Shao-Hua Sun

@shaohua0116

Deep Learning, Computer Vision, Reinforcement Learning, Robot Learning, Meta-learning | PhD student in CS | GitHub: | From Taiwan

Los Angeles, CA
Vrijeme pridruživanja: listopad 2015.

Tweetovi

Blokirali ste korisnika/cu @shaohua0116

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @shaohua0116

  1. Prikvačeni tweet
    19. pro 2019.

    Our paper "Program Guided Agent" is accepted as a spotlight paper! We propose a modular framework that can accomplish tasks specified by programs, structured in a formal language, and achieve zero-shot generalization to more complex tasks. Paper:

    Poništi
  2. proslijedio/la je Tweet

    We need to be working to together to help combat the spread of the coronavirus Instead, & refuse to share information with Taiwan and is blocking critics on Twitter. This is unacceptable behavior and exemplifies undue Chinese influence in the UN.

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet

    The International Civil Aviation Organization, , which is a specialized U.N. agency, is apparently now blocking people on Twitter for making references to the fact that it excludes Taiwan (Taiwan's inclusion in international organizations a big issue for China)

    Poništi
  4. proslijedio/la je Tweet

    Also read more diverse books, the world is big and interesting

    Prikaži ovu nit
    Poništi
  5. 23. sij

    Sorry it’s actually meningococcal vaccine instead of polio

    Prikaži ovu nit
    Poništi
  6. 21. sij

    My preparation today: - Got four vaccinations at once for yellow fever, hepatitis A, typhoid, and polio - Got the medicine for malaria - Registered at

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    21. sij

    🚨New lecture series🚨 We've teamed up with to bring you the Deep Learning Lecture Series: 12 lectures covering a range of topics in Deep Learning - all led by DeepMind researchers, all free, and all open to everyone. Info & tickets:

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    18. sij

    that one collaborator on the paper that’s just happy to be there

    Poništi
  9. proslijedio/la je Tweet
    18. sij

    Tried mixed precision yet? Took 10 min to set up and my model runs almost 2x faster with same results. Vars and grads are still 32 bits so it usually doesn't affect predictive performance. E.g. in TF2, set option and make all input to your layers float16 (data, RNN states, ..):

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    17. sij

    The best career tip: Start writing. Publicly. No one will notice at first. Which is good. You’ll build the habit. You’ll find your voice. You’ll build a platform slowly, brick by brick. Over 5-10 years, you’ll build an asset very few have. The opportunities will shock you.

    Poništi
  11. 19. sij

    This is so sad. I have been a reviewer for 10+ conferences/journals. Each time, I block out N entire days on my calendar for N assigned papers. I make sure nothing can get in the way to delay my reviews. If I can't guarantee this, I choose not to accept the reviewer's invitation.

    Poništi
  12. proslijedio/la je Tweet
    26. pro 2019.
    Poništi
  13. 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
  14. proslijedio/la je Tweet
    24. pro 2019.

    since everyone loves/hates ranking authors based on their submissions/acceptances to , here's another rank i just created based on the # of letters in their names. does not show up in top-50 (one with all names, and the other with the first name segments only)

    Poništi
  15. 25. pro 2019.

    It’s sad that as a Taiwanese you can’t vote abroad. I’m gonna fly back to Taiwan next week from Los Angeles to vote for the Taiwanese Presidential election on 11 January 2020.

    Poništi
  16. 24. pro 2019.

    As asked by and others authors with >= 5 submissions sorted by acceptance rate Zhiyuan Li, Mingyuan Zhou, Deva Ramanan 4/5 80.0% Le Song 7/9 77.8% Jimmy Ba 6/8 75.0% Martin Jaggi 5/7 71.4% Abhinav Gupta 5/7 71.4% Pushmeet Kohli 6/9 66.7% Max Welling 5/8 62.5%

    Poništi
  17. 23. pro 2019.

    authors with > 5 submissions: 32 Sergey Levine 20 Yoshua Bengio 16 Cho-jui Hsieh 14 Pieter Abbeel 13 Liwei Wang, Tom Goldstein, Chelsea Finn, Bo Li, Jun Zhu # of accepted papers: 13 Sergey Levine 7 Le Song, Jun Zhu 6 Cho-jui Hsieh, Jimmy Ba, Liwei Wang, Pushmeet Kohli

    Poništi
  18. 23. pro 2019.

    My Gmail app just found a hotel reservation in Hawaii from today’s arXiv update. Apparently, 1. Everyone is going to Hawaii for their holidays. 2. There's a hotel named Dong Yang Inn in Hawaii. 3. There's a paper author named Dong Yang from today's arXiv papers. Thanks,

    Poništi
  19. 21. pro 2019.

    Besides this amazing AC, I am also super grateful for my reviewers who are extremely thorough and responsive. All 3 reviewers replied to our rebuttal within 36 hours after I posted it two days before the deadline. The scores went from 6 1 6 to 8 6 8.

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

    For the last 10 months, I have been working on Program Synthesis. Our (me and ) first paper was just accepted () and I think it's my favorite research I have done?: (1/N)

    Prikaži ovu nit
    Poništi
  21. proslijedio/la je Tweet
    20. pro 2019.

    In ICLR last year, we show that existing few-shot classification methods perform poorly across different domains (). This year in , we show how we can make few-shot learners generalize better to unseen domains ()!

    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:

    ·