Workshop on Approximate Inference in Stochastic Processes and Dynamical Systems, Cumberland Lodge 2008

Approximate system identification: Misfit versus latency

author: Ivan Markovsky, School of Electronics and Computer Science, University of Southampton
published: Aug. 5, 2008, recorded: May 2008, views: 446

Slides

Slides
0:00	Approximate system identification: Misfit versus latency
0:47	Example
2:44	Linear or nonlinear, deterministic or stochastic?
5:35	In this talk . . .
6:35	Least squares ↔ Latency
7:54	Total least squares ↔ Misfit
9:46	Geometric interpretation of latency
10:09	Geometric interpretation of latency (2)
10:46	Geometric interpretation of misfit
11:02	Geometric interpretation of misfit (2)
11:15	Notes
12:33	Regression ↔ Latency
13:09	Errors-in-variables regression ↔ Misfit
13:45	System identification: wd 7→ b B ∈M
14:59	Stochastic estimation vs deterministic approximation
16:24	Misfit vs latency
18:49	Statistical interpretation of misfit and latency
19:58	Identification problems
22:33	Conclusions (1)
23:14	Conclusions (2)
23:27	Conclusions (3)
23:33	Conclusions (2)
23:36	Conclusions (3)
23:48	Thank you

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.

Lecture popularity: You need to login to cast your vote.

Description

Two fundamentally different approaches in system identification, which are used for quantification of the model–data mismatch, are misfit and latency. The aim of this talk is to explain the rationale behind them and link them to statistical estimation methods—errors-in-variables regression and classical regression—respectively.