Adaptive Weighting Push-SUM for Decentralized Optimization with Statistical Diversity
Yiming Zhou, Yifei Cheng, Linli Xu, Enhong Chen* · 2024
Read Adaptive Weighting Push-SUM for Decentralized Optimization with Statistical Diversity by Yiming Zhou, Yifei Cheng, Linli Xu, Enhong Chen* (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-07252.txt.
Description
Networks whose members hold unlike data can still agree on a shared answer, Yiming Zhou and co-authors show in a 2024 paper. Decentralized training lets devices learn together without a central server. The classic Push-SUM gossip protocol falters when data differ across members.
Their adaptive weighting version contains the old protocol as a special case. With enough chatter, the agreement gap shrinks far faster than the old bound. Two standard training methods on the new protocol reach good answers at rates the old one cannot match when data are skewed. Smarter gossip helps sensor fleets and phone swarms learn jointly.
Catalog record
| CREATOR | Yiming Zhou, Yifei Cheng, Linli Xu, Enhong Chen* |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 106 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-07252