提出DICE方法，量化去中心化学习中节点贡献的级联影响。
DICE: Data Influence Cascade in Decentralized Learning
- 基于数据、拓扑与损失曲率，建模节点间影响级联
- 首次实现去中心化网络中贡献度的可计算估计
- 适用于选合作伙伴和检测恶意行为，适合分布式系统研究者
去中心化学习通过点对点网络将地理分布的计算资源协同起来，应对日益增长的数据与算力需求。然而，缺乏合理的激励机制严重抑制参与意愿。我们提出，公平激励依赖于对参与节点贡献的准确归因，但局部连接导致影响呈现级联传播，带来挑战。为此，我们设计了首个在去中心化环境中估算数据影响级联（DICE）的方法。理论分析表明，影响级联由数据、通信拓扑和损失函数曲率共同决定，并推导出任意邻居跳数下的可计算近似。DICE为选择合适合作者和识别恶意行为等应用奠定基础。项目主页见 https://raiden-zhu.github.io/blog/2025/DICE/。
原文摘要 · Abstract (English)
Decentralized learning offers a promising approach to crowdsource data consumptions and computational workloads across geographically distributed compute interconnected through peer-to-peer networks, accommodating the exponentially increasing demands. However, proper incentives are still in absence, considerably discouraging participation. Our vision is that a fair incentive mechanism relies on fair attribution of contributions to participating nodes, which faces non-trivial challenges arising from the localized connections making influence ``cascade'' in a decentralized network. To overcome this, we design the first method to estimate \textbf{D}ata \textbf{I}nfluence \textbf{C}ascad\textbf{E} (DICE) in a decentralized environment. Theoretically, the framework derives tractable approximations of influence cascade over arbitrary neighbor hops, suggesting the influence cascade is determined by an interplay of data, communication topology, and the curvature of loss landscape. DICE also lays the foundations for applications including selecting suitable collaborators and identifying malicious behaviors. Project page is available at https://raiden-zhu.github.io/blog/2025/DICE/.
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。