Book Image

Learning Probabilistic Graphical Models in R

Book Image

Learning Probabilistic Graphical Models in R

Overview of this book

Probabilistic graphical models (PGM, also known as graphical models) are a marriage between probability theory and graph theory. Generally, PGMs use a graph-based representation. Two branches of graphical representations of distributions are commonly used, namely Bayesian networks and Markov networks. R has many packages to implement graphical models. We’ll start by showing you how to transform a classical statistical model into a modern PGM and then look at how to do exact inference in graphical models. Proceeding, we’ll introduce you to many modern R packages that will help you to perform inference on the models. We will then run a Bayesian linear regression and you’ll see the advantage of going probabilistic when you want to do prediction. Next, you’ll master using R packages and implementing its techniques. Finally, you’ll be presented with machine learning applications that have a direct impact in many fields. Here, we’ll cover clustering and the discovery of hidden information in big data, as well as two important methods, PCA and ICA, to reduce the size of big problems.

Related Content you might be interested in

Current Title:

Small book image
Learning Probabilistic Graphical Models in R
Apr, 2016 | 250 pages
No titles found
Table of Contents (15 chapters)
Learning Probabilistic Graphical Models in R
Learning Probabilistic Graphical Models in R
Credits
Credits
About the Author
About the Author
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
Probabilistic Reasoning
Probabilistic Reasoning
Machine learning
Representing uncertainty with probabilities
Probabilistic graphical models
Summary
2
Exact Inference
Exact Inference
Building graphical models
Variable elimination
Sum-product and belief updates
The junction tree algorithm
Examples of probabilistic graphical models
Summary
3
Learning Parameters
Learning Parameters
Introduction
Learning by inference
Maximum likelihood
Learning with hidden variables – the EM algorithm
Principles of the EM algorithm
Summary
4
Bayesian Modeling – Basic Models
Bayesian Modeling – Basic Models
The Naive Bayes model
Beta-Binomial
The Gaussian mixture model
Summary
5
Approximate Inference
Approximate Inference
Sampling from a distribution
Basic sampling algorithms
Rejection sampling
Importance sampling
Markov Chain Monte-Carlo
MCMC for probabilistic graphical models in R
Summary
6
Bayesian Modeling – Linear Models
Bayesian Modeling – Linear Models
Linear regression
Bayesian linear models
Summary
7
Probabilistic Mixture Models
Probabilistic Mixture Models
Mixture models
EM for mixture models
Mixture of Bernoulli
Mixture of experts
Latent Dirichlet Allocation
Summary
Appendix
Appendix
References
Index
Index

Rejection sampling

Suppose we want to sample from a distribution that is not a simple one. Let's call this distribution p(x) and let's assume we can evaluate p(x) for any given value x, up to a normalizing constant Z, that is:

In this context, p(x) is too complex to sample from but we have another simpler distribution q(x) from which we can draw samples. Next, we assume there exists a constant k such that for all values of x. The function kq(x) is the comparison function as shown in the following figure:

The distribution p(x) has been generated with a simple plot:

0.6*dnorm(x,1)+0.4*dnorm(x,5)

The rejection sampling algorithm is based on the following idea:

  • Draw a sample z0 from q(z), the proposal distribution

  • Draw a second u0 sample from a uniform distribution on [0, kq(z0)]
  • If then the sample is rejected otherwise u0 is accepted

In the following figure, the pair (z0, u0) is rejected if it lies in the gray area. The accepted pairs are a uniform distribution under the curve of p(z) and therefore...

Previous Section
End of Section 4
Next Section