Barret Zoph

@barret_zoph

Senior Research Scientist at Google Brain.

San Francisco, CA
barretzoph.github.io
Vrijeme pridruživanja: studeni 2016.
1 fotografija ili videozapis Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @barret_zoph

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

  1. proslijedio/la je Tweet
    18. stu 2019.

    RandAugment: Practical automated data augmentation with a reduced search space Decreasing the search space in a clever way avoids the need to perform highly expensive computation search. ie. NAS→EfficientNets, AutoAugment→RandAugment Might be useful for domain randomization.

    *New paper* RandAugment: a new data augmentation. Better & simpler than AutoAugment. Main idea is to select transformations at random, and tune their magnitude. It achieves 85.0% top-1 on ImageNet. Paper: Code:
    1 reply 25 proslijeđenih tweetova 137 korisnika označava da im se sviđa
    Poništi
  2. proslijedio/la je Tweet
    13. pro 2019.

    Tomorrow I'll be talking about JAX MD: a hardware accelerated, end-to-end differentiable, molecular dynamics library at the ML4PS at 9:20am (along with tons of amazing speakers). Paper: Code: Colab:

    "We aim to bring together computer scientists, mathematicians and physical scientists who are interested in applying machine learning to various outstanding physical problems, including in inverse problems and approximating physical processes." ✨🌍✨
    38 proslijeđenih tweetova 123 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  3. 1. pro 2019.

    Slides and video of my talk at the Neural Architects workshop at ICCV this year!

    17 proslijeđenih tweetova 54 korisnika označavaju da im se sviđa
    Poništi
  4. 22. stu 2019.

    This is a great description of RandAugment! Thanks so much.

    This video explains the new RandAugment AutoML Data Augmentation algorithm from , improving on previous techniques (AutoAugment/PBA) on ImageNet and dramatically reducing the search space, making AutoML for Data Aug much easier!
    1 reply 5 proslijeđenih tweetova 23 korisnika označavaju da im se sviđa
    Poništi
  5. 18. stu 2019.

    *New paper* RandAugment: a new data augmentation. Better & simpler than AutoAugment. Main idea is to select transformations at random, and tune their magnitude. It achieves 85.0% top-1 on ImageNet. Paper: Code:

    5 replies 169 proslijeđenih tweetova 622 korisnika označavaju da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    29. lip 2019.

    Automatically learned data augmentation policies can train more accurate models using fewer labeled examples, letting you stretch the amount you get from each labeled example. Work by 's , , Golnaz Ghiasi, Tsung-Yi Lin, Jonathon Shlens, &

    Data augmentation is even more crucial for detection. We present AutoAugment for object detection, achieving SOTA on COCO validation set (50.7 mAP). Policy transfers to different models & datasets. Paper: , Code: , details in thread.
    Prikaži ovu nit
    5 replies 103 proslijeđena tweeta 467 korisnika označava da im se sviđa
    Poništi
  7. proslijedio/la je Tweet
    26. lip 2019.

    Data augmentation is even more crucial for detection. We present AutoAugment for object detection, achieving SOTA on COCO validation set (50.7 mAP). Policy transfers to different models & datasets. Paper: , Code: , details in thread.

    5 replies 114 proslijeđenih tweetova 382 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    18. lip 2019.

    Nice article in Daily about AutoAugment. Thanks Ralph Anzarouth for the interview! Come see our talk [1-1A] and poster [1-1P-12] if you want to learn more.

    4 proslijeđena tweeta 16 korisnika označava da im se sviđa
    Poništi
  9. proslijedio/la je Tweet
    22. tra 2019.

    Exciting new work on replacing convolutions with self-attention for vision. Our paper shows that full attention is good, but loses a few percents in accuracy. And a middle ground that combines convolutions and self-attention is better. Link:

    5 replies 111 proslijeđenih tweetova 476 korisnika označava da im se sviđa
    Poništi
  10. proslijedio/la je Tweet
    22. tra 2019.

    Automatic Speech Recognition (ASR) struggles in the absence of an extensive volume of training data. We present SpecAugment, a new approach to augmenting audio data that treats it as a visual problem rather than an audio one. Learn more at →

    13 replies 341 proslijeđeni tweet 871 korisnik označava da mu se sviđa
    Poništi
  11. proslijedio/la je Tweet
    22. tra 2019.

    Wanted to apply AutoAugment to speech, but a handcrafted augmentation policy already improves SOTA. Idea: randomly drop out certain time & frequency blocks, and warp input spectrogram. Results: state-of-art on LibriSpeech 960h & Switchboard 300h. Link:

    Automatic Speech Recognition (ASR) struggles in the absence of an extensive volume of training data. We present SpecAugment, a new approach to augmenting audio data that treats it as a visual problem rather than an audio one. Learn more at →
    41 proslijeđeni tweet 166 korisnika označava da im se sviđa
    Poništi
  12. proslijedio/la je Tweet
    16. tra 2019.

    Latest version of the AutoAugment paper is up: . Stop by our oral presentation at CVPR to learn more! Joint work with Vijay Vasudevan and .

    9 proslijeđenih tweetova 47 korisnika označava da im se sviđa
    Poništi
  13. proslijedio/la je Tweet
    2. stu 2017.

    AutoML for large scale image classification and object detection

    275 proslijeđenih tweetova 560 korisnika označava da im se sviđa
    Poništi

@barret_zoph 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:

    ·