Abstract
In this letter, we consider a class of generalized multivariate Huber (GMH) loss functions. Our goal is parameter estimation in linear models contaminated by non-Gaussian noise. We assume access to a secondary dataset of independent noise realizations, and we use these data to fit a convex GMH function that will then lead to efficient parameter estimation. Our framework includes the classical weighted least squares and Huber's function as special cases. We demonstrate its advantages in heavy-tailed noise distributions.
| Original language | English |
|---|---|
| Pages (from-to) | 1647-1651 |
| Number of pages | 5 |
| Journal | IEEE Signal Processing Letters |
| Volume | 23 |
| Issue number | 11 |
| DOIs | |
| State | Published - Nov 2016 |
Keywords
- Maximum likelihood estimation
- multidimensional signal processing
- parameter estimation
- signal detection
All Science Journal Classification (ASJC) codes
- Signal Processing
- Electrical and Electronic Engineering
- Applied Mathematics
Fingerprint
Dive into the research topics of 'Fitting Generalized Multivariate Huber Loss Functions'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver