Kevin Mader

@KevinSMader

Software engineer, lecturer, and occasional Twitter troll

Zurich
Vrijeme pridruživanja: veljača 2018.

Tweetovi

Blokirali ste korisnika/cu @KevinSMader

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

  1. Prikvačeni tweet
    2. ožu 2018.
    Odgovor korisnicima

    DL papers that plot their loss curves in Excel make me nervous. From the results there seems to be only marginal improvements with CapsuleNet potentially more due to network initialization or ordering. I would wait for a significant improvement on a well established benchmark.

    Poništi
  2. proslijedio/la je Tweet

    Much of what’s being sold as "AI" today is snake oil. It does not and cannot work. In a talk at MIT yesterday, I described why this happening, how we can recognize flawed AI claims, and push back. Here are my annotated slides:

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    6. stu 2019.

    New essay: “Is your startup idea taken?" -- and why we love X for Y startups

    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    26. lis 2019.

    I saw a guy debugging his model today. No idpb. No unittest. No visualization. No disabling regularization. He just sat there staring at every line of code and cursing TensorFlow. Like every machine learning researcher I know.

    Poništi
  5. proslijedio/la je Tweet
    22. lis 2019.

    I’ve read almost every book on startups, and somehow have just completely missed the chapters that covered this part of the strategy.

    Poništi
  6. 9. lis 2019.

    I haven't seen startup tasks prioritized in that order, but I can't say I disagree

    Poništi
  7. 13. kol 2019.

    "We encourage and expect quality and diversity in speakers. Tracks with all male speakers will be sent back to the 19th century." Great job !

    Poništi
  8. 2. kol 2019.

    Seterra: The U.S.: 50 States; My Score: 93% 4:06

    Poništi
  9. 2. kol 2019.

    Silly ob- and nidwalden

    Prikaži ovu nit
    Poništi
  10. 2. kol 2019.

    Seterra: Switzerland: Cantons; My Score: 93% 3:15

    Prikaži ovu nit
    Poništi
  11. 3. ožu 2019.

    "Pull requests that have a failing status can’t be merged on a phone", bravo for preventing terrible accidentally clicked buttons

    Poništi
  12. proslijedio/la je Tweet
    Odgovor korisniku/ci

    When teaching yesterday, I realised one advantage of using the cloud is that you can reassure students there is absolutely nothing they can do to break their computer

    Poništi
  13. proslijedio/la je Tweet
    8. lis 2018.

    Many people believe that 21st century neural networks will soon transform doctors' approaches to medical reasoning; the 18th century Bayes' rule has yet to.

    Poništi
  14. proslijedio/la je Tweet
    10. lis 2018.
    Poništi
  15. proslijedio/la je Tweet

    Great medical example of multivolume rendering using ipyvolume by , showing CT scan (grey) with PET scan overlayed (red) and label data (green). Thanks Kevin! Try it out on mybinder:

    Poništi
  16. 20. lip 2018.

    And a rarity among medical papers, they published the code and pretrained models so you can run it yourself on

    Poništi
  17. proslijedio/la je Tweet
    25. svi 2018.

    I get messages a lot that start with "Have you ever thought of..." or "Did you consider..." Trust me, I've heard your idea before. The world isn't lacking ideas, it's lacking good execution. If you believe in your idea, go out and try to make it successful.

    Poništi
  18. proslijedio/la je Tweet

    Eight datasets for machine learning just released by : millions of images with labeled objects, 400K handwritten names for OCR, medical term relationships, parking signs, disaster response etc. Download them from

    Poništi
  19. 23. tra 2018.
    Poništi
  20. proslijedio/la je Tweet
    13. tra 2018.

    '“Is curing patients a sustainable business model?” Goldman Sachs analysts ask' - sigh

    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:

    ·