Tweetovi

Blokirali ste korisnika/cu @nestedsoftware

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

  1. 8. sij

    I was really happy to get to be a part of this project! Glad to see that is also benefitting!

    Poništi
  2. proslijedio/la je Tweet
    5. sij

    Consider: millions of years ago our antecedents gave a massive sacrifice of their left hemisphere. We lost a tremendous amount of short term memory and replaced it with Broca’s, Wernicke & the phonological loop. But why? So we can—talk. Thus chimpanzees can do this—we can’t:

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    28. pro 2019.

    Let’s develop a tic-tac-toe player using a neural network and reinforcement learning. { author: }

    Poništi
  4. 27. pro 2019.

    Phil Ashby's personal account of the heady days of the dawn of online gaming with 's Wireplay service

    Poništi
  5. 14. pro 2019.

    I would add that there was a tremendous production of experimental data throughout the 20th century around quantum behaviour of particles. That data was key to the development of the standard model (for e.g. vis parity violation).

    Poništi
  6. 1. pro 2019.

    I liked the first part of Gary's book. It does a good job of demonstrating the limitations of present machine learning techniques. The second half, about what to do next, is not convincing. I suspect that only careful analysis of biological models will gradually lead to progress.

    Poništi
  7. 25. stu 2019.

    For me, the OpenWorm project illustrates the correct paradigm for making progress toward strong AI: The idea is to increase the level of detail in our understanding of a worm, C. Elegans, and its 302 neurons.

    Poništi
  8. 23. stu 2019.

    My sense is that a much more detailed understanding (real time, 3-dimensions, molecular scale) of the nervous system, starting with simpler organisms like worms and insects will be needed to advance toward strong/general AI.

    Poništi
  9. 8. stu 2019.
    Poništi
  10. proslijedio/la je Tweet
    Poništi
  11. 3. lis 2019.

    Thought would appreciate this: I think deep learning can definitely be useful, but the old saying always applies: When all you have is a hammer, everything looks like a nail.

    Poništi
  12. proslijedio/la je Tweet

    Incrementally calculate the mean, variance, and standard deviation over a sliding window of incoming values. { author: }

    Poništi
  13. 23. ruj 2019.

    As one of the guest authors, it's very exciting that there's now a print edition of ! The ebook is great. Yet somehow there's still something romantic and satisfying about holding a book as a concrete, physical object.

    Poništi
  14. 11. ruj 2019.

    I'm no expert in machine learning, but I also didn't consider myself naive about the limitations of deep learning with respect to strong or general AI. Still, I am finding 's book, Rebooting AI (), fascinating and refreshing!

    Poništi
  15. proslijedio/la je Tweet

    Using PyTorch to reach for 99.6% accuracy at MNIST image recognition. { author: }

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

    Focus on fundamentals. 🛠️ Identify core concepts & building blocks underlying any technology or paradigm. What problem is it trying to solve? What is the underlying motivation of its developers?

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    30. kol 2019.
    Poništi
  18. 20. kol 2019.

    My article PyTorch Hello World, made the roundup of top Python posts for the week!

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

    Your First Year in Code - a community book for new (and not so new) developers. I interviewed the main author and editor Isaac Lyman () about the book and the people that made this community project a reality. View on YouTube at:

    Poništi
  20. proslijedio/la je Tweet

    Machine learning involves processing large amounts of data, which can make the results confusing. Before getting into anything more complicated, let's replicate a really basic backpropagation as a sanity check.

    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:

    ·