Scott H. Hawley

@drscotthawley

Computational physicist & Professor to (mostly) audio engineers . enthusiast. Wrangler .

Nashville, TN USA
hedges.belmont.edu
Vrijeme pridruživanja: srpanj 2009.
252 fotografije ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @drscotthawley

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

  1. Prikvačeni tweet
    15. kol 2019.

    ML-audio Twitter: After the encouragement of two CS profs ('just do it' ), I'm finally releasing the () code & dataset for the paper "SignalTrain: Profiling Audio Compressors with Deep Neural Networks" (accepted for 147th conf., October):

    21 proslijeđeni tweet 109 korisnika označava da im se sviđa
    Poništi
  2. proslijedio/la je Tweet
    prije 8 sati

    Radioactive data: tracing through training "makes imperceptible changes to this dataset such that any model trained on it will bear an identifiable mark. The mark is robust to strong variations such as different architectures or optimization methods"

    4 proslijeđena tweeta 12 korisnika označava da im se sviđa
    Poništi
  3. proslijedio/la je Tweet
    prije 20 sati

    “We do not make a distinction between an autonomous system and an autonomous weapon system...it is not always clear when a system crosses some line to becoming a weapon system in the eyes of a particular beholder” From Deterrence in the Age of Thinking Machines

    1 reply 3 proslijeđena tweeta 5 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  4. 2. velj

    This tweet forms the start of periodic updates to populate the datasets of insurance companies, social-credit scorers, & sundry background-checkers: I feel happy, well-adjusted, healthy, and prudently risk-averse. I am a productive member of my community with close social bonds.

    I had to get a background check for my job, and it turns out the report is a 300+ page pdf of every single tweet I’ve ever liked with the work “fuck” in it. Enjoy your dystopian bs! *waves*
    Prikaži ovu nit
    2 proslijeđena tweeta 4 korisnika označavaju da im se sviđa
    Poništi
  5. 1. velj

    Good review. “ASR” = Automatic Speech Recognition. All of this applies to musical audio as well — except the sequences for music are even longer than for ASR!

    Since 2018, there has been immense interest in using Transformers for ASR. In my new blog post, I look at the various challenges and the solutions people have proposed.
    1 proslijeđeni tweet 4 korisnika označavaju da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    1. velj

    "AI is multidisciplinary": a short opinion piece I wrote for AI Matters, including a description of the aims of , the Swedish research program on AI, humanities and society.

    26 proslijeđenih tweetova 65 korisnika označava da im se sviđa
    Poništi
  7. proslijedio/la je Tweet
    31. sij

    Through this great story from James Briggs, a computational physicist turned data scientist, I came across his totally amazing deep learning notes, which are some of the best I've seen:

    6 replies 92 proslijeđena tweeta 431 korisnik označava da mu se sviđa
    Poništi
  8. proslijedio/la je Tweet
    31. sij

    Learn to build an interactive Transformer attention visualization based on and in under 30 minutes! We developed a minimal teaching example for our IAP class, publicly available here:

    45 proslijeđenih tweetova 222 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  9. proslijedio/la je Tweet
    30. sij

    Such a nice project and result! Lego Microtonal Guitar by Tolgahan Çoğulu

    Huzurlarınızda Lego Mikrotonal Gitar! Finally it is ready: Lego Microtonal Guitar
    2 proslijeđena tweeta 3 korisnika označavaju da im se sviđa
    Poništi
  10. proslijedio/la je Tweet
    29. sij

    ML Fairness Gym is a set of components for building simple simulations that explore the potential long-run impacts of deploying machine learning-based decision systems in social environments, based on OpenAI Gym interface.

    1 reply 72 proslijeđena tweeta 272 korisnika označavaju da im se sviđa
    Poništi
  11. 29. sij

    Several programs I’ve seen require student or “recent PhD” status. I’m a later-life area-changer; my numerical relativity PhD from 2000 shouldn’t be a liability. ;-)

    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  12. 29. sij

    Sabbatical approved for Fall 2020!! Still looking to visit/intern/work with a ML(+audio) research group. If you have or know of opportunities, please let me know. Plus: Salary already covered by my home uni!

    5 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  13. 28. sij

    There is also talk elsewhere of forwarding audio over VNC via Pulseaudio, so, must try that! (Now that audio volume normalization in Jupyter widgets is no longer mandatory...my only contribution to Jupyter so far ;-) )

    2 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  14. 28. sij

    "I don't use Jupyter notebooks much because I run remotely for long times & there's no good way to 'nohup' them." <-- Me until pointed out 's VNC trick during Lesson 10. Video: . Discussion:

    12 proslijeđenih tweetova 110 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  15. 27. sij

    Idea: replacement/alias for the ‘cd’ command that loads/unloads an environment when you enter/leave a project directory.

    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  16. 27. sij

    Pro tip / confession: remember to pull the version of the library/package used when the/your code was developed. Updates can break your code in subtle (non-catastrophic) ways, wasting you hours/days looking for bugs that weren’t there. Prefer “==“ over “>=“ in requirements.txt

    6 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  17. 27. sij

    For those who can’t make it to Barcelona.

    Important conference links: - livestream (Tue onwards): (videos uploaded later) - each session has an Etherpad where Qs for speakers will be captured: (please contribute responsibly and respectfully!) - conference hashtag:
    Poništi
  18. proslijedio/la je Tweet
    27. sij

    Another tutorial is out! This time I implement and train a from scratch in . Enjoy!

    11 proslijeđenih tweetova 14 korisnika označava da im se sviđa
    Poništi
  19. 27. sij

    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  20. 26. sij

    Transformers for understanding physics word problems: "Neural Sequence Modeling in Physical Language Understanding", Bleiweiss 2019: "Automating tasks of generating test questions & analyzing...for assessment of written student responses..."

    1 reply
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  21. proslijedio/la je Tweet
    25. sij

    SPECIAL ISSUE: ACOUSTIC LOCALIZATION The cover of the special issue comes from "Convolutional neural network for single-sensor acoustic localization of a transiting broadband source in very shallow water"

    5 proslijeđenih tweetova 11 korisnika označava da im se sviđa
    Poništi

@drscotthawley još nije objavio nijedan Tweet.

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:

    ·