Debanjan Mahata

@debanjanbhucs

Applied Research Scientist, Machine Learning, NLP, Data Mining, Information Retrieval, Tabla, Hindusthani Classical Music... enthusiast. Opinions are my own.

New York, USA
Vrijeme pridruživanja: rujan 2009.

Tweetovi

Blokirali ste korisnika/cu @debanjanbhucs

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

  1. prije 8 sati

    To the reviewers - Attention is all you need.

    Poništi
  2. proslijedio/la je Tweet

    Absolutely perfect reminder. Thanks

    Poništi
  3. proslijedio/la je Tweet
    2. velj

    I’m confused. How did we conclude that the best way to prepare kids for the future is to cluster them into a setting where they are organized by age, into grades, and forced to learn the same things, at the same time and pace, 7 hours a day, 5 days a week, for 12 + years? Huh?

    Poništi
  4. 2. velj

    Someone just expressed what I wouldn’t have expressed. Thanks! Currently in O1 after 4 H1B lottery attempts and in the queue with a EB1A accepted. The wait simply doesn’t end.

    Poništi
  5. 2. velj

    Looking forward to my trip to CMU.

    Poništi
  6. proslijedio/la je Tweet
    2. velj
    Odgovor korisnicima

    Thanks! I've recently come across George Dyson's formulation: "Any system simple enough to be understandable will nto be complicated enough to behave intelligently, while any system complicated enough to behave intelligently will be too complicated to understand".

    Poništi
  7. proslijedio/la je Tweet
    2. velj

    Wow: Google's "Meena" chatbot was trained on a full TPUv3 pod (2048 TPU cores) for **30 full days** - That's more than $1,400,000 of compute time to train this chatbot model. (! 100+ petaflops of sustained compute !)

    Poništi
  8. proslijedio/la je Tweet
    2. velj

    Happy multi-cultural palindrome day! Works whether your preferred date format is MM/DD/YYYY or DD/MM/YYYY or YYYY/MM/DD.

    Poništi
  9. proslijedio/la je Tweet
    31. sij

    Today we announce a novel, open-source method for text generation tasks (e.g., summarization or sentence fusion), which uses edit operations instead of generating text from scratch, leading to less errors and faster model execution. Read about it below.

    Poništi
  10. proslijedio/la je Tweet
    31. sij

    Reduce Bias in Text? Find it out in our podcast with 📢

    Poništi
  11. proslijedio/la je Tweet
    31. sij

    All papers to the Sixth IEEE International Conference on Multimedia Big Data ( 2020) will be accepted through EasyChair

    Poništi
  12. proslijedio/la je Tweet
    30. sij

    New dataset coming soon for recsys and Twitter will release a large public dataset of ~200M public tweets, obtained by subsampling within ~2 weeks, that contains engagement features, user features, and tweet features.

    Poništi
  13. proslijedio/la je Tweet
    29. sij

    news prompt me to post this upcoming paper I was lucky to contribute to (led by ), on the complicated perceptions of misinformation and censorship in China. In our (qualitative, interview-based) study, health misinformation was a top-level concern 1/

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    28. sij

    Pleased to announce the release of PMIndia, a parallel corpus featuring 13 languages of India paired with English. Corpus: Paper: .Part of the project.

    Poništi
  15. proslijedio/la je Tweet
    25. sij
    Poništi
  16. proslijedio/la je Tweet
    25. sij

    Check out our recent work on multilingual pretraining!

    Poništi
  17. proslijedio/la je Tweet
    24. sij

    This is an invaluable resource. Thanks and team! 2 wishlist items: discussion of the coverage (how can we monitor, estimate on ongoing basis) and perhaps an estimate of future availability, plans or needs (how to ensure this continues to be available forever!)

    Poništi
  18. proslijedio/la je Tweet
    21. sij

    102: Dina Demner-Fushman tells us about processing biomedical scientific literature, understanding clinical notes, answering consumer health questions, and the challenges involved in each of these applications.

    Poništi
  19. 19. sij

    No! I will not ask the authors why you didn't use BERT when there is no need to do so.

    Poništi
  20. 18. sij

    A snowy afternoon leading to a snowy evening - What could be a better time to start reviewing for

    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:

    ·