Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | SS-8.6 |
| Session: | Innovations in Sampling Theory and Applications |
| Time: | Thursday, May 20, 14:40 - 15:00 |
| Presentation: |
Special Session Lecture |
| Topic: |
Special Sessions: Innovations in Sampling Theory and Applications |
| Title: |
QUANTITATIVE L^2 APPROXIMATION ERROR OF A PROBABILITY DENSITY ESTIMATE GIVEN BY IT SAMPLES |
| Authors: |
Thierry Blu; Swiss Federal Institute of Technology (EPFL) | | |
| | Michael Unser; Swiss Federal Institute of Technology (EPFL) | | |
| Abstract: |
We present a new result characterized by an exact integral expression for the approximation error between a probability density and an integer shift invariant estimate obtained from its samples. Unlike the Parzen window estimate, this estimate avoids recomputing the complete probability density for each new sample: only a few coefficients are required making it practical for real-time applications. We also show how to obtain the exact asymptotic behavior of the approximation error when the number of samples increases and provide the trade-off between the number of samples and the sampling step size. |
| |
| Back | |