How to solve (almost) any problem using Markov Chain Monte Carlo

Abstract:
MCMC sampling has allowed for simulation from complex, massively multivariate distributions even when that distribution is only specified up to an unknown normalizing constant. This talk presents the theory behind maximum likelihood parameter estimation using simulation (MCMC-MLE), and little known but vitally important implementation issues and heuristics focusing on exponential families of distributions. In particular, a useful approximation to the normalizing constant in the the exponential family likelihood is presented which increases the stability and accuracy of the MCMC-MLE algorithm. We also show how MCMC standard errors can be used as a measure of when to trust this approximation. Finally, simple examples are used to illustrate how this algorithm can be used to perform inference on a new class of models for social networks dubbed Exponential-Family Random Network Models (ERNM).

Bio:

Dr. Ian Fellows is the founder an president of Fellow Statistics Inc., a statistical consulting firm providing advanced analytic support to the corporate and government worlds. His research interests range over many sub-disciplines of statistics, including statistical visualization where his work won the John Chambers Award in 2011, artificial intelligence where his Texas Hold'em AI programs were ranked second in the world [masked]), and Markov Chain Monte Carlo methods which were used extensively in his research in social network theory. For more information, see http://www.fellstat.com.

 

Note:

There is no validated parking. You can find free parking at: 1036 Broxton Ave Los Angeles, CA 90024. Also please check in at the front desk before going upstairs.

Join or login to comment.

  • Jeff W.

    I'm new to the group but I see a few request for someone to video tape. I work as a writer but also a videographer and editor. If no one else is doing it (and I'd be happy if they are) I will be glad to videotape if I have permission. (and I can post to YouTube - where we can require a link from the group to read or we can make it public (much easier). Let me know. Jeff

    4 · April 15, 2013

    • David M.

      It was recorded, but I haven't seen a link yet.

      April 18, 2013

    • Jeff W.

      yes, I video taped it. Hope to have it up this week on this group's meetup channel. Sorry for delay.

      1 · April 22, 2013

  • RN

    I am really new to machine learning. It would have been great to have more of an idea of what he was talking about. To fellow newbies taking Andrew Ng's machine learning class through Coursera beginning April 22 and wanting to form a study group, I say "Let's do it!"

    April 17, 2013

    • lee

      We have a location in Lincoln Heights called Droplabs that would be perfect. Whiteboards, conference room, etc. We can totally host the class.

      April 18, 2013

    • lee

      Someone who knows I'm interested in this just pointed me to this thread. My apologies. Will take to forum.

      April 18, 2013

  • RN

    Are there any other newbies taking Andrew Ng's Machine Learning Class through Coursera beginning April 22?

    1 · April 14, 2013

    • Leela Krishna K

      I am enrolled to this class too. Yay! for a study group on ML!

      April 14, 2013

    • RN

      Does Peter's study group pertain to Ng's up-coming Machine Learning class? If so, what are the details?

      April 17, 2013

  • Daniel G.

    Just like a graduate seminar, wonderful!

    1 · April 16, 2013

  • Rob Z.

    The entrance is on Westwood.

    April 16, 2013

  • Yasmin L.

    Finally got here, but can't get inside.

    April 16, 2013

  • Andrei K.

    How do I get into the building? The front doors are locked and there ate no attendants.

    2 · April 16, 2013

  • Aran D.

    Fun fun!

    April 16, 2013

  • Dante D.

    Just found out I have to work late this time - looking forward to next event!

    April 15, 2013

  • Tim Triche, J.

    have to babysit! :-|

    April 15, 2013

  • Rudiger L.

    Just found out that I can't make it to this talk. Is there any chance it will be filmed (and posted online later)?

    April 14, 2013

    • Yuli L.

      agree. This is such an interesting topic and I can't make it. It will be great to be able see it on-line later on.

      1 · April 15, 2013

  • Eric K.

    Sorry to miss it!

    April 15, 2013

  • Victor W.

    We have 159 folks signed up. Will the venue support this size group?

    April 15, 2013

    • Rob Z.

      Usually about 50% of people don't show up. We should be fine.

      April 15, 2013

  • Rupesh

    Great topic and I would like to join except I'm out of town during that whole week. Any chance for video conferencing or recording for viewing later?

    April 12, 2013

  • Dante D.

    Great Topic!

    April 9, 2013

  • Andrew P.

    Looking forward to it!

    April 8, 2013

  • A former member
    A former member

    Cool, 10 minutes from home

    April 8, 2013

  • Dustin B.

    sounds like a great topic!

    April 5, 2013

  • Andrei K.

    MCMC is a very exciting topic. Looking forward to it!

    April 5, 2013

  • Jeff W.

    I 2nd that emotion. Westwood would be great.

    April 4, 2013

  • A former member
    A former member

    West-Side++

    1 · April 4, 2013

Our Sponsors

People in this
Meetup are also in:

Imagine having a community behind you

Get started Learn more
Henry

I decided to start Reno Motorcycle Riders Group because I wanted to be part of a group of people who enjoyed my passion... I was excited and nervous. Our group has grown by leaps and bounds. I never thought it would be this big.

Henry, started Reno Motorcycle Riders

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