Charlie Marx

@chtmarx

Fairness and interpretability in machine learning. Student

Vrijeme pridruživanja: srpanj 2018.

Tweetovi

Blokirali ste korisnika/cu @chtmarx

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

  1. proslijedio/la je Tweet
    17. pro 2019.

    Congrats to the winners of the CRA undergrad research award! Winners include lots of theory & ML people. Meena and Alex (not on Twitter?) are in theory (both advised by ). Others on Twitter include @CharlieMarx9 1/2

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

    “We typically assume that there’s one best model, but in practice there can be many models that produce different results.” -

    Poništi
  4. proslijedio/la je Tweet
    12. pro 2019.

    Check out @CharlieMarx9 and at NeurIPS poster #107! Feature correlation is confusing in influence methods and they have answers!

    Poništi
  5. proslijedio/la je Tweet
    6. stu 2019.

    Wondering if you can game explainability methods (e.g. LIME/SHAP) to say whatever you want to? Our recent research suggests this is possible.

    Poništi
  6. proslijedio/la je Tweet

    CompSci/math double major Charlie Marx '20 is headed to the NeurIPS conference in Vancouver to present his published work (co-authored w/ & Richard Phillips ’18) on reducing bias in machine-learning models to produce more equitable results.

    Poništi
  7. proslijedio/la je Tweet
    28. stu 2019.

    If you go to , consider attending our workshop on human-centric ML , we have a terrific set of invited speakers (, Deirdre Mulligan, , Finale Doshi-Velez, ), contributed papers and panels!

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

    Philly folks: keynote by danah boyd this Thursday at 4:30pm. See you there! Vulnerabilities: How Social Media and Data Infrastructure are Exploited for Fun, Profit, and Politics Open to the public. On SEPTA regional rail and NHSL.

    Poništi
  9. proslijedio/la je Tweet

    My very first oped. Hopefully not my last. On AI-based tech surveillance in Utah.

    Poništi
  10. proslijedio/la je Tweet
    4. ruj 2019.

    Super proud that student @CharlieMarx9 had a paper accepted to ! Disentangling Influence applies disentangled reps to indirect influence of features on a model. Joint w/ alum , me, , and .

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    9. lip 2019.

    1/ shapley values (and the Python shap package) have become an *integral* part of my machine learning methodology I’m surprised by how many people still don’t know about them

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    5. lip 2019.
    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:

    ·