Andrzej Wąsowski

@AndrzejWasowski

A context-free professor conjugates a well-typed glass of higher-order prilimubary students.

Vrijeme pridruživanja: lipanj 2009.

Tweetovi

Blokirali ste korisnika/cu @AndrzejWasowski

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

  1. Poništi
  2. 3. velj

    Dear Hive, I seek trustworthy sources on the economical and cost of , especially (if such exist) the environmental cost of in robotics. If you know a paper, or a reasonable news article, point me to them. Please retweet. Thanks!

    Poništi
  3. proslijedio/la je Tweet
    3. velj
    Odgovor korisniku/ci

    It's very interesting, here there is an explanation about this,

    Poništi
  4. proslijedio/la je Tweet
    31. sij

    Some absolutely amazing ROS 2 tutorials. Check them out.

    Poništi
  5. proslijedio/la je Tweet
    2. velj
    Odgovor korisnicima

    Below you find the example from the slides. (sorry for German text) The first part is for 32bit x86 FP format: that 0.1 is not representable (you basically get the same with 64bit or IEEE FP) The problem is that they are all base 2 and for 0.1 you need base 10.

    Poništi
  6. proslijedio/la je Tweet
    2. velj
    Odgovor korisniku/ci

    Actually none of the numbers is represantable in standard floating point format. 0.1 is off. The trick is that already the 0.1 is rounded on output. Then you get error further errors by additions (just taught this our first graders last week 😅)

    Poništi
  7. proslijedio/la je Tweet
    2. velj
    Odgovor korisniku/ci

    rascal>0.1 + 0.1 + 0.1 real: 0.3 rascal>0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 real: 0.6 rascal>0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 real: 0.9 😎

    Poništi
  8. 2. velj

    0.2 + 0.1 works as well. The interesting thing is that all the summed numbers are re-presentable, and so is the result. Still the addition produces something different.

    Prikaži ovu nit
    Poništi
  9. 1. velj

    If you think that numerical imprecision affects calculations rarely, when you iterate many times, on very small numbers, or on very large numbers, try "0.1 + 0.1 + 0.1" in your favorite[*] programming language. [*] Some might be better than others, if you really try. Most won't

    Prikaži ovu nit
    Poništi
  10. 1. velj

    Thinking of it, most of my research is done in free-time (after teaching, admin, and nuisance) . I edit papers myself. Reviewers work for free (wait that's me...). The $1750 cost of hosting a PDF file on a web-server exceeds the cost of performing the research reported in it.

    Prikaži ovu nit
    Poništi
  11. 1. velj

    A mere US$1400 + ca. 200$ yearly membership (or $1750 in one shot if your prefer) is a real bargain to get your research published in a Plan S compliant OA journal within 10 weeks of submission. Another hallmark of quality from :P

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    31. sij

    Interested in software engineering for AI/ML-enabled systems? After teaching a course on the topic last semester, I finally managed to compile a reading list in the form of an annotated bibliography: (PRs welcome)

    Poništi
  13. proslijedio/la je Tweet
    27. sij

    We are seeking a post-doc affiliated with the BARC center for basic algorithms research(). Solid track record in algorithms theory? Join our ambitious, creative, collaborative, and fun environment!

    Poništi
  14. 28. sij

    Come to . We will have a chat about your dark side.

    Poništi
  15. proslijedio/la je Tweet
    27. sij
    , , i još njih 7
    Poništi
  16. 27. sij

    TIL: A proof by magic. From in the . Also in the words of John Kruschke himself (is he on Twitter?): 'This magic trick is great entertainment at parties.'

    Poništi
  17. proslijedio/la je Tweet
    27. sij

    Come join my lab as a PhD. Less than a week left to apply! 🏃‍♀️⌚️

    Poništi
  18. 26. sij
    Poništi
  19. 24. sij

    Oh. It turns out there are even papers on this (a negative result!)

    Prikaži ovu nit
    Poništi
  20. 24. sij

    A few months of reading various probability book and the main thing I learnt is the fallacy that it is easy to make biased coins.

    Prikaži ovu nit
    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:

    ·