Learning Over Multitask Graphs—Part I: Stability Analysis

Learning Over Multitask Graphs—Part I: Stability Analysis

This paper formulates a multitask optimization problem where agents in the network have individual objectives to meet, or individual parameter vectors to estimate, subject to a smoothness condition over the graph. The smoothness condition softens the transition in the tasks among adjacent nodes and...

Full description

Bibliographic Details
Main Authors: Roula Nassif, Stefan Vlaski, Cedric Richard, Ali H. Sayed
Format: Article
Language:English
Published: IEEE 2020-01-01
Series:IEEE Open Journal of Signal Processing
Subjects:
Multitask distributed inference
diffusion strategy
smoothness prior
graph Laplacian regularization
gradient noise
stability analysis
Online Access:https://ieeexplore.ieee.org/document/9075197/

Internet

https://ieeexplore.ieee.org/document/9075197/

Similar Items