Decentralized Parameter-Free Online Learning with Compressed Gossip
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913166384955392 |
|---|---|
| author | Ortega, Tomas Jafarkhani, Hamid |
| author_facet | Ortega, Tomas Jafarkhani, Hamid |
| contents | We study decentralized online convex optimization when agents communicate over a graph and messages may be compressed. Classical decentralized online methods typically require learning-rate choices that depend on the horizon, comparator scale, or other problem parameters, while compressed communication introduces additional disagreement that must be controlled. We propose DECO-EF (DEcentralized COin-betting with Error Feedback), a decentralized parameter-free online learning algorithm that combines coin-betting predictions with compressed difference-based gossip. Each agent maintains a clean accumulated state and a compressed tracker, and communicates only compressed state differences during gossip steps. The method is parameter-free in the online-learning sense: it does not tune to the horizon, the comparator norm, or the learning rate. We prove expected comparator-adaptive network-regret bounds for DECO-EF under compressed communication. To the best of our knowledge, this gives the first expected sublinear network-regret guarantees for parameter-free decentralized online learning under compressed communication. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_27831 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Decentralized Parameter-Free Online Learning with Compressed Gossip Ortega, Tomas Jafarkhani, Hamid Machine Learning Signal Processing Optimization and Control 68W10, 68W15, 68W40, 90C06, 90C35, 90C26 G.1.6; F.2.1; E.4 We study decentralized online convex optimization when agents communicate over a graph and messages may be compressed. Classical decentralized online methods typically require learning-rate choices that depend on the horizon, comparator scale, or other problem parameters, while compressed communication introduces additional disagreement that must be controlled. We propose DECO-EF (DEcentralized COin-betting with Error Feedback), a decentralized parameter-free online learning algorithm that combines coin-betting predictions with compressed difference-based gossip. Each agent maintains a clean accumulated state and a compressed tracker, and communicates only compressed state differences during gossip steps. The method is parameter-free in the online-learning sense: it does not tune to the horizon, the comparator norm, or the learning rate. We prove expected comparator-adaptive network-regret bounds for DECO-EF under compressed communication. To the best of our knowledge, this gives the first expected sublinear network-regret guarantees for parameter-free decentralized online learning under compressed communication. |
| title | Decentralized Parameter-Free Online Learning with Compressed Gossip |
| topic | Machine Learning Signal Processing Optimization and Control 68W10, 68W15, 68W40, 90C06, 90C35, 90C26 G.1.6; F.2.1; E.4 |
| url | https://arxiv.org/abs/2605.27831 |