Sourav Mishra

@srvmshr

EECS PhD at University of Tokyo. Previously Microsoft Research VRF. VirginiaTech & Stanford Biodesign alum. Apolitical. Advised by

Tokyo-to, Japan
souravmishra.net
Vrijeme pridruživanja: travanj 2013.
26 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @srvmshr

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

  1. 4. velj

    Real life Twitter threads

    When a cleaner finds 5 euros at a Finnish office
    1 korisnik označava da mu se sviđa
    Poništi
  2. proslijedio/la je Tweet
    4. velj

    Adding an independent Gaussian variable is the same as doing a heat diffusion on the density.

    143 proslijeđena tweeta 637 korisnika označava da im se sviđa
    Poništi
  3. 3. velj

    "Hello, do you need help with debugging?"

    1 reply 2 korisnika označavaju da im se sviđa
    Poništi
  4. 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 :)

    13 replies 72 proslijeđena tweeta 451 korisnik označava da mu se sviđa
    Poništi
  5. 1. velj

    Next new cool addition in fastai

    0:04
    This is a nice package for making pyplot animations more intuitive: All you do is call "camera.snap()" every time you re-do the plot.
    Poništi
  6. proslijedio/la je Tweet
    30. sij

    Why aren’t there more Country songs about neural nets? They are all about taking in loss, embracing that loss, making the loss flow through you until you become better

    15 replies 67 proslijeđenih tweetova 440 korisnika označava da im se sviđa
    Poništi
  7. 31. sij

    Not trying to hijack this discussion (hence the quoted post) but quite often I have seen arxiv submission as a means to beat the blind review system. Changing the abstract/title would greatly help the process for a start.

    Odgovor korisnicima @tdietterich @roydanroy
    But what if you want your paper to be publicly accessible already before the conference submission deadline?
    1 reply 3 korisnika označavaju da im se sviđa
    Poništi
  8. 30. sij

    Sorting never grows old, eh?

    Sorting Algorithms in C
    Poništi
  9. proslijedio/la je Tweet
    29. sij

    About a dozen of some of the most brilliant people in Science and Machine Learning I know left Tokyo in the past few months, after living here for years. Career opportunities are so much more interesting overseas. Japan needs to figure out a way to keep talent.

    12 replies 37 proslijeđenih tweetova 206 korisnika označava da im se sviđa
    Poništi
  10. proslijedio/la je Tweet
    27. sij

    I ran into every gym as a teenager, screamed "3-2-1-Kobe", and took shots I couldn't possibly make. Honestly, it never even crossed my mind the man could die. I mean, I knew he *could* like all of us. But to me, Kobe was larger than life.

    1 proslijeđeni tweet 5 korisnika označava da im se sviđa
    Poništi
  11. 26. sij

    Accepted papers at ALT 2020. Interesting set of papers.

    Poništi
  12. 25. sij

    Undergrads have much more creativity and think out of the box often. Our research regimens tend to make us think hard & deep with very narrow focus. Please encourage them. They are valuable peers having starters' issues, just like we did at some point.

    I overheard PhD folk subtly mocking undergraduate students, please don't do this. Don't forget that we all started somewhere. Undergraduates are the future, so treat them like it. Bring them into the conversation, help them feel respected & valued as people & as researchers.
    1 korisnik označava da mu se sviđa
    Poništi
  13. 24. sij

    True game changer for ML. Thanks to

    Google Dataset Search is now officially out of beta. "Dataset Search has indexed almost 25 million of these datasets, giving you a single place to search for datasets & find links to where the data is." Nice work, Natasha Noy and everyone else involved!
    Prikaži ovu nit
    1 korisnik označava da mu se sviđa
    Poništi
  14. 23. sij

    Need some CI measures for that. Can't believe the division is sharp without any standard deviation.

    There's a fine line between getting lucky often and actually being really good at probability and statistics.
    Poništi
  15. 23. sij

    Found a pretty useful thread for things being done except the Deep Learning warfront.

    Poništi
  16. proslijedio/la je Tweet
    22. sij

    The "Elements of Data Science" curriculum is now substantially complete: It is a notebook-based introduction to Data Science in Python for people with no prior experience in programming or statistics.

    12 replies 243 proslijeđena tweeta 802 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  17. 22. sij

    What do you use for your ML projects?

    1 reply 1 korisnik označava da mu se sviđa
    Poništi
  18. 22. sij

    JAX looks like the new PyTorch

    2 korisnika označavaju da im se sviđa
    Poništi
  19. 21. sij

    When LaTeX compiler said figure didn't fit the margins, I wasn't expecting to find it in the hallway.

    I feel like I stumbled into a SIGGRAPH paper by accident.
    Prikaži ovu nit
    2 korisnika označavaju da im se sviđa
    Poništi
  20. proslijedio/la je Tweet
    12. stu 2019.

    Method is also super simple: 1) Train a classifier on ImageNet 2) Infer labels on a much larger unlabeled dataset 3) Train a larger classifier on the combined set 4) Iterate the process, adding noise

    12 replies 38 proslijeđenih tweetova 215 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi

@srvmshr još nije objavio nijedan Tweet.

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:

    ·