align-toparrow-leftarrow-rightbackbellblockcalendarcamerachatcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-crosscrosseditemptyheartfacebookfullheartglobegoogleimagesinstagramlocation-pinmagnifying-glassmailmoremuplabelShape 3 + Rectangle 1outlookpersonplusImported LayersImported LayersImported Layersshieldstartwitteryahoo

Talk by Prof. Michael Jordan (UC-Berkeley)

Title:  On the Computational and Statistical Interface and "Big Data"

Speaker: Michael I. Jordan is the Pehong Chen Distinguished Professor in the Department of Electrical Engineering and Computer Science and the Department of Statistics at the University of California, Berkeley. 

Abstract: The rapid growth in the size and scope of datasets in science and technology hascreated a need for novel foundational perspectives on data analysis that blendthe statistical and computational sciences.  That classical perspectives from thesefields are not adequate to address emerging problems in "Big Data" is apparentfrom their sharply divergent nature at an elementary level---in computerscience, the growth of the number of data points is a source of "complexity"that must be tamed via algorithms or hardware, whereas in statistics, the growthof the number of data points is a source of "simplicity" in that inferences aregenerally stronger and asymptotic results or concentration theorems can be invoked.We present several research vignettes on topics at the computation/statisticsinterface, an interface that we aim to characterize in terms of theoretical tradeoffsbetween statistical risk, amount of data and "externalities" such as computation,communication and privacy.  [Joint work with Venkat Chandrasekaran, John Duchi,Martin Wainwright and Yuchen Zhang.]


Join or login to comment.

  • DB T.

    Here is the video for Prof. Jordan's talk at youtube. https://www.youtube.com/watch?v=zdavG9xbVp0&feature=youtu.be

    September 22, 2014

    • Shivani R.

      thanks so much DB Tsai! :) this is awesome

      1 · September 23, 2014

  • Chester C.

    Here is video for Prof. Michael Jordan't talk. Thanks for Yelp again. https://www.dropbox.com/s/l4wib40jmqru8l5/SFMachineLearning_MichaelIJordan_082014sbs.mp4?dl=0

    August 28, 2014

    • Chester C.

      From Yelp: Clicking on the dropbox link opens a 15 minute streaming video as a preview for the downloadable file, and then you'll have click the download button at the top of that page for the full video.

      September 18, 2014

    • DB T.

      I upload the video to youtbe, and it's easier for audience to access. https://www.youtube.co...­

      September 22, 2014

  • Douglas L.

    I received a notice that I was confirmed off the waiting list less than an hour before the meet up was scheduled to begin. At that point it would've been impractical to make it to the city in time.

    August 22, 2014

    • Sara A.

      I'm sorry about that! Our system for managing the waitlist didn't work very well for this event. We'll try to figure out something better before the next event.

      August 28, 2014

  • Jack k.

    This was a great talk. It's a shame there were empty seats because some people didn't show up.

    1 · August 21, 2014

    • Martin F.

      Another Meetup request for enhancement is to show which # we are in the waitlist queue (e.g., next up, 2 more in front of you). Or give us a Bayesian apriori probability of us getting off the waitlist and getting a spot. Those with a chance can monitor meetup more often or bet on getting a spot and can be at the event just in case. For those that travel some distance only with adequate quantitative knowledge can we adequately gauge a risk of driving a long way and have a chance versus get turned away.

      August 22, 2014

    • Jacob Zekai G.

      I got off the checklist like 10 days ago but they failed to find my name two days ago. Had to show the confirmation email to get in. :/

      August 22, 2014

  • Nathan L.

    Great talk!

    August 21, 2014

  • Erin L.

    Here is a link to the slides (from another talk, but same slidedeck): http://www.stat.harvard.edu/NRC2014/MichaelJordan.pdf

    1 · August 21, 2014

    • Pieter de J.

      Thanks, that's great! Any additional material is welcome.

      August 21, 2014

  • David W.

    And this article linked from Prof Jordan's Publications page seems a pretty good summary of the first example he offered. I was too shy to raise the question of whether their work covered continuous distributions, but this paper does have sections on density estimation, so I guess that answer is "yes": http://www.cs.berkeley.edu/~jordan/papers/duchi-jordan-wainwright-nips14.pdf

    August 21, 2014

  • David W.

    If you are in a hurry to see the slides, this PPT presentation looks like a fair representation of the third example Prof Jordan offered. http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&ved=0CF8QFjAG&url=http%3A%2F%2Fwww.samsi.info%2Fsites%2Fdefault%2Ffiles%2FJordan_september2012.ppt&ei=fCv2U67_AaSCjAK54YCYBg&usg=AFQjCNHV-4_9AGcnVw9gWN47Ujpyk0We0A&sig2=tKRcPjYw2RrZGAnerX6AFQ

    (I do wish Google would go back to offering direct links.)

    August 21, 2014

  • Mauricio R.

    Prof. Jordan mentioned 3 ways to find confidence intervals: the bootstrap, the central limit theorem, but I did not catch the third one. Does anyone remember?

    August 21, 2014

    • David W.

      My memory: Cramer-Rao bounds.

      August 21, 2014

  • Bill Z.

    is there pizza?

    August 20, 2014

  • Pieter de J.

    Will the slides be posted anywhere?

    August 21, 2014

  • Chris N.

    Great talk tonight -- thanks Michael, Argyle and Yelp!

    If anyone is interested in open-source, distributed deep learning in the JVM, Deeplearning4j will be giving two workshops next month. Feel free to contact me if you have questions. :)

    September 13-14 @9:00am-4:30pm
    (Saturday-Sunday)
    Workshop: Practical Deep Learning & Its Applications (Hosted by Zipfian)
    Learn how to leverage deep learning with a curriculum built by industry experts.
    452 Shotwell St.
    San Francisco, CA 94110

    http://www.zipfianacademy.com/workshops/practical-deep-learning-its-applications

    September 25-26 @9:00am-5:00pm
    (Thursday-Friday)
    Workshop: Distributed Deep Learning with Deeplearning4j
    Academy by the Bay continues with its hands-on, two-day professional training courses this fall with its first course focused on deep learning.

    http://www.meetup.com/Academy-By-the-Bay/events/197376172/

    1 · August 20, 2014

  • Sara A.

    Thank you everyone for coming to our meetup yesterday! We will be posting the video as soon as it's available.

    Also, someone left a pair of headphones at the event. Please let me know if they're yours.

    Regards,
    Sara

    August 21, 2014

  • A former member
    A former member

    In San Francisco visiting from London. Would be interested to meet other data scientists while I'm out there

    August 3, 2014

    • Mauricio R.

      Hello Jahan, I would be interested in meeting. Please contact me at [masked]

      August 21, 2014

  • Chester C.

    In few days once we get the video from yelp

    1 · August 21, 2014

  • Warren

    When will the talk be available online?

    1 · August 21, 2014

  • A former member
    A former member

    Great talk.

    August 20, 2014

  • Phillip W.

    Any hope for those who are wait listed?

    1 · August 20, 2014

  • Lauren M.

    Is there a cutoff time to be checked in or can we arrive up until the talk starts at 7p?

    August 20, 2014

  • Conny

    N/a

    August 20, 2014

  • Sara A.

    Hi Raymond. Meet and greet is from 6:30-7:00, and the talk will run from 7:00 to 8:00. The meetup should be finished by 8:30.

    August 19, 2014

  • Raymond P.

    What exactly is the schedule? Meet and greet 6:30-7:30 and then talks 7:30-8:30 or 7:30-9:00?

    1 · August 19, 2014

  • Taufiq H.

    I hope I can get in. Am really interested in this talk

    August 17, 2014

  • Chester C.

    Just talked to Yelp and confirmed that Video will be available few days after the event. Streaming is not available.

    5 · July 3, 2014

  • Chester C.

    The venue has 250 capacity limit, and we need leave some for people at yelp who will work at event. But considering there is always certain % will not show up , some on the waiting list might be able to get in

    1 · July 3, 2014

  • Chester C.

    Yelp has great recording system and engineering, they usually record all events. We will contact them to make sure this is recorded and publish to the members

    1 · July 3, 2014

  • louis d v.

    any chance of the talk being recorded or available in live streaming ?

    1 · July 3, 2014

  • Austin D. T.

    How do I get off the wait list and on the RSVP?

    1 · July 2, 2014

    • henry c.

      well, that filled up quickly. O.o

      1 · July 3, 2014

  • David s.

    great! I attended one of his talk several years ago about Dirichlet Process. It's wonderful!

    July 2, 2014

  • Ted F.

    Prof. Jordan has an impressive group of advisees:

    Yoshua Bengio (Postdoc)
    Andrew Ng
    David Blei
    Alice Zheng

    ...to name a few!

    1 · July 2, 2014

Our Sponsors

  • Yelp

    venue, food & drinks

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy