Monte Carlo Simulation for Statistical Inference, Model Selection and Decision Making

author: Nando de Freitas, Department of Computer Science, University of British Columbia
published: March 13, 2008,   recorded: March 2008,   views: 28438


Related content

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.
Lecture popularity: You need to login to cast your vote.
  Delicious Bibliography

 Watch videos:   (click on thumbnail to launch)

Watch Part 1
Part 1 51:48
Watch Part 2
Part 2 56:50
Watch Part 3
Part 3 50:22
Watch Part 4
Part 4 1:02:26
Watch Part 5
Part 5 42:25
Watch Part 6
Part 6 59:02


The first part of his course will consist of two presentations. In the first presentation, he will introduce fundamentals of Monte Carlo simulation for statistical inference, with emphasis on algorithms such as importance sampling, particle filtering and smoothing for dynamic models, Markov chain Monte Carlo, Gibbs and Metropolis-Hastings, blocking and mixtures of MCMC kernels, Monte Carlo EM, sequential Monte Carlo for static models, auxiliary variable methods (Swedsen-Wang, hybrid Monte Carlo and slice sampling), and adaptive MCMC. The algorithms will be illustrated with several examples: image tracking, robotics, image annotation, probabilistic graphical models, and music analysis.
The second presentation will target model selection and decision making problems. He will describe the reversible-jump MCMC algorithm and illustrate it with application to simple mixture models and nonlinear regression with an unknown number of basis functions. He will show how to apply this algorithm to general Markov decision processes (MDPs). The course will also cover other Monte Carlo simulation methods for partially observed Markov decision processes (POMDPs) using policy gradients, common random number generation, and active exploration with Gaussian processes. An outline to some applications of these methods to robotics and the design of computer game architectures will be given. The presentation will end with the problem of Monte Carlo simulation for Bayesian nonlinear experimental design, with application to financial modeling, robot exploration, drug treatments, dynamic sensor networks, optimal measurement and active vision.

See Also:

Download slides icon Download slides: mlss08au_freitas_asm.pdf (14.4┬áMB)

Help icon Streaming Video Help

Link this page

Would you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !

Reviews and comments:

Comment1 Ilyas, March 14, 2008 at 4:19 p.m.:

Good presentation. The cameraman is sleepy

Comment2 Lev, March 16, 2008 at 9:08 a.m.:

Excellent... Prof.'s Examples are really useful and make us easy to apply in real time

Comment3 Crt, March 20, 2008 at 6:45 p.m.:

Where are the other 3 lectures?

Comment4 Peter Kese (staff), March 20, 2008 at 7:16 p.m.:

There are altogether 6 videos in this lecture. Only three of them have been sent directly from the event. The other three will be published once we get the tapes from Australia - probably within a week or so.

Comment5 Ciamak, April 24, 2008 at 7:20 p.m.:

Very good lecture. Thank You. I'm also interested in the other 3 lecutres, but they are not published yet app. one month later. Is there still an chance to get them ?

Comment6 Yoba, May 10, 2008 at 11:24 p.m.:

Excellent teacher!

Comment7 Nina (staff), May 15, 2008 at 4:31 p.m.:

The lecture from Nando de Freitas is now complete.

Comment8 mlss08 participant, June 13, 2008 at 3:39 p.m.:

Awesome cameraman! He worked very hard to always be in the lecture theater and delivered us good quality video lectures which will not be possible without the joint effort of team. Thanks so much for the great work, guys!

Comment9 Max Dama, March 2, 2009 at 9:11 p.m.:

Very good

Comment10 goth, July 28, 2009 at 5:53 p.m.:

why only no.2, 3, 4 available??

Comment11 james, August 25, 2009 at 2:31 p.m.:

good course but how can I download these videos?

Comment12 Jack Tanner, September 8, 2009 at 10:34 p.m.:

The slides in the PDF don't match the slides in the lecture. For example, the slide in the video at 32.00 is not in the PDF.

Comment13 Gurusaran, November 1, 2009 at 8:56 p.m.:

That was very Good...
But could any one help me downloading this video???
i.e.How to download this video??

Comment14 Hyoungjin Kim, January 7, 2010 at 12:28 p.m.:

VERY Goooooooooooooooooood !

Comment15 Miroslav Kopecky, February 5, 2010 at 7:41 a.m.:

Very nice presentation, You are very good Nando ! Thanks

Comment16 Josh, March 12, 2010 at 3:51 p.m.:

Inappropriate see-through shirt..
He shouldn't come to class in gym clothing

Comment17 kimo, March 23, 2010 at 2:44 p.m.:

This guy is doing a striptease rather than teaching showing of his body biceps ... ridiculous

Comment18 Robin75, December 14, 2010 at 9:25 p.m.:

I developed a risk analysis tool called Statscorer which allows do to Monte Carlo simulations within Excel and in-depth stochastic modeling, while remaining very simple.

You can download a 15-day evaluation version freely (no personal information required).

You can visit for very detailed examples of how to create stochastic models in Excel.


Comment19 tetris, June 10, 2011 at 2:44 p.m.:

what's with the gun show ?

Comment20 Shubham, August 12, 2011 at 10:38 a.m.:

Is there anyway that i can download .m file (MATLAB) to understand it further? Eg. for Rejection Sampling or SIR etc. etc. used in these lectures?

Comment21 subodh kant dubey, January 17, 2012 at 1:16 p.m.:

Thank's for putting the presentation online. It's great. I am not able to watch it online in a go. So, can you provide the link to download following video lectures

Markov Chain Monte Carlo Methods
author: Christian Robert, Paris Dauphine University

Monte Carlo Simulation for Statistical Inference, Model Selection and Decision Making
author: Nando de Freitas, Department of Computer Science, University of British Columbia

Comment22 Garo, January 30, 2012 at 11:39 p.m.:

I was not expecting that this video lecture are so good.

Thank you.

Comment23 shima, June 22, 2012 at 7:43 p.m.:

Amazing! thank you :)
I wish I had .m codes

Comment24 Shuma-Tuma, February 18, 2014 at 9:47 p.m.:

Fire the camera man. The lecture was great. thanks

Comment25 Sithal, May 10, 2015 at 12:55 p.m.:

Nice Videos on MCMC methods.Can you please make the .m codes available.

Write your own review or comment:

make sure you have javascript enabled or clear this field: