Tweetovi

Blokirali ste korisnika/cu @RachitSaluja

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

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

    Just published - Our work using 3D convolutional neural networks for multiclass tissue segmentation of brain tumors and white matter hyperintensities

    Poništi
  2. proslijedio/la je Tweet
    4. pro 2019.

    Chefs also have to do a lot of hyperparameter tuning

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    26. stu 2019.

    Transformers v2.2 is out, with *4* new models and seq2seq capabilities! ALBERT is released alongside CamemBERT, implemented by the authors, DistilRoBERTa (twice as fast as RoBERTa-base!) and GPT-2 XL! Encoder-decoder with ⭐Model2Model⭐ Available on

    Poništi
  4. proslijedio/la je Tweet
    9. stu 2019.

    Stuck in Dubai airport for over 24 hours now without a get away, with my uncle( senior), sister, her husband and their 17 month old infant. Please help

    Prikaži ovu nit
    Poništi
  5. proslijedio/la je Tweet

    quality gets worse with each upcoming conference. Our reviews has extreme swings (8s and 1s). Reviewers giving low scores are nasty and bitter. Its like they are out on a war and want to destroy competition. Not healthy. What are the alternatives?

    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet
    3. stu 2019.

    Excellent early morning talk by showing how we can adapt neural nets for patients in the real world, where more than one disease may coexist.

    Poništi
  7. proslijedio/la je Tweet
    30. lis 2019.

    Which functions a neural network is capable to implement? Trees are the sparsest network, we study them in-depth! Ironically, understanding tree functions translate to peers in NN. To present at Read the thread for the previous work. In press:

    Poništi
  8. proslijedio/la je Tweet

    Our opinion piece that deep learning should guide neuroscience research just came out:

    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    25. lis 2019.

    If you run "pkill -f bert" and you were training one model with BERT and one with RoBERTa, then you kill two berts with one stone.

    Poništi
  10. proslijedio/la je Tweet

    Detectron2 is Facebook AI’s new object-detection platform, written in PyTorch and featuring a new, more modular design. The new library is flexible and extensible, scalable, easy to use, and able to provide fast training on single or multiple GPU servers.

    Poništi
  11. proslijedio/la je Tweet
    8. lis 2019.
    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    2. lis 2019.

    In new work with , we reconstruct the architecture, weights, and biases of unknown deep ReLU networks by observing their output. ReLU nets are piecewise linear functions - a lot of info is in the boundaries between linear regions. Full paper:

    Poništi
  13. proslijedio/la je Tweet

    Congrats to Professor Josh Tenenbaum for being named a recipient of a 2019 MacArthur Fellowship!

    Poništi
  14. proslijedio/la je Tweet
    25. ruj 2019.

    We ( and I) are curious to know your experience with calendar apps. Please fill this survey out. Shouldn't take more than 5-7 minutes. And please RT so we can reach a larger number + variety of people!

    Poništi
  15. proslijedio/la je Tweet
    25. ruj 2019.

    New preprint! Does V4’s tuning to hue on artificial stimuli tell us about its tuning to hue on natural images? Not really. (Twitterified in the thread below) with , , Matthew Smith, and

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet

    . is my role model. I feel so inspired and energized to be here at

    Poništi
  17. proslijedio/la je Tweet

    At today, we announced Google Research India - a new AI research team in Bangalore that will focus on advancing computer science & applying AI research to solve big problems in healthcare, agriculture, education, and more.

    Poništi
  18. proslijedio/la je Tweet
    10. ruj 2019.

    Computer vision to automatically assess infant neuromotor risk

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

    Preprint out on assessing infant neuromotor risk with computer vision  with

    Poništi
  20. proslijedio/la je Tweet
    9. kol 2019.

    What's the worst grant acronym you can remember? Can't possibly beat ARPA's current call for "Design Intelligence Fostering Formidable Energy Reduction and Enabling Novel Totally Impactful Advanced Technology Enhancements" (DIFFERENTIATE)

    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:

    ·