Serverless gossip training of LSTM failure detectors: A matched-protocol comparison with federated, local and centralized learning on NASA C-MAPSS

Not provided in the content

federated learninggossip learningpredictive maintenanceLSTM

2609.35792

Builder Relevance

80%

1h ago

Abstract

This paper compares serverless gossip learning with federated averaging and other training methods for LSTM failure detection models on the NASA C-MAPSS dataset.

Reality Card

Core Claim

Gossip learning can achieve comparable performance to federated averaging in failure detection while eliminating the need for a central aggregation server.

Method / Result

Gossip reached a terminal-window F1 of 89.6 +/- 1.3%, closely matching FedAvg's 89.9 +/- 1.1%.

Limitations

The study's results may vary with different data heterogeneity levels and larger rings degrade gossip performance.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.

← Back to all papers