提出无需调参的去中心化在线学习算法，支持压缩通信
Decentralized Parameter-Free Online Learning with Compressed Gossip
- 用硬币投注法结合压缩差值通信，实现无参数自适应
- 首次在压缩通信下证明期望次线性网络后悔界
- 适合分布式系统中资源受限的在线学习场景
我们研究了在图结构通信且消息可压缩的条件下，去中心化在线凸优化问题。传统去中心化在线方法通常需要依赖时间跨度、比较器尺度或其他问题参数设置学习率，而压缩通信会引入额外偏差。本文提出 DECO-EF（DEcentralized COin-betting with Error Feedback），一种结合硬币投注预测与压缩差值型共识的去中心化无参数在线学习算法。每个代理维护一个干净的累积状态和一个压缩追踪器，仅在共识步骤中传输压缩的状态差值。该方法在在线学习意义上完全无参数：不需针对时间跨度、比较器范数或学习率进行调优。我们在压缩通信下证明了 DECO-EF 的期望比较器自适应网络后悔界。据我们所知，这是首个在压缩通信下对无参数去中心化在线学习给出期望次线性网络后悔保证的工作。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。