1

I'm interested in learning approximate message passing from the paper "Message Passing Algorithms for Compressed Sensing: I. Motivation and Construction". My background is in computer science and engineering and I have never taken a course on measure theory.

I've noticed that the paper mentioned above requires knowledge in measure theory. So I searched for resources and I found this https://people.eecs.berkeley.edu/~wainwrig/Papers/WaiJor08_FTML.pdf which I still find it based on pure mathematics.

Can anyone recommend a readable introduction or papers that are prerequisites to the sources above?

William
  • 97
  • 8

1 Answers1

2

I think that Structured Belief Propagation for NLP (especially starting from Section 2) is a wonderful resource that could fit your background.

The slides offer plenty of illustrations and elementary examples. They are very close to what you would actually program.

TheCG
  • 822
  • 4
  • 13
  • It's really good reference you provided. But it still doesn't explain some lemmas in the reference "Message Passing Algorithms for Compressed Sensing: I. Motivation and Construction". – William Nov 01 '20 at 10:27