(30 minutes to learn)
The Swedsen-Wang algorithm is an MCMC algorithm for sampling from Ising models. It is an auxiliary variable model, where we define a set of "bond" variables which determine which states are coupled, and we alternate between sampling the states and the bond variables. It mixes much faster than Gibbs sampling in models where the variables are tightly coupled.
This concept has the prerequisites:
- Markov random fields (Swedson-Wang is a method for sampling from MRFs.)
- Gibbs sampling (Swedson-Wang is a special case of Gibbs sampling.)
Core resources (read/watch one of the following)
→ Machine Learning: a Probabilistic Perspective
A very comprehensive graudate-level machine learning textbook.
Location: Section 24.5.3, pages 866-868
Supplemental resources (the following are optional, but you may find them useful)
→ Bayesian Reasoning and Machine Learning
A textbook for a graudate machine learning course.
→ Machine learning summer school: Markov chain Monte Carlo (2009)
A video tutorial on MCMC methods.
Location: Part 2, 22:58 to 27:26
-No Additional Notes-
- create concept: shift + click on graph
- change concept title: shift + click on existing concept
- link together concepts: shift + click drag from one concept to another
- remove concept from graph: click on concept then press delete/backspace
- add associated content to concept: click the small circle that appears on the node when hovering over it
- other actions: use the icons in the upper right corner to optimize the graph placement, preview the graph, or download a json representation