...

    Poništi
  20. proslijedio/la je Tweet
    10. pro 2019.

    Imagine discussing machine learning and medicine in Bermuda with other top researchers across informatics, policy, clinical translation, and ML. That's what this meeting is about! Abstracts due soon. If you have questions, send me a note.

    Poništi
  21. proslijedio/la je Tweet
    8. pro 2019.

    In Vancouver all week 🙌🏻 Hit me up to chat probabilistic programming, Gaussian processes, Bayesian optimization, Bayesian NNs. Especially if it's applied to neurodegen diseases, and life sciences --

    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:

    ·