HumanForest

@HumanForests

is testing our (pat.pending) method under an NSF grant to harness human brain power and beat the predictive accuracy of machine models

Brooklyn, NYC
Vrijeme pridruživanja: kolovoz 2019.

Tweetovi

Blokirali ste korisnika/cu @HumanForests

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

  1. 14. sij

    Great conversations at about our NSF-funded work on Human Forest, and its applications in biotech/pharma.

    Poništi
  2. proslijedio/la je Tweet
    26. stu 2019.

    Albania is one of europes poorest countries and has been hit by two catastrophic earthquakes in the past two months. hundreds and hundreds of people are injured. dozens have died. homes have been turned to dust. please please get the media to talk about this and

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    25. ruj 2019.

    Some psychological effects are depressingly robustly replicable.

    Poništi
  4. proslijedio/la je Tweet
    25. ruj 2019.

    ( & ) we are trying 2 test that question 2gether w/ at w/National Science Foundation grant. Our vs JHU's random forest model in predicting clinical trial outcomes 4 the nxt 2.5 years

    Poništi
  5. proslijedio/la je Tweet
    23. ruj 2019.
    Poništi
  6. proslijedio/la je Tweet
    9. ruj 2019.
    Poništi
  7. proslijedio/la je Tweet
    9. ruj 2019.
    Poništi
  8. proslijedio/la je Tweet
    11. ruj 2019.
    Poništi
  9. proslijedio/la je Tweet
    7. ruj 2019.

    Uh-oh? Definitely worth a look, esp. his response to my Q that combining market predictions with his predictions led to further 10% gain in AUC -- so independent signal: we should assimilate him into the market. ;-)

    Poništi
  10. proslijedio/la je Tweet
    10. ruj 2019.

    I Work for N.S.A. We Cannot Afford to Lose the Digital Revolution.

    Poništi
  11. proslijedio/la je Tweet
    23. kol 2019.

    What is a portfolio? Why does it matter for ? explain in great new article. Imagine if whenever a trial was planned, reviewed, published, we could see its portfolio like this?

    Poništi
  12. 9. ruj 2019.

    take note: will use human collective brain power 2 predict success of new cancer drugs as part of our new research. Can it beat random forests? We think it might!

    Poništi
  13. proslijedio/la je Tweet

    Happy to participate in the next round of clinical trials ✋

    Poništi
  14. proslijedio/la je Tweet
    6. ruj 2019.
    Poništi
  15. proslijedio/la je Tweet
    6. ruj 2019.

    It's pretty shocking that Scientific American would publish such a weakly researched and poorly written article. Assertions can and should be rebutted with the considerable existing evidence against. Algorithmic Foreign Policy via

    Poništi
  16. proslijedio/la je Tweet
    3. ruj 2019.

    All else equal, I suspect this is true. But all else is never remotely equal in these research designs. What matters is what you read—and how you read it.

    Poništi
  17. proslijedio/la je Tweet
    6. ruj 2019.

    I like this juxtaposition. “Trust-my-judgment” is, less & less, something Experts can demand by virtue of status alone. “Show-me-your-track-record” is a request that even High-Priesthood Experts will find it ever harder to dismiss.

    Poništi
  18. 6. ruj 2019.

    Trust in algorithms always dependent on the circumstance. We focus on building better AI (development of which is concentrated in extractive wealth conditions), while we beggar our humanity by relegating it 2 automation & synthetic dominion. At humans r atop HCI hybrids

    Poništi
  19. 6. ruj 2019.

    We at ( and ) agree and with HumanForest (pat. pending) we aim to make it clear why humans matter.

    Poništi
  20. 6. ruj 2019.
    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:

    ·