Mirco Musolesi

@mircomusolesi

Professor at UCL and University of Bologna. Turing Fellow at the Alan Turing Institute. ML/AI, computational models of human behaviour, and social systems.

London, England
Vrijeme pridruživanja: listopad 2008.

Tweetovi

Blokirali ste korisnika/cu @mircomusolesi

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

  1. prije 3 sata

    Reinforcement Learning meets Network Science (and vice versa).

    Poništi
  2. proslijedio/la je Tweet
    3. velj

    Less than five days left to apply for this fully funded PhD studentship! See: Leverhulme Doctoral Training Programme for the Ecological Study of the Brain: Apply: Deadline: 7 Feb 2020 23:59 BST

    Poništi
  3. proslijedio/la je Tweet

    When PC World reviewed email in 1996, we covered all the key factors--such as how much it cost and how long it took to get there.

    Poništi
  4. proslijedio/la je Tweet
    31. sij

    Here’s something you don’t see every day on campus. A couple of Indian Elephants during the filming of 'Doctor at Large' in 1957.

    Poništi
  5. 2. velj

    An application of this method to causality analysis applied to mobile interaction can be found in this paper using mobile sensing data (\w ): 5/5

    Prikaži ovu nit
    Poništi
  6. 2. velj

    Having said that, I would suggest people to have a look at the method for correlated/separable variables proposed by Fani. Then you can happily move to MMC if the assumptions in your case do not hold (i.e., your variables are not correlated/separable). :) 4/5

    Prikaži ovu nit
    Poništi
  7. 2. velj

    I would argue that in many practical cases you have variables that show correlation in experimental studies. But I understand your point that this might not be true in general and you might need a more complex method like MMC you use in your paper. 3/5

    Prikaži ovu nit
    Poništi
  8. 2. velj

    The type of analysis proposed in this paper seems necessary in case of non separability of random variables and when the correlation tests fail. In that case, you probably need a non-linear model as that described in the paper, but not in general. 2/5

    Prikaži ovu nit
    Poništi
  9. 2. velj

    Related to this, I would like to point out (and promote!) the PhD work of my student Fani Tsapeli on causality vs correlation analysis of longitudinal data for ubiquitous computing/mobile sensing/HCI/etc. The methodological framework is described here: 1/5

    Prikaži ovu nit
    Poništi
  10. 29. sij

    The deadline for applications to the Doctoral Training Programme for the Ecological Study of the Brain is coming up soon (6 Feb)! I am personally interested in discussing projects in the area of computational models of human behaviour.

    , , i još njih 6
    Poništi
  11. proslijedio/la je Tweet
    27. sij
    Poništi
  12. proslijedio/la je Tweet
    24. sij
    Odgovor korisniku/ci
    Poništi
  13. proslijedio/la je Tweet
    24. sij

    It would be interesting to have anti-acknowledgements in papers: "This work was completed despite being declined for funding by the following agencies..."

    Poništi
  14. 24. sij

    Timely book found lying around in the department: “Should Britain join the single currency?”

    Poništi
  15. 22. sij

    In case you don't have institutional access to Pervasive Computing, you can find the pre-print here:

    Prikaži ovu nit
    Poništi
  16. 22. sij

    Our paper "Interpretable Machine Learning for Privacy-Preserving Pervasive Systems" with is now out in .

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet

    Our amazing will run to become a World Heritage Site! Info: Foto: Giorgio Bianchi

    Poništi
  18. 21. sij

    The 3rd edition of the UK Mobile, Wearable and Ubiquitous Systems Research Symposium (MobiUK 2020) will take place at on 6-7 July 2020. Submit your 1-page abstract by 1 April!

    , , i još njih 7
    Poništi
  19. proslijedio/la je Tweet
    17. sij

    More about politics of location data and AI are in my forthcoming book!

    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:

    ·