Program

PROGRAM OF THE 3rd SPRING SCHOOL on

Data-Driven Model Learning of Dynamic Systems

 

 

GENERAL INFORMATION

The time schedule of the Spring School is available here

For the computer exercices,  participants  should bring their own laptop with one of the latest versions of Matlab (version R2014a at least)  installed with stand alone license. The Matlab System Identification Toolbox must be available.

 

PROGRAM AT A GLANCE

MONDAY 1 APRIL (14:00) - WEDNESDAY 3 APRIL (12:00)

Lecturer: Xavier Bombois, CNRS Research Director, Laboratoire Ampère, Ecole Centrale de Lyon 

A two-day course on linear system identification

 

Theme 1: Introduction;concepts; identification cycle

Theme 2: Parametric (prediction error) identification methods: prediction criterion and model structures, linear and pseudo-linear regressions, conditions on data, statistical and asymptotic properties, model set selection and model validation

Theme 3: Non-parametric identification (ETFE)

Theme 4: Experiment design.

Exercises : getting hands on the different concepts using exercises on paper and computer exercises (Matlab System Identification toolbox).

 

WEDNESDAY 3 APRIL (afternoon)

Lecturer: Laurent Bako, Associate Professor, Laboratoire Ampère, Ecole Centrale de Lyon

An half-day introductory course on hybrid system identification

 

Theme 1: From sparsity-inducing optimization to robust regression

Theme 2: Application to hybrid system identification

 

THURSDAY 4 APRIL (all day)

Lecturer: Håkan Hjalmarsson, Professor, KTH, Stockholm, Sweden

A one-day course on dynamic model learning

 

Theme 1: Fundamental parameter estimation concepts: Sufficient statistics, the Cramér-Rao bound, the maximum likelihood estimator, estimator-based methods

Theme 2: Minimum MSE estimators: Bayes estimators, empirical Bayes methods, risk estimation methods, Gaussian processes, asymptotic analysis

Theme 3: Application to dynamical models: linear models, non-linear input-output models, non-linear state-space models

Theme 4: Computational tools: Sampling, Markov Chain Monte Carlo methods, particle filtering and smoothing

 

 

 

e
Online user: 1