glial

42 ms·

1. 481.▲
   by glial 6y ago

   Can you give some examples?
2. 482.▲
   by glial 6y ago

   It seems correct to a first order of approximation for the general population. What is your objection? I know people who lift weights regularly can have high BMI but low body fat. But until those people make up a large share of the populati
3. 483.▲
   by glial 6y ago

   It can predict, but that’s a different thing entirely.
4. 484.▲
   by glial 6y ago

   I think it’s more that the worldview held by western institutions and police in particular does not equip police to actually solve the problems they are trying to solve - through no real fault of the individuals involved.
5. 485.▲
   by glial 6y ago

   So glad to hear that! I loved her Odyssey translation. Her introduction alone is worth the price of admission.
6. 486.▲
   by glial 6y ago

   Digital signal processing maybe?
7. 487.▲
   by glial 6y ago

   Facebook’s (very good) Prophet forecasting library is a wrapper for Stan models. https://facebook.github.io/prophet/
8. 488.▲
   by glial 6y ago

   Is there a record of the address he gave? I would love to read it.
9. 489.▲
   by glial 6y ago

   There is strong incentive to do so, so I would wager a yes.
10. 490.▲
    by glial 6y ago

    I have never heard before about such a contention. Both Shannon and Wiener were geniuses IMO with very pivotal contributions.
11. 491.▲
    by glial 6y ago

    Felons often cannot vote, cannot get jobs working with children, have difficulty finding housing, etc.
12. 492.▲
    by glial 6y ago

    Inside a faraday cage to prevent RAM from emitting radio waves.
13. 493.▲
    by glial 6y ago

    That may say as much about our egocentric media landscape as anything.
14. 494.▲
    by glial 6y ago

    Those “invincible” young people don’t live by themselves, you know.
15. 495.▲
    by glial 6y ago

    Mine required a PDF submission that met very strict formatting guidelines, but they didn't have a LaTeX template available either. I asked the graduate student help desk if I could use one from a former student's GitHub repo, and
16. 496.▲
    by glial 6y ago

    Good idea. I've found the best way to engage (if you want to engage) is to just take the argument seriously, and ask what kind of society they envision, in detail. It gets awkward quickly.
17. 497.▲
    by glial 6y ago

    How would that have helped, in this case? Should the judge's son have answered the door with a gun in his hand?
18. 498.▲
    by glial 6y ago

    I found it difficult to read too. Here's an annotated version, with code, which helps: https://nlp.seas.harvard.edu/2018/04/03/attention.html
19. 499.▲
    by glial 6y ago

    Having had a similar experience, my new strategy is to set up a server and just ssh in to do ML development. That way I don't have to worry about mouse/display/wifi drivers.
20. 500.▲
    by glial 6y ago

    > you can look at magnitude and direction of various coefficients and conclude certain obvious things. Sometimes, but if you have correlated predictors, the 'explainability' can be distributed across coefficients of those predi
21. 501.▲
    by glial 6y ago

    Curious where you get your numbers, because I think they're wrong.
22. 502.▲
    by glial 6y ago

    What is a probability? Nearly always, a probability isn’t a statement about the world, it’s a statement about your knowledge and the information you have. Even for frequentist statisticians, equating probabilities with proportions of outcom
23. 503.▲
    by glial 6y ago

    Same here, but Reed and University of Chicago.
24. 504.▲
    by glial 6y ago

    > is anyone really training models on a Macbook or a Mac Mini? In an academic environment, absolutely. Scaling is the second problem to solve, but iterating on model architecture is the first.
25. 505.▲
    by glial 6y ago

    Asking that question isn’t conspiracy theorizing, it’s just due diligence. Like currency, elections live or die by peoples faith in them. It’s really, really important that everybody believes the outcome.
26. 506.▲
    by glial 6y ago

    Make the tax so high that the only feasible use of plastic is medical supplies and (very) durable goods.
27. 507.▲
    by glial 6y ago

    Thanks!
28. 508.▲
    by glial 6y ago

    Thanks for sharing. My own pet theory is very similar to yours: humans evolved consciousness in order to ‘observe’ our cognitive and emotional state, so that we can build models and apply them to reason about other people’s cognitive and em
29. 509.▲
    by glial 6y ago

    There's a reasonable argument to be made against the parent, but "what about" isn't it.
30. 510.▲
    by glial 6y ago

    Having just finished a 10 year PhD stint, I second this advice. It is absolutely critical to have a supportive department, secure funding sources, and an advisor with whom you can work well. If any of those is not the case, it’s a recipe f