## Introduction To Bayesian Inference

author: Christopher Bishop, Microsoft Research
published: Nov. 2, 2009,   recorded: August 2009,   views: 35733
Categories
You might be experiencing some problems with Your Video player.

# Slides

0:00 Slides INTRODUCTION TOBAYESIAN INFERENCE –PART 1 Please ask questions Research First Generation Second Generation Third Generation Probability Theory The Rules of Probability Bayes’ Theorem Oranges and Apples Probability Densities Bayesian Inference Why is prior knowledge important? Probabilistic Graphical Models Decomposition Directed Graphs MAAS Graph - 1 Graph - 2 Graph - 3 Graph - 4 Factor Graphs From Directed Graph to Factor Graph Inference on Graphs Factor Trees: Separation Messages: From Factors To Variables Messages: From Variables To Factors What if the graph is not a tree? What if marginalisations are not tractable? Illustration: Bayesian Ranking Two Player Match Outcome Model Two Team Match Outcome Model Multiple Team Match Outcome Model Graph - 5 Graph - 6 Skill Dynamics TrueSkillTM Graph - 7 infer.net Infer.Net demonstration Graph - 8

# 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.

# Watch videos:   (click on thumbnail to launch)

Part 1 1:17:02
!NOW PLAYING

Part 2 1:34:01

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

1 Roger Orobi, November 17, 2009 at 6:58 p.m.:

I liked the slides on binary variable - distribution of placeba/Treated relationship of drugs. The question is: how can it be further studied and quantified to be made a standard measure to treat patients?

2 murugan manoj, January 23, 2010 at 8:08 a.m.:

this is really good 4 me....

3 Ian, September 2, 2010 at 3:51 a.m.:

Ummm.. I must be missing something. If you know how many apples and oranges there are in each box, can't you just calculate the probability based on the relative number of apples and oranges? Aren't you just three times more likely to have gotten the orange from the red box? Why have a supposition at all? If the supposition is 1/5 then the result is .43, i.e., less likely to have come from the red box, which goes against what we would expect given the relative number of oranges in that box!

4 Benny, February 7, 2011 at 12:17 p.m.:

Chris Bishop is a fantastic author and scientist. His book: Pattern Recognition and Machine Learning is the definitive text, but this lecture is, I am sorry to say, disappointing given who is giving it.

I think one should avoid statements like "I'm sure that's all clear, but any questions ?", and saying things like: given the quality of the applicants there is no need to go through proofs of simple stuff like Baye's theorem is there ? It puts people off interacting. Prof Bishop says himself that questions are important, but if you imply that the stuff is easy, with what you say and how you say it, you make people scared of showing themselves up.

Baye's theorem is not easy, simple concepts in probability can trip up the smartest of people and it is through discussions and example that one gains an intuitive insight.

5 Chris, November 3, 2011 at 11:19 p.m.:

While an interesting talk, there are some historical notes that are not quite accurate. A lot more work was done in the 1980's than mentioned here.

But first let me explain we have to take into the account the political and cold war status that existed at the time. The being the existence and proliferation of nuclear weapons. Often in job interviews, one was at that time asked if you had moral objections in working in this field.

Further to that the technology actually was very advanced, in some ways it was perfect for message passing, because there existed the cpu called the transputer. That was a cpu that was a hybrid designed to do just that, pass messages at extremely high speed. With the significant benefit of be able to connect extra cpu's with great ease and allowing very large arrays of them to be build, only further enhancing the capabilities.

So why is this progress not seen today? Simply put there was a phrase that maybe was stuck in our minds more than today's culture, that being a quote from Oppenheimer, "I have become the destroyer of worlds".

With the military extremely interested in this technology applications, a lot of people just left the field altogether.

I posted this because yes, the technology is exciting, yes the potential is great, but the moral issues today are just as valid as the ones in the past, in fact with financial markets using this technology the risk in terms of damage has just shifted, and we can see what that has caused just recently.

So do think about what the consequences of what your work will cause, before you do it, not like Oppenheimer who only saw the consequences after Hiroshima was obliterated.

6 Doug, February 27, 2013 at 9:22 p.m.:

At 5:37 he asks what the probability is that the box is blue, given the fruit is an orange, but at 7:54 he solves the probability that the box is red (answer 2/3).
So to answer the question at 5:37, p(blue|orange) the answer is 1-2/3= 1/3.