Mahima Pushkarna

@mahimapushkarna

🕴Human/AI Design🤖 , 🔭Converting Research to Tools 📐, Design Lead, , Big Picture Data Visualization, PAIR | Tweets own

Gradient Descent Valley
Vrijeme pridruživanja: rujan 2008.

Tweetovi

Blokirali ste korisnika/cu @mahimapushkarna

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

  1. Prikvačeni tweet
    27. stu 2019.

    🙋🏿🙋🏼‍♂️🙋🏽‍♀️🙋🏿‍♂️The What-If Tool team wants to hear from you! Fill out this form to participate in an upcoming study to help us improve the ! 👇🏽👇🏿👇

    Poništi
  2. 26. sij

    If you can ask a question that you hadn't already thought to ask of your model when using the , we consider that a meaningful step in the right direction! My colleagues and unpack WIT x SHAPLEY values at .

    Poništi
  3. 26. sij

    "Historically, minimalism tells us to focus on what doesn’t at first seem pleasant or beautiful and turn it into art instead of creating a worldview based only on what we already like."

    Poništi
  4. proslijedio/la je Tweet
    24. sij

    You can now search for datasets on Google: And, because one must test data tools with topics of importance, I looked for "chocolate+UK." Who knew that the top UK chocolate bars for all ages are Mars and Galaxy and Snickers? Congrats, Natasha Noy & team!

    Poništi
  5. proslijedio/la je Tweet
    14. sij

    spent last summer with and PAIR exploring using DeepDream-style techniques on the BERT language architecture, as a model understanding probe. Check out the blog post for insights and code pointers:

    Poništi
  6. 20. pro 2019.
    Poništi
  7. proslijedio/la je Tweet
    12. pro 2019.

    Poster today at 10:45am! #83 "Toward automatic conference base explanations"- work with my awesome intern , and two awesome jamess - James Zou. And James Wexler

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    22. stu 2019.

    How does a neural net see depth in paintings? Does it "resolve" the depth depicted in these images in the same way we do? Read about depth predictions: PAIR project by , ,

    Poništi
  9. proslijedio/la je Tweet
    21. stu 2019.

    What-If Tool 1.5 is now released, with full support for models that return feature-wise attributions! This accompanies the Explainable AI release. Let's dive in...

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    14. stu 2019.

    just kicked off with a welcome note from ! Join us in this discussion about AI and via livestream 💻 ➡️ – mjesto: Google UK

    Poništi
  11. proslijedio/la je Tweet
    11. stu 2019.

    Want to learn how your ML models work under the hood? and I have a new mini-series for you on the What-If Tool!

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

    Very pleased to release Alibi with and - an open source library for model inspection and interpretation. Currently includes Anchors (), contrastive explanations () and trust scores ():

    Poništi
  13. proslijedio/la je Tweet
    5. stu 2019.
    Poništi
  14. 22. lis 2019.

    If you're at today, come check out talk about what makes the What-if Tool from at Ballroom A in the session track "A tour of VAST"! Also, we have stickers!

    Poništi
  15. proslijedio/la je Tweet
    10. lis 2019.

    The war between ML frameworks has raged on since the rebirth of deep learning. Who is winning? 's data analysis shows clear trends: PyTorch is winning dramatically among researchers, while Tensorflow still dominates industry.

    Poništi
  16. proslijedio/la je Tweet
    2. lis 2019.

    is hiring a UX designer. This is a great opportunity to help improve prosecutor transparency & work with great folks like ! (, , can you help get the word out?)

    Poništi
  17. 30. ruj 2019.

    Is it weird to like all the posts in the thread? 👇👇👇 Well it can't be helped, is now loaded with even neater capabilities :)

    Poništi
  18. 27. ruj 2019.

    +1 to this ENTIRE thread. Organizers & hosts - Pick up the phone and rehearse introductions if you must. Speakers, esp POCs - point audience to your bios in your slide annotations, and scope background info to your presentations.

    Poništi
  19. 27. ruj 2019.

    If fairness IRL is that elusive piece of the puzzle, ML Fairness is the one that just doesn't seem to fit! 🧩😕 demystifies it a bit in her blog post!

    Poništi
  20. proslijedio/la je Tweet
    24. 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:

    ·