Greg Koytiger

@GregKoytiger

VP, Head of AI Products @

Boston, MA
Vrijeme pridruživanja: listopad 2015.

Tweetovi

Blokirali ste korisnika/cu @GregKoytiger

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

  1. proslijedio/la je Tweet
    23. sij

    Here is the promised blogpost: on the motivation behind our modeling / ML approach for protein-peptide interactions. We'll likely have another post soon focused more on the (structural) biology.

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    15. sij

    Glad to see ’s AlphaFold paper finally out. I had the pleasure of being one of the reviewers and getting to write the accompanying article. The future of protein structure prediction is looking very bright!

    Poništi
  3. proslijedio/la je Tweet
    10. sij

    Researchers developed a bespoke machine-learning approach, hierarchical statistical mechanical modelling, for the accurate prediction of protein–peptide interactions across multiple protein families.

    Poništi
  4. 8. sij

    For me, the most important take away is Figure 5 - that current machine learning approaches are of higher fidelity than high throughput experiments! For another example, see Figure 2b from the Nature Methods paper yesterday

    Prikaži ovu nit
    Poništi
  5. 8. sij

    Dream Kinase prediction challenge paper is now on Biorxiv! (Unofficial) top performing model by yours truly

    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet

    Our field isn't quite "artificial intelligence" -- it's "cognitive automation": the encoding and operationalization of human-generated abstractions / behaviors / skills. The "intelligence" label is a category error

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    7. sij

    I’m late to my own party but excited to share our new work on predicting SLiM-mediated protein-protein interactions, out today in with Joe Cunningham, , and ! A blogpost is forthcoming but for now a tweetstorm (1/8)

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    7. sij

    Joseph Cunningham Peter Sorger and use energy-based ML to predict protein-peptide interactions. Their model is interpretable, naturally incorporates physical "priors", and outperforms high-throughput experiments!

    Poništi
  9. proslijedio/la je Tweet
    6. sij

    The path of a maturing thinker. In order to get to Grown-Up Mountain and start real learning, you have to brave the cold winds of Insecure Canyon. If you're not willing to say "I don't know" for a while, you might spend your whole life on Child’s Hill.

    Poništi
  10. proslijedio/la je Tweet
    31. pro 2019.

    A piece of holiday-time reflection: one thing I’m grateful about in science is the existence of a real field-wide community, made more visible by Twitter. I suspect this is less true in other professions and is a genuinely positive feature.

    Poništi
  11. proslijedio/la je Tweet
    24. pro 2019.

    Happy holidays from the Cascade team (minus a few)! It's been an amazing year with these incredible people, here's to an even better 2020!

    Poništi
  12. proslijedio/la je Tweet

    Reminder: student evaluations of teaching tend to show stark racial and gender biases, and don't actually measure teaching effectiveness. A statement endorsed by 22 scholarly associations calls for limiting the role of student ratings in faculty review:

    Poništi
  13. proslijedio/la je Tweet
    27. stu 2019.
    Poništi
  14. proslijedio/la je Tweet

    Cool paper on combining meta- and active learning to efficiently learn protein function from protein sequence, from Rainier Barrett and

    Poništi
  15. proslijedio/la je Tweet
    19. stu 2019.

    Great work using inter-residue orientations to exceed AlphaFold’s performance on protein structure prediction by Jianyi Yang, Ivan Anishchenko, and others from the Baker lab: . First heard about this at RosettaCon and I’m very glad to finally see it out!

    Poništi
  16. proslijedio/la je Tweet

    This is how octopuses use camouflage in the wild

    Poništi
  17. proslijedio/la je Tweet
    4. stu 2019.
    Prikaži ovu nit
    Poništi
  18. proslijedio/la je Tweet
    31. lis 2019.

    Your annual reminder of arguably the greatest thread of science Twitter poetry of all time. To all scientists this Halloween eve: may the sparrow of doubt hover close enough to guide you to rigor, but not so close that you are paralyzed into inaction. 🎃

    Poništi
  19. proslijedio/la je Tweet
    29. lis 2019.

    Engineering orthogonal signalling pathways reveals the sparse occupancy of sequence space. Congrats on the great paper

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

    Scouring through openreview for ICLR submissions, noticed that there are a lot of interesting GNN and chemistry submissions. Here is a short list of some that caught my attention for now (and that I managed to pick the main idea).

    Prikaži ovu nit
    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:

    ·