Diffusion LMS for multitask problems with overlapping hypothesis subspaces

There are many important applications that are multitask-oriented in the sense that there are multiple optimum parameter vectors to be inferred simultaneously by networked agents. In this paper, we formulate an online multitask learning problem where node hypothesis spaces partly overlap. A cooperat...

Full description

Saved in:
Bibliographic Details
Main Authors: Jie Chen, Richard, Cedric, Hero, Alfred O., Sayed, Ali H.
Format: Conference Proceeding
Language:English
Subjects:
Adaptive systems
collaborative processing
Convergence
diffusion strategy
distributed optimization
Estimation
Least squares approximations
Multitask learning
Optimization
Signal processing algorithms
Vectors
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:There are many important applications that are multitask-oriented in the sense that there are multiple optimum parameter vectors to be inferred simultaneously by networked agents. In this paper, we formulate an online multitask learning problem where node hypothesis spaces partly overlap. A cooperative algorithm based on diffusion adaptation is derived. Some results on its stability and convergence properties are also provided. Simulations are conducted to illustrate the theoretical results.
ISSN:1551-2541
2378-928X
DOI:10.1109/MLSP.2014.6958929