Phelma Formation 2022

Bayesian methods for data image analysis (SIGMA S9) - WPMTBMD7

  • Number of hours

    • Lectures 13.5
    • Projects 0
    • Tutorials 1.5
    • Internship 0
    • Laboratory works 4.0

    ECTS

    ECTS 2.0

Goal(s)

The aim is to introduce fundamentals on Bayesian inference, and to develop applications in the framework of image and signal processing.

Content(s)

  • Introduction
  • Bayesian estimators
  • Priori choice
  • Approximate Bayesian inference
    • Deterministic approximation methods
    • Stochastic approximation methods
  • Case study: Bayesian inference for speech recognition


Prerequisites

Basic notion in both estimation and detection theory

Test

Semester 9 - The exam is given in english only 

Lab work + Written exam (First session)
+ Written or Oral exam for the second session



Contrôle continu *30% + DS 70%

Additional Information

Semester 9 - This course is given in english only EN

Curriculum->Double-Diploma Engineer/Master->Semester 9
Curriculum->Master->Semester 9

Bibliography

[1] Robert, C. (2007). The Bayesian choice: from decision-theoretic foundations to computational implementation. Springer Science & Business Media.
[2] Šmídl, V., & Quinn, A. (2006). The variational Bayes method in signal processing. Springer Science & Business Media.
[3] Gilks, W. R. (2005). Markov chain monte carlo. John Wiley & Sons, Ltd.
[4] Hjort, N. L., Holmes, C., Müller, P., & Walker, S. G. (Eds.). (2010). Bayesian nonparametrics (Vol. 28). Cambridge University Press.