Manjari Narayan

@NeuroStats

Interested in inferential validity of network science. Also graphical models, statistical causal inference. Maker of skggm. Ph.D | B.S

bit.ly/narayan2017
Joined November 2014
48 Photos and videos Photos and videos

Tweets

You blocked @NeuroStats

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @NeuroStats

  1. Pinned Tweet
    10 Oct 2016

    My side project with "skggm: Bringing the best of Gaussian graphical models to sklearn"

    2 replies 24 retweets 32 likes
  2. Jun 2

    Did not realize people systematically use "gauntlet" to mean "gantlet". The distinction seems nearly gone now.

    1 like
  3. Retweeted
    May 27
    Replying to

    16/Most Americans are decent ppl like Ricky Best and Taliesin Namkai-Meche. But they have little representation among the shouting classes.

    19 replies 197 retweets 811 likes
  4. May 28
    Replying to

    Or even if they agree in theory, people tend to give one or other more credit then it deserves.

    2 likes
  5. May 28

    Consensus ≠ Truth Contrarianism ≠ Truth. So many quick to agree with one but not the other. Both useful but not substitute for evidence.

    2 replies 30 retweets 52 likes
  6. May 28

    Yeah like what the hell is an "Asian dialect"? Even more egregious than being asked if Indians speak Indian. Why not Asian languages?

    The comments on this thread cracked me up. (But full disclosure: I confess to having enjoyed The DaVinci Code)
    2 replies 2 likes
  7. Retweeted
    May 28
    Replying to

    5/ A moment of great character can produce other moments like it when folks take the time to keep that memory alive.

    1 reply 189 retweets 852 likes
  8. May 27

    Selling Long Life. "Not to worry. The film, called The Immortalists, is coming to a theater near you."

  9. Retweeted
    May 26

    Lord, grant me the confidence of a Hacker News commenter criticizing an article they haven't read

    20 replies 660 retweets 2,419 likes
  10. May 25

    So much for my playing around with . Given how often the clip slipped off this past month, losing it for good was inevitable.

    3 replies
  11. May 25

    "the only way to do great work, in any field, is to find time to consider the larger questions."

    My pre-grad school job I asked to work 3 days/week 10 hours/day for 75% of my salary. Spent the extra time coding :)
    6 likes
  12. May 23
    Replying to

    I meant Mark Newman the physicist.

    1 like
  13. May 23
    Replying to

    References: Wonderful/classic paper from and

    2 replies 1 like
  14. May 23
    Replying to

    4/ So it is tricky business to interpret your scaling estimates without knowing that they are unbiased & quantifying precision.

    1 reply 1 like
  15. May 23
    Replying to

    3/ In particular, you need something better than the usual line-fitting /least-squares estimates.

    1 reply 2 likes
  16. May 23
    Replying to

    2/ Difficult to estimate scaling parameter of distributions accurately at finite samples.

    1 reply 1 like
  17. May 23
    Replying to

    1/ Tricky to determine whether data merely consistent with power-law AND other distributions or uniquely explained by power-law

    2 replies 5 likes
  18. May 23

    "So you think you have a power law?" Some required reading for the fans

    1 reply 1 retweet 10 likes
  19. Retweeted
    May 23

    Loving math as a teen is just as essential for getting to grips with machine learning as Ancient Greek is for finance.

    6 replies 3 retweets 10 likes
  20. Retweeted
    May 23

    Today's science spam: telling bad jokes and getting them wrong.

    1 reply 2 retweets 1 like
  21. May 23

    Rare indeed. "I have not only stepped back from an entire research area, but also allowed and facilitated the transfer of grant funds ..."

    We should all be so lucky to receive this level of mentorship from senior scientists -
    1 retweet 3 likes

@NeuroStats hasn't Tweeted yet.

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·