LSTM故障检测器的无服务器八卦训练：与联邦、本地及集中式学习在NASA C-MAPSS上的匹配协议比较
Serverless gossip training of LSTM failure detectors: A matched-protocol comparison with federated, local and centralized learning on NASA C-MAPSS
浏览论文内容
中文总结 AI 辅助
该研究在NASA C-MAPSS基准上比较了环形八卦、联邦、本地与集中式LSTM故障检测训练，证明八卦在数据异质性适中时是无服务器的可行替代方案，且性能接近联邦学习。
中文摘要 AI 辅助
工业预测性维护越来越依赖于从分布在不同地点的设备中学习，而这些设备的传感器数据难以轻易汇集。联邦平均（FedAvg）通过中央聚合服务器解决此问题；八卦学习则去除了服务器，但其在受控条件下针对循环故障检测模型的行为尚未得到测量。我们针对堆叠LSTM在NASA C-MAPSS涡扇发动机基准上检测即将发生的故障，比较了同步环形八卦与FedAvg、隔离的本地训练以及集中式参考方法。所有方法共享同一开放实现、架构、初始化、优化器、数据划分和训练预算，主要评估指标使用每台测试发动机的一个终端窗口，以避免重叠窗口的统计依赖性。在FD001上（五个随机种子），八卦方法达到了终端窗口F1分数89.6±1.3%，而FedAvg为89.9±1.1%，本地训练为83.6±6.7%，集中式训练为93.5±2.1%，同时在不设协调器的情况下传输与FedAvg相同的负载。节点模型高度一致但并非完全相同（成对决策分歧为1.8%，而无通信时为5.6%）。在FD002-FD004上，节点间通信相比本地训练将终端窗口F1提升了13-28个百分点；八卦方法在FD003和FD004上与FedAvg持平，但在多条件的FD002子集上低4.3个百分点。模拟的消息丢失、节点故障和服务器中断对两种方法均无明显影响，而更大的环形拓扑则使八卦方法退化更快。因此，在数据异质性适中时，环形八卦是一种实用的无服务器替代方案，而随着异质性增长，更快混合的拓扑结构变得重要。
英文摘要
Industrial predictive maintenance increasingly depends on learning from equipment spread across sites whose sensor data cannot easily be pooled. Federated averaging (FedAvg) solves this with a central aggregation server; gossip learning removes the server, but its behaviour for recurrent failure-detection models has not been measured under controlled conditions. We compare synchronous ring gossip with FedAvg, isolated local training and a centralized reference for a stacked LSTM that detects imminent failure on the NASA C-MAPSS turbofan benchmark. All methods share one open implementation, architecture, initialization, optimizer, data split and training budget, and the primary endpoint uses one terminal window per test engine to avoid the statistical dependence of overlapping windows. On FD001 (five seeds), gossip reached a terminal-window F1 of 89.6 +/- 1.3%, compared with 89.9 +/- 1.1% for FedAvg, 83.6 +/- 6.7% for local training and 93.5 +/- 2.1% for centralized training, while transmitting the same payload as FedAvg without a coordinator. Node models agreed closely but not exactly (1.8% pairwise decision disagreement versus 5.6% without communication). Across FD002-FD004, peer communication improved terminal-window F1 over local training by 13-28 points; gossip matched FedAvg on FD003 and FD004 but was 4.3 points lower on the multi-condition FD002 subset. Simulated message loss, node failure and server outage changed neither method appreciably, whereas larger rings degraded gossip faster. Ring gossip is therefore a practical serverless alternative when data heterogeneity is moderate, and faster-mixing topologies become important as heterogeneity grows.
发表机构
- Antalya Bilim University（安塔利亚科学大学）
- Ankara University（安卡拉大学）
- Northwestern University（西北大学）
机构由 AI 辅助整理，请以论文原文为准。
