Summary of Detection Of Global Anomalies on Distributed Iot Edges with Device-to-device Communication, by Hideya Ochiai and Riku Nishihata and Eisuke Tomiyama and Yuwei Sun and Hiroshi Esaki
Detection of Global Anomalies on Distributed IoT Edges with Device-to-Device Communication
by Hideya Ochiai, Riku Nishihata, Eisuke Tomiyama, Yuwei Sun, Hiroshi Esaki
First submitted to arxiv on: 16 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Distributed, Parallel, and Cluster Computing (cs.DC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a fully distributed collaborative scheme, Wireless Ad Hoc Federated Learning (WAFL-Autoencoder), to detect anomalies in IoT applications. This approach enables multiple devices installed at a remote site to work together and identify rare events (anomalies) from their local observations. The authors introduce the concept of Global Anomaly detection, which considers samples that are not only rare locally but also globally across all devices. A distributed threshold-finding algorithm is proposed for efficient detection. Evaluation on standard benchmarks confirms the effectiveness of WAFL-Autoencoder in training anomaly detectors and finding thresholds with low false positive rates while achieving high true positive rates. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine you have many devices connected to each other, like sensors or cameras, that can detect unusual events happening at the same time. This is called anomaly detection. The problem is that these devices are usually far from each other and need to communicate wirelessly to share information. In this paper, researchers propose a new way for devices to work together to find rare events. They call it Wireless Ad Hoc Federated Learning (WAFL-Autoencoder). It helps devices identify unusual events and set a threshold for what counts as an anomaly. The authors tested their method on standard benchmarks and found that it works well. |
Keywords
» Artificial intelligence » Anomaly detection » Autoencoder » Federated learning
