Howard Chen

@howard50b

Cornell CS / ASAPP NLP Research

New York, NY
Vrijeme pridruživanja: kolovoz 2013.

Tweetovi

Blokirali ste korisnika/cu @howard50b

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @howard50b

  1. proslijedio/la je Tweet
    15. sij

    We are excited to announce the release of the panoramas for the Touchdown tasks to the StreetLearn dataset! This update adds 29k panoramas, and we verify results for both VLN and SDR consistent with the original paper. Details in this arxiv paper:

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

    If you live in Taiwan ROC, please vote this weekend. We cannot leave things to chance.

    Poništi
  3. proslijedio/la je Tweet
    26. pro 2019.
    Poništi
  4. proslijedio/la je Tweet
    24. pro 2019.

    I support for President. I just finished reading his book "The War on Normal People". I don't agree with everything, but it is a refreshingly coherent plan for addressing the inexorable tide of automation that is progressive in the truest sense of that word.

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

    Interviewer: What's your biggest strength? Me: Expert in Machine learning Interviewer: What's 9 + 10? Me: 5 Interviewer: Nope. 19 Me: It's 14 Interviewer: Wrong. 19 Me: It's 19. Interviewer: What's 2 + 2? Me: 19 Interviewer: You'll overfit right in!

    Poništi
  6. proslijedio/la je Tweet

    Had on Deep Cuts to talk UBI, automation, and being goth.

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

    Can we use deep RL to learn from data, rather than from online interaction? Aviral Kumar discusses challenges and recent work in this area, including our algorithm BEAR for fully-off policy RL:

    Poništi
  8. proslijedio/la je Tweet
    25. stu 2019.

    This has been 2 years and 3 papers in the making: direct mapping of natural language instructions and first-person observations to continuous velocity control. Yep, we learn the entire pipeline with a single interpretable neural model! ❤️

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    21. stu 2019.

    Machine learning conferences getting more geopolitical. Apparently, organizers of this year's ICCV conference were forced to change Taiwan from a country to a region, after the usual slide that listed accepted papers / country listed Taiwan as a country.

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    18. ruj 2019.

    Most NLP models treat numbers (e.g., “91”) in the same way as other tokens, i.e., they embed them as vectors. Is this a good representation for downstream numerical tasks such as DROP, math questions, etc? Yes! Pre-trained vectors (BERT, GloVe, ELMo) know numbers.[1 / 6]

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet

    In today's edition of the is the most incredibly creative group of supporters of any candidate ever, here is 's unofficial anime opening credits featuring anime and among others.

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    Poništi
  13. proslijedio/la je Tweet

    Who is John Yang?

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet

    I did the , looks like may have miscounted 🤦🏻‍♂️🙋🏻‍♂️

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    3. ruj 2019.
    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    17. kol 2019.

    “HK social media users have started writing posts in phonetically-spelled Cantonese words as a means to shut out mainland internet trolls. People who are not Cantonese speakers will be baffled by the messages, which are likely to confuse even NLP systems.”

    Poništi
  17. proslijedio/la je Tweet

    An academic walks into a bar and Reviewer 3 says this is not the joke he would have told.

    Poništi
  18. proslijedio/la je Tweet
    9. kol 2019.
    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    31. srp 2019.

    Hong Kong protestors are on another level. Here they’re using lasers to avoid facial recognition cameras. A cyber war against Chinese artificial intelligence.

    Poništi
  20. proslijedio/la je Tweet

    If you care more about your family and your kids than my neckwear, enter your zip code at and see what $1,000 a month would mean to your community. I have done the math, and it's a big difference.

    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:

    ·