Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
410422 | Neurocomputing | 2013 | 15 Pages |
Abstract
In this work, we analyze the generalization ability of distributed online learning algorithms under stationary and non-stationary environments. We derive bounds for the excess-risk attained by each node in a connected network of learners and study the performance advantage that diffusion has over individual non-cooperative processing. We conduct extensive simulations to illustrate the results.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Artificial Intelligence
Authors
Zaid J. Towfic, Jianshu Chen, Ali H. Sayed,