Martin WattenbergOvjeren akaunt

@wattenberg

Human/AI interaction. Visualization as design, science, art. Co-lead Google's PAIR (People+AI Research) initiative in . Tweets are my own.

Vrijeme pridruživanja: srpanj 2008.

Tweetovi

Blokirali ste korisnika/cu @wattenberg

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

  1. And what happens when you apply this idea to other artworks across history? Read more here: And try it yourself with this interactive visualization:

    Prikaži ovu nit
    Poništi
  2. What does a neural net see when it looks at an Escher print? Explore illusions in an automatic depth perception system: From , and

    Prikaži ovu nit
    Poništi
  3. 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
  4. proslijedio/la je Tweet

    On Nov. 14th we'll be live at the to discuss transparency in AI & . Want to get involved? 💬 Join the conversation by tweeting questions to Google speakers 💻 Watch the livestream->

    Poništi
  5. proslijedio/la je Tweet
    7. stu 2019.

    Really excited to share that we launched a new version of today! It's a fast easy way to train machine learning models on-device, in your browser. You can now save models for your projects, use sound/pose, and more. Try it out →

    Prikaži ovu nit
    Poništi
  6. UMAP is a real advance in visualizing high dimensional data. But how should we read it? And when should we use it? Take a look at this interactive essay by and !

    Poništi
  7. proslijedio/la je Tweet
    1. stu 2019.

    Curious about AI and machine learning? Share what you want to know. What do you most want to learn about machine learning?

    Poništi
  8. “Time was up for Molly and Charlie. They had one chance, and they had to make it count.” What happens next in the story? Now what if I started, "Time was up for Deborah and Steven” instead? If you ask a neural network (GPT-2), the different answers tell a story of their own.

    Poništi
  9. proslijedio/la je Tweet
    15. kol 2019.

    How are semantics and word senses represented in contextual word embeddings? Visualizations, blogpost, and paper by , , , , and .

    Poništi
  10. Prikaži ovu nit
    Poništi
  11. The blog at explores the shape of these representations, with math and visualization. A mysterious “squared distance” effect they found turns out to be surprisingly natural!

    Prikaži ovu nit
    Poništi
  12. Some context: Recently and discovered certain neural networks (e.g., BERT) build internal geometric representations of syntax trees. It’s beautiful: the full tree is actually embedded in a high-dimensional vector space.

    Prikaži ovu nit
    Poništi
  13. How does a neural net represent language? See the visualizations and geometry in this PAIR team paper and blog post

    Prikaži ovu nit
    Poništi
  14. One of many fascinating insights from the new baby name data that came out today. Read more at !

    Poništi
  15. Designing with AI requires a big change in mindset—in my opinion, just as big as when graphical user interfaces were introduced. I'm delighted to announce the People + AI Guidebook, which lays out recommendations for designing human-first AI products.

    Poništi
  16. What does it really mean to design for human/AI collaboration? How can doctors and ML work together effectively? If you're at CHI, go to this talk for some answers! (And if you're not, the paper's online.)

    Poništi
  17. proslijedio/la je Tweet
    17. tra 2019.

    3rd and last Award Winner 2019: TCAV by , Senior scientist at Brain: ‘I created a tool for everyone to understand how work’. is Interpretable for and

    Prikaži ovu nit
    Poništi
  18. I would recommend this essay to anyone starting to work with data, not just journalists! Data is rarely what it seems at first.

    Poništi
  19. Bach to future! Compose your own music, with the help of the AI on today's home page. Great collaboration with the Magenta, Doodle and TensorFlow.js teams. Congrats to all involved!

    Poništi
  20. proslijedio/la je Tweet
    12. ožu 2019.

    The Periodic Table of Elements turns 150 years old this month and the history of its creation is astonishing. In celebration, here's a visualization our team made a few years ago:

    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:

    ·