一篇新论文介绍了一种名为MUFFLe的方法，旨在降低联邦学习相关的通信成本。MUFFLe通过将广义去重集成到FedAvg流程中，有效地识别和移除重复模式来压缩模型更新。在MNIST数据集上的初步测试表明，与8位量化和Top-k稀疏化等其他压缩技术相比，MUFFLe显著降低了上行通信需求，同时仍能达到目标精度。 AI
影响 这项研究通过降低通信开销，可能促使更高效的联邦学习部署，从而在资源受限的环境中得到更广泛的应用。
排序理由 该集群包含一篇详细介绍联邦学习新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.14354v1 Announce Type: new Abstract: Federated learning is well suited to edge environments but is often limited by the uplink cost of transmitting model updates. This Work-in-Progress paper presents MUFFLe, a communication-efficient update compression scheme that inte…
Federated learning is well suited to edge environments but is often limited by the uplink cost of transmitting model updates. This Work-in-Progress paper presents MUFFLe, a communication-efficient update compression scheme that integrates generalized deduplication (GD) into the F…