Summary of Fedmse: Federated Learning For Iot Network Intrusion Detection, by Van Tuan Nguyen and Razvan Beuran
FedMSE: Federated learning for IoT network intrusion detection
by Van Tuan Nguyen, Razvan Beuran
First submitted to arxiv on: 18 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 A novel federated learning approach is proposed for improving IoT network intrusion detection, addressing concerns about data availability, computational resources, transfer costs, and privacy preservation. The semi-supervised federated learning model combines Shrink Autoencoder and Centroid one-class classifier (SAE-CEN) to enhance performance by representing normal network data and identifying anomalies in a decentralized strategy. A mean square error-based aggregation algorithm (MSEAvg) is introduced to prioritize more accurate local models, improving global model performance. The approach demonstrates significant improvements in real-world heterogeneous IoT networks, with detection accuracy increasing from 93.98±2.90 to 97.30±0.49, while reducing learning costs and maintaining robustness. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way is found to help detect intrusions on Internet of Things (IoT) networks. This is important because as more devices connect to the internet, there are more ways for hackers to attack. The new approach uses a combination of two existing methods: Shrink Autoencoder and Centroid one-class classifier. This helps the system better understand what normal network data looks like and catch unusual patterns that might be attacks. Another important part is an algorithm that decides which local models to use based on how accurate they are, making the overall model more effective. The results show a big improvement in detecting intrusions, from 93.98% to 97.30%, while also reducing the cost of training and keeping it reliable. |
Keywords
» Artificial intelligence » Autoencoder » Federated learning » Semi supervised
