Bayesian inversion, spring 2016

Last modified by purisha@helsinki_fi on 2024/03/27 10:26

Bayesian inversion, spring 2016

Nut3web.jpg

Uuzi.gifPlease send the poster file to zenith.purisha@helsinki.fi before 18 May 2016. The posters will be printed and you may take it on Thursday at 9 am in C131. The poster presentation is from 9-11, but let them there to be seen by others until 4 pm.

Teacher: Tapio Helin ja Samuli Siltanen 

Scope: 15 cr

Type: Advanced studies

Teaching: Lectures, exercises, and project work with measured data

Topics: Theory and computational methods of Bayesian inversion

Prerequisites for the theoretical part: Measure and integration theory

Prerequisites for the computational part: Basics of mathematical probability, some Matlab skills such as given by course Applications of matrix computations

The [toc] macro is a standalone macro and it cannot be used inline. Click on this message for details.

Inverse problems are about interpreting indirect measurements. The scientific study of inverse problems is an interdisciplinary field combining mathematics, physics, signal processing, and engineering.  Examples of inverse problems include

  • Three-dimensional X-ray imaging (more information, also see this video and this video)
  • Recovering the inner structure of the Earth based on earthquake measurements 
  • Sharpening a misfocused photograph ( more information   )
  • Reconstructing electric conductivity from current-to-voltage boundary measurements (see  this page  and  this page )
  • Finding cracks inside solid structures
  • Prospecting for oil and minerals
  • Monitoring underground contaminants
  • Finding the shape of asteroids based on light-curve data (see  this page )

The common features of all this problems are the need to understand indirect measurements and to overcome extreme sensitivity to noise and modelling inaccuracies.

The topic of the course is statistical inverse problems. The lectures consist of two parts:

Theoretical modelling in Bayesian inverse problems (Tapio Helin)
Computational methods (prof. Samuli Siltanen) 

 The goals of the course are

(computational part)

- introduce the framework for statistical Bayesian inverse problems

- understand main ideas of uncertainty quantification via the Bayes formula

- learn efficient computational methods for exploring the posterior distribution.
In particular, two central Markov chain Monte Carlo algorithms are studied and implemented in Matlab: Metropolis-Hastings algorithm and the Gibbs sampler.

(theoretical part)

- learn the general Bayes formula

- show well-posedness of Bayesian inverse problems

- study continuity properties of the solution when discretization is refined

News

  •  

Teaching schedule

Period III:  Lectures as follows:

Wednesday 10-12 in room C124 Thursday 10-12 in room B120 Friday 10-12 in room C124.

Two hours of exercise classes per week.

Period IV:  Lectures continue as long as needed (but not for the whole period). There is a project work, which is reported as a poster in a poster session in May. The exact date will be decided later.


Lectures

Wednesday 27.1.2016 Samuli Siltanen
 
Introduction to inverse problems and Bayes formula. Motivating examples: X-ray tomography and Glottal Inverse Filtering (GIF). Practical information about the course.
Introductory lecture slides (PDF)

Thursday 28.1.2016 Samuli Siltanen
 
Real-valued random variable, probability density function, and computational sampling.

Friday 29.1.2016 Tapio Helin
 
Motivation, sigma-algebras, Radon-Nikodym derivative and conditional expectation.

Wednesday 3.2.2016 Tapio Helin
 
Conditional expectation and probability continued, Bayes formula.

Thursday 4.2.2016 Tapio Helin
 
Proof of Bayes formula, some principals of Bayesian inference, example with Gaussian prior and noise.

First version of the lecture notes from the theoretical part: Bayes_theoretical_notes_v1.pdf

Friday 5.2.2016 Samuli Siltanen
 
Conditional probability in the case of probability density functions. Linear measurement model. 

The newest version of the lecture note is here: BayesNotes_v2.pdf

The complete package including Matlab and LaTeX files is here: LectureNotes.zip

Wednesday 10.2.2016 Tapio Helin 
Gaussian posterior, Posterior consistency in over/underdetermined systems

Thursday 11.2.2016 Samuli Siltanen

Introduction of one-dimensional convolution as a measurement model leading to ill-posed inverse problems.
Here are the discrete convolution demonstration slides (with source files in the zip folder): ConvDemo_v1.pdfConvDemo.zip.
Here is the explanation regarding continuous and discrete model: 1D_convolution.pdf

We started building a Matlab library for studying one-dimensional convolution and deconvolution.
DC_convmtx.mtarget1.mPSF.m
deconv1_cont_comp.mdeconv1_cont_plot.m
deconv2_discretedata_comp.mdeconv2_discretedata_plot.m
deconv3_naive_comp.mdeconv3_naive_plot.m 

 

Friday 12.2.2016 Samuli Siltanen

Singular Value Decomposition (SVD) for matrices. Detecting ill-posedness. See the document SVD.pdf.

Computation of SVD for the 1D convolution problem:
deconv4_SVD_comp.mdeconv4_SVD_plot.m

Illustration of Gaussian likelihood, prior and posterior distributions in a 2-dimensional toy example:
GaussianPostDemo.m 

Screen Shot 2016-02-12 at 13.16.25.png

Wednesday 17.2.2016 Tapio Helin
Posterior consistency in underdetermined systems, metrics on probability space

Thursday 18.2.2016 Samuli Siltanen

Definition and discussion of CM and MAP estimates. Derivation of a formula for the MAP estimate in the case of a Gaussian posterior (note that in that case the CM and MAP estimates are the same). 

Introduction of a simple 2-dimensional example. It is a measurement scenario with two temperatures. Indoor temperature f1 is measured with thermometer T1 inside without connection to outdoor temperature. The outdoor temperature f2 is measured with thermometer T2, located in the windowpane. The reading of thermometer T2 is a linear combination of indoor and outdoor temperatures. Both thermometers are corrupted by additive Gaussian noise with standard deviation sigma. The prior is Gaussian. This Matlab file shows the situation: TwoTemps.m

Screen Shot 2016-02-18 at 16.49.35.png

Newest version of the notes: BayesNotes_v3.pdf

Friday 19.2.2016 Samuli Siltanen

Markov chain Monte Carlo sampling using a simple Metropolis-Hastings method.

For explanation of the Metropolis-Hastings method, see Section 5.3.3 of the old lecture note IPnotes14.pdf.

The two-dimensional example of measuring indoor and outdoor temperatures using two thermometers is here in a revised form:
TwoTempsMH1.mposterior.mlogposterior.m

Screen Shot 2016-02-19 at 14.00.41.png

 

Wednesday 24.2. Guest Lecturer: Dr. Marko Laine (Finnish Meteorological Institute)

  The slides of the first part: ML-slides-2016-02-24.pdf
And the MCMC toolbox can be found here: http://helios.fmi.fi/~lainema/mcmc/

Thursday 25.2. Tapio Helin 
Hellinger distance of two Gaussians, well-posedness of the posterior with perturbed measurement and prior

Friday 26.2. Tapio Helin 
Well-posedness of the posterior with perturbed prior continued, introduction to probability in Banach spaces 

Newest version of the notes: Bayes_theoretical_notes_v2.pdf

Wednesday 2.3. Tapio Helin 
Introduction to probability in Banach spaces, Gaussian random variables

Thursday 3.3. Samuli Siltanen

MCMC computation using the Gibbs sampler.

Matlab file: TwoTempsGibbs1.m

Friday 4.3. Tapio Helin
Gaussian random variables, Fernique theorem, Cameron-Martin space

Wednesday 16.3. Samuli Siltanen

Computation of MAP estimates for the 1D deconvolution problem. We use several choices of priors.
Newest version of the lecture notes: BayesNotes_v4.pdf, see the new chapter 2.7.

Matlab routines used in the lecture (some of these are posted again here in the same form than above):


Thursday 17.3. Samuli Siltanen

Computation of CM estimates for the 1D deconvolution problem. We use the Metropolis-Hastings algorithm with several choices of priors.

Matlab routines:


Friday 18.3. Samuli Siltanen

Practicalities about the project work:

  • forming the teams (two students in each team)
  • discussion of the two-phase structure of the project work
  • choosing the topics of the project work, involving real data measurement (tomography, photographic, other)
  • agreeing upon the mid-project deadline
  • setting the date for the final poster session

Wednesday 23.3. Tapio Helin
Cameron-Martin spaces, stability of Bayesian inversion for non-linear problems

The newest version of the lecture note is here: Bayes_theoretical_notes_v3.pdf

Thursday 31.3. Tapio Helin
Stability and approximation properties

Friday 1.4. Tapio Helin 
Bayesian inversion for the inverse heat equation

The final version of the lecture notes: Bayes_theoretical_notes_final.pdf


 

 


Exams

Uuzi.gif Uuzi.gif Uuzi.gif Uuzi.gif Uuzi.gif Uuzi.gif 

HOME EXAM IS HERE (updated version with typos corrected at 13:50 on Monday, April 18)

Deadline for returning answers is 12 o'clock noon on Monday, April 25, 2016.

Course material

Lecture notes will be updated here as the course progresses.

The book Mueller&Siltanen: Linear and nonlinear inverse problems with practical applications (SIAM 2012) explains the computational models used in the course. These lecture notes cover pretty much everything in the computational part: IPnotes14.pdf.

For the theoretical part we partly use the lecture notes by Masoumeh Dashti and Andrew Stuart. Also, an excellent review on Bayesian inverse problems in Banach spaces is available in this paper by Andrew Stuart (only available on UH network).

A slide with preliminaries and a few basics about probability distributions (in particular Gaussian) from the lecture Statistical Inverse Problems at the University of Eastern Finland: 1_Prelim.pdf2_Processes.pdf (We thank Janne Huttunen and Ville Kolehmainen for providing the slides).

Some courses in Inverse Problem in Imaging and one of Bayesian Inversion lectures: Bayesian Inversion by Bangti Jin.

Registration

 
Did you forget to register?   What to do?

Exercises

Teaching assistants: Andreas Hauptmann & Zenith Purisha

Assignments

The assignments are to be prepared for the exercise class, where the students are expected to present the solutions.

Exercise classes

 Tuesday 10-12: 09.02.2016 - classic - C124

 Tuesday 10-12: 16.02.2016 - computer - C128

 Tuesday 10-12: 23.02.2016 - classic - C124

 Tuesday 10-12: 01.03.2016 - computer - C128

 Tuesday 10-12: 15.03.2016 - classic - C124

 Tuesday 10-12: 22.03.2016 - computer - C128

 Tuesday 10-12: 05.04.2016 - computer - C128 (Last exercise session: we go through computational measurement models needed in the project works.)


Project work

Project work assistants: Andreas Hauptmann and Zenith Purisha.

The idea is to study a Bayesian inverse problem both theoretically and computationally in teams of two students. The end product is a scientific poster that the team will present in a poster session on May 19 (at 9-11, Exactum first floor corridor).  The poster can be printed using the large-scale printer of the Industrial Mathematics Laboratory

 

The classical table of contents is recommended for structuring the poster:

 

1 Introduction
2 Materials and methods
3 Results
4 Discussion

 

Section 2 is for describing the data and the inversion methods used. In section 3 those methods are applied to the data and the results are reported with no interpretation; just facts and outcomes of computations are described. Section 4 is the place for discussing the results and drawing conclusions.

The recommended measurement context of the project is X-ray tomography. You can measure a dataset yourself in the X-ray facility of the Industrial Mathematics Laboratory:

IMG_1313.jpg   IMG_1345.jpg  IMG_1362.jpg  Rose03B.jpg

The project work has two phases, each with a specific goal.

First goal (deadline April 15) consists of two things: (a) two first sections should be preliminary written in LaTeX (not necessarily in poster format yet) and (b) the Matlab codes related to the measurement should be run and studied. Two things will be graded in the meeting about the first goal: (a) the draft of project work and (b) your understanding of the available Matlab codes relevant to your topic. The grade represents 30% of the final grade of the project work. Please agree on a meeting time with the lecturer for reviewing and grading the first goal.

Second and final goal (deadline May 19): poster is presented in the poster session. The poster will be printed in size A1. You may create your own poster (from scratch), or you can use e.g. this template as a starting point and edit its layout, colors, fonts, etc. as much as you like.

Example posters are shown on this page.


Course feedback

Course feedback can be given at any point during the course. Click here.