Mihaela Rosca

@MihaelaCRosca

Research Engineer at DeepMind. Machine learning. views are my own.

London
Vrijeme pridruživanja: listopad 2011.

Tweetovi

Blokirali ste korisnika/cu @MihaelaCRosca

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

  1. 22. stu 2019.
    Poništi
  2. 22. stu 2019.

    The code reproducing the experiments is this paper is now available at:

    Poništi
  3. 22. stu 2019.

    Code reproducing our review paper on stochastic gradient estimation:

    Prikaži ovu nit
    Poništi
  4. 22. stu 2019.

    Code reproducing our ICML paper on improving GAN training with latent optimization:

    Prikaži ovu nit
    Poništi
  5. 22. stu 2019.

    Code reproducing our Neurips paper on training language gans from scratch: .

    Prikaži ovu nit
    Poništi
  6. 22. stu 2019.

    The code for three of our papers is now out! Interested in learning language GANs from scratch? Stochastic gradient estimation? Improving GANs through latent optimization? Follow this thread!

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet

    I have two open phd positions in my group at / Both in deep learning, one in computer vision to combine multi-sensors for scene reconstruction, and the other for weather forecast and air traffic control.

    Poništi
  8. 6. srp 2019.

    At unconference session. Such a great idea! The topics chosen were diverse and the discussions enlightening for all of us!

    Poništi
  9. 5. srp 2019.

    A must for anyone interested in training VAEs.

    Poništi
  10. 1. srp 2019.

    has started! Looking forward to an amazing week!

    Poništi
  11. 26. lip 2019.

    I really enjoyed working on this review paper with my amazing co workers! So much to learn!

    Poništi
  12. proslijedio/la je Tweet
    23. svi 2019.

    The issue isn't deep fakes or autogenerated text. The issue is that not a single bit of our media is digitally signed. We sign software to ensure updates to our machines haven't been tampered with unexpectedly but don't have any similar process for updating fellow humans.

    Prikaži ovu nit
    Poništi
  13. 17. svi 2019.

    Excited about this collaboration with my colleagues Yan Wu and Timothy Lilicrap. Compressed sensing, GANs and meta learning. What's not to like?

    Poništi
  14. 25. tra 2019.

    EU citizens living in the UK can choose to vote in the UK European parliament elections instead of their EU country of citizenship. If interested, register to vote by 7th of May. Gov info at

    Poništi
  15. proslijedio/la je Tweet

    We ( ,, , & myself) are organizing a workshop on "Uncertainty & Robustness in Deep Learning" at . See for more info. Please submit your work (deadline: April 30, 2019) and attend the workshop! :)

    Poništi
  16. 4. ožu 2019.

    Cannot recommend this enough. Apply before the 29th of March!

    Poništi
  17. 16. velj 2019.

    Variational inference is a gift that keeps on giving. Been looking at it for two years, and it can still amaze me at times. Kids, know your ELBO.

    Poništi
  18. proslijedio/la je Tweet
    30. sij 2019.

    The application period for EEML 2019 is now open! Detailed instructions for applying are available online . If you want to meet our amazing speakers and network with like-minded people, please apply before March 29th!

    Poništi
  19. 28. sij 2019.

    Excited to be at and looking forward to my talk tomorrow! In the meantime, I am enjoying the great talks by an excellent set of speakers

    Poništi
  20. proslijedio/la je Tweet
    26. sij 2019.

    Wow wow wow! now supports the fastai library directly :D That means you can run all the lessons at , and do your own research and development with fastai, all for free. That's... that's just amazing. To get started:

    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:

    ·