Kevin Yang 楊凱筌

@KevinKaichuang

Machine learning scientist at Generate Biosciences. He/him.

Massachusetts, USA
Vrijeme pridruživanja: studeni 2016.

Tweetovi

Blokirali ste korisnika/cu @KevinKaichuang

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

  1. Prikvačeni tweet

    We (, ) wrote a review about machine learning for protein engineering, but there's so many new papers that we can't update it quickly enough. So Zach and I made a public list of papers covering machine learning on proteins.

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    4. velj

    Would like to write about my family text thread which is entirely in Chinese and how I have to copy & paste to translate into English to understand what they’re all laughing/哈哈哈 about and how I’m always lagging behind

    Poništi
  3. A nice review on using ML to find GPCR ligands, by and Benjamin Kaufman

    Poništi
  4. proslijedio/la je Tweet

    Congratulations for winning the 2020 Prize for Creative Promise in Biomedical Science

    Poništi
  5. proslijedio/la je Tweet
    2. velj

    does coronavirus come from bananas if they both have a methionine at the start here

    Prikaži ovu nit
    Poništi
  6. Why didn't I think to BLAST random parts of 2019-nCoV and write papers about the hits?

    Poništi
  7. 3. velj

    You use gradient descent to learn parameters, so why not use it to learn hyperparameters? And hyper-hyperparameters?! by Kartic Chandra, , Samantha Andow, Emilio Arroyo-Fang, , ...

    Poništi
  8. proslijedio/la je Tweet
    2. velj

    Here are some thoughts on stats “vs” ML. Rel btwn the fields, why debates on topic are almost always fundamentally flawed, why I think there’s so much talking past each other, and why I wish we could focus instead on *specific* methods + problems 2/3

    Prikaži ovu nit
    Poništi
  9. 2. velj

    The only wet-lab experiments I do nowadays.

    餃子
    Poništi
  10. proslijedio/la je Tweet
    1. velj

    Me: Blah blah probability distribution blah blah blah. Prospective visiting graduate student: Sometimes it helps to think out of the simplex. Me:

    Poništi
  11. proslijedio/la je Tweet
    31. sij

    alright all you terrified dorks let me show you how to search for matching protein sequences because there is a CERTAIN GARBAGE PAPER OUT THERE

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet

    Westerners think Asia is exotic, but I just spent a week in Uppsala, Sweden, in university culture, and THIS place is exotic. Tuxedoes, ball gowns, Latin, opera, student choirs, artillery, laurel crowns for doctorates, and diplomas with gold rings. Someone make a movie of this.

    Prikaži ovu nit
    Poništi
  13. 31. sij

    I wonder what the correlation is between being able to see things in your head and having an internal monologue.

    Poništi
  14. 31. sij

    I don't know about you, but I carry bat jerky around all the time.

    Poništi
  15. 30. sij

    Everything in latex is fun except for making tables.

    Poništi
  16. 29. sij

    My coworkers and I never baby talk to our language models. We use full sentences and a wide vocabulary including complex words. My transformer trained for 3 epochs can carry a full conversation. I stopped my LSTM after her second epoch because nobody cares about LSTMs anymore.

    Poništi
  17. 29. sij

    My wife and I never baby talk our kid. We use full sentences and a wide vocabulary including complex words in 3 languages. My son isn't even 2 and he's already learned to ignore us in English, Japanese, and Mandarin.

    Poništi
  18. 29. sij

    So I use a blanket my grandma brought from Taiwan when I was a kid. It's not really big enough for two adults: that's what the other 3 blankets on the bed are for.

    Poništi
  19. proslijedio/la je Tweet
    27. sij

    tw americans it has come to my attention . that u guys don’t wash ur rice PLEASE WASH UR RICE there have been cases where crazy sh!t like mice feces and like . bugs have been found in ur packet of rice not to mention dust and other weird stuff in there PLEASE WASH UR RICE

    Prikaži ovu nit
    Poništi
  20. 28. sij

    Common benchmarks for image generation can be gamed by memorizing the training set. , , demonstrate a neural network divergence metric that measures both sample quality and coverage of the true distribution.

    Poništi
  21. 28. sij

    I... Don't know? Whatever the ICML template specifies or my current markdown editor defaults to I guess.

    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:

    ·