Wednesday, February 15, 2012

Sloans and More

The Alfred P. Sloan Research Fellows were announced today including Northwestern's own Nicole Immorlica. Other winners in theoretical computer science include Xi Chen, Nate Foster and Prasad Raghavendra. A shout out to TTIC who have their second Sloan Fellow in Jinbo Xu.

Computer science did well in the president's budget for FY 2013. CISE head Farnam Jahanian gives the details. Of course now the budget has to get through congress.

Tomorrow there will be a celebration of twenty years of the NITRD (Networking and Information Technologies Research and Development), an interagency program that has heavily supported CS research over the past two decades. Quite an impressive list of speakers. There will be a live webcast of the event.

FOCS call for papers is out. Submission deadline is April 4.

7 comments:

  1. Congratulations to all Sloan Fellows this year.

    However, I want to mention that we should not exaggerate the importance of this award and consider it very seriously, esp. because Sloan organization itself already tries to exaggerate this award by calling it a “Prestigious Award” or publishing the results in NYTimes. I will consider this award like other media awards, e.g., TR35, etc. Of course not mentioning the amount of this award which is very small comparing to other grants/awards like NSF CAREER, I don’t think the selection process is also very wise and comparable at all to major awards like NSF CAREER, PECASE, or other governmental awards. As I have been in several panels for such major NSF awards, usually there are at least 10 experts in each area as the panelists coming from different regions, discussing all applications in details, and considering very seriously conflict of interests. However e.g. by looking at the selection committee of Sloan this year at http://www.sloan.org/fellowships/page/9 there are only three persons in the committee for the whole CS (in this year Alex Aiken, Stanford University; Margo I. Seltzer, Harvard University; Éva Tardos, Cornell University) and each of them are selected for 3 years. Also as far I know there is no concept of conflict of interests. So as a result in theory for three consecutive years one person has the main power to select the theory Sloan fellows (similarly in other areas). Indeed due to lack of conflict of interest and also self-interests (e.g. for promoting their departments), it is easy to check that usually there are much more people from the same departments of the selection committee members as a result (just imagine what happens in a conference in which one person alone, namely the chair, selects all papers for three years:)).

    So I think other well-organized awards like NSF CAREER or PECASE awards should be considered much more seriously than awards like Sloan with not a
    very well-organized and comprehensive selection process.

    ReplyDelete
    Replies
    1. Not to take any credit away from the awardees, but the argument above has a degree of truth to it.

      For example, there is at least one advisor/advisee pair in the selection committee/awardees (go figure it out yourself!). Moreover, I cannot help but notice that the theory awards were to folks working in algorithmic game theory. That makes sense too, considering the research interests of the theory person on the selection committee.

      Delete
  2. I don't think Prasad Raghavendra works in Algorithmic Game Theory unless you think AGT includes "Unique Games".

    WRT faculty having conflict of interests: many awards have conflicts of interests. There have been recent major awards such as the Goedel prize in which one of the recipients was the former student of the chair of the committee! No one seemed to have a problem with that. At MIT, there are best thesis awards, which is an award for the student, but which also helps the advisor, but the advisor is the nominator. So it's not just a problem with Sloan Awards.

    ReplyDelete
  3. I agree with the first Anonymous. Any award without a way to manage the conflict of interests has these serious issues. I know as a fact that the statement of the first Anonymous is correct for Sloan. About Goedel I'm not sure. Note that Goedel's selection committee this year has 6 members http://www.sigact.org/Prizes/Godel/goedel_call12.pdf
    and all in the same area of theoretical computer science and at least two of them will change each year. This is a bit different from Sloan with one CS theory person for 3 years.

    Apart from this, there are only three theory persons this year Nicole, Xi, and Prasad (two in game theory) which is strange since usually there are at least 5-6 people in theory.

    ReplyDelete
    Replies
    1. French food: tastes horrible and the portions are too small

      Sloans in theory: biased choices and not enough of them

      Delete
  4. What's with French food? Their pastries are awesome, but what else?

    ReplyDelete
  5. ``At MIT, there are best thesis awards, which is an award for the student, but which also helps the advisor, but the advisor is the nominator. So it's not just a problem with Sloan Awards.''

    This is irrelevant. Surely advisers can nominate their advises for an award, however selecting them would be conflict of interests.

    ReplyDelete