Robust Distributed Learning under Resource Constraints: Decentralized Quantile Estimation via (Asynchronous) ADMM
Anna van Elst, Igor Colin, Stephan Clémençon · 2026
Read Robust Distributed Learning under Resource Constraints: Decentralized Quantile Estimation via (Asynchronous) ADMM by Anna van Elst, Igor Colin, Stephan Clémençon (2026) 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-2601-20571.txt.
Description
Anna van Elst, Igor Colin, and Stephan Clémençon study robust distributed learning on resource-constrained edge devices. Algorithms there must be communication-efficient, robust to corrupted data, and light in memory. Gossip-based methods meet the first need, but robustness is harder to achieve.
The authors turn to asynchronous decentralized ADMM methods for estimating quantiles, starting from the median, which resists corruption better than the mean. Edge networks, from sensors to phones, must learn without a central server and despite bad actors. Robust quantile estimation gives such networks a trustworthy summary of their data.
Catalog record
| CREATOR | Anna van Elst, Igor Colin, Stephan Clémençon |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 86 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-20571