Ian Williamson

@ianadwilliamson

A budding photonicist and optimizician / Currently a postdoc at / PhD from

San Francisco Bay Area, CA
ianwilliamson.me
Vrijeme pridruživanja: listopad 2011.
5 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @ianadwilliamson

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

  1. Prikvačeni tweet
    20. pro 2019.

    Waves are recurrent neural nets! 🌊 Our work was published in Science Advances today showing that analog machine learning computations can be performed by propagating light and sound through programmed materials! Code on GitHub:

    8 replies 51 proslijeđeni tweet 176 korisnika označava da im se sviđa
    Poništi
  2. proslijedio/la je Tweet
    20. sij

    This is a terrible take. It totally misses the key point: handles >10k submissions and >25m downloads per month at a cost of only $1.3m /year. spends $193m (!) on publications, and spends $33.7m. Something stinks.

    10 replies 162 proslijeđena tweeta 483 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  3. 20. sij

    Developed a vectorized transfer matrix method to compute the optical reflection / transmission of multi-layer stacks for another project. Uses Tensor Flow to vectorize over omega/k and is quite fast. Releasing into the wild in case anyone finds it useful.

    3 proslijeđena tweeta 13 korisnika označava da im se sviđa
    Poništi
  4. 20. sij

    Liking xarray a lot for handling history / results / checkpoints in optimizations. So much cleaner than stuffing various tensors into pandas dataframes 😬

    1 reply 1 proslijeđeni tweet 12 korisnika označava da im se sviđa
    Poništi
  5. proslijedio/la je Tweet
    10. sij

    as an

    29 proslijeđenih tweetova 64 korisnika označavaju da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    5. sij

    A hell game which poses sickening question: What aged Liverpool youth team would Everton have to play so they could finally win at Anfield? Under 9s? Under 8s?

    59 replies 319 proslijeđenih tweetova 1.621 korisnik označava da mu se sviđa
    Poništi
  7. proslijedio/la je Tweet
    4. sij

    In 2020, I predict we sill see the first ***ConvPuter***—a fully-functioning computer system where every function in every application, from the OS kernel to user-facing apps, is learned w. deep convolutional neural networks.

    5 replies 24 proslijeđena tweeta 274 korisnika označavaju da im se sviđa
    Poništi
  8. proslijedio/la je Tweet
    2. sij

    is what drove the development of libraries like , but we are beginning to realize these tools can find applications way beyond ML! Here's a cool example from Research (even if I don't understand the details...)

    1 reply 3 proslijeđena tweeta 5 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    23. pro 2019.

    Visualizing MNIST classification. Neat.

    19 replies 1.797 proslijeđenih tweetova 5.729 korisnika označava da im se sviđa
    Poništi
  10. proslijedio/la je Tweet
    21. pro 2019.

    Our recent work on synthetic dimensions is out on ! Higher-dimensional physics, which previously required complex structures, can now be studied in a very simple photonic structure. with Qian, Luqi, , Meng & Shanhui Fan

    5 replies 14 proslijeđenih tweetova 90 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    14. pro 2019.

    The lesson of the decades long failure of the wars of 9/11 must still be carved into collective consciousness. U.S. officials misled the public about the war in Afghanistan, confidential documents reveal - Washington Post

    1 reply 1 proslijeđeni tweet
    Poništi
  12. 13. pro 2019.

    Ever look at a nice figure in a paper and forget about the data / science for a minute as you take in the subtle pastel color gradients, clean lines, and consistent fonts while a soothing calmness washes over you?

    5 replies 2 proslijeđena tweeta 29 korisnika označava da im se sviđa
    Poništi
  13. proslijedio/la je Tweet
    12. pro 2019.

    Before we create a Space Force we should make sure everyone has health care.

    2.758 replies 20.290 proslijeđenih tweetova 150.229 korisnika označava da im se sviđa
    Poništi
  14. 2. pro 2019.

    writing this review paper got me like

    1 reply 7 korisnika označava da im se sviđa
    Poništi
  15. 22. stu 2019.

    Interested in photonic inverse design and automatic differentiation? ➡️ Example notebooks and slides from our workshop are on GitHub! The simulator package uses Python/HIPS autograd and could also be of interest to machine learning people who do physics!

    11 proslijeđenih tweetova 34 korisnika označavaju da im se sviđa
    Poništi
  16. 12. stu 2019.

    when the gradients agree

    9 korisnika označava da im se sviđa
    Poništi
  17. proslijedio/la je Tweet
    7. stu 2019.

    i would like more frens. so i have made a list of my best qualities. for your consideration 6. am fluffy 1. can jump (and sit) C. sometimes i sneeze i look forward. to hearing from you

    4.331 reply 22.902 proslijeđena tweeta 224.279 korisnika označava da im se sviđa
    Poništi
  18. 6. stu 2019.

    Currently learning about all of the fancy new advances in web design that have taken place since when I was last serious about making websites (~middle school)

    4 korisnika označavaju da im se sviđa
    Poništi
  19. 20. ruj 2019.

    Accidentally typed 'githug' into the address bar instead of 'github'. I don't know what a 'githug' is, but I think I need one

    6 korisnika označava da im se sviđa
    Poništi
  20. proslijedio/la je Tweet
    4. ruj 2019.

    check out our electromagnetic simulation package 'ceviche', which implements Maxwell's Equations with automatic differentiation for effortless optimization and integration with machine learning models.

    3 proslijeđena tweeta 4 korisnika označavaju da im se sviđa
    Poništi

@ianadwilliamson 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:

    ·