Summary of Adaptive Nad: Online and Self-adaptive Unsupervised Network Anomaly Detector, by Yachao Yuan et al.
Adaptive NAD: Online and Self-adaptive Unsupervised Network Anomaly Detector
by Yachao Yuan, Yu Huang, Yali Yuan, Jin Wang
First submitted to arxiv on: 30 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Signal Processing (eess.SP)
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 This paper addresses the pressing need for Anomaly Detection Systems (ADSs) that can adapt to evolving cyber threats in the Internet of Things (IoT). Current offline unsupervised learning methods are insufficient for real-world applications, as they rely on assumptions about known legitimate data and lack interpretability. The proposed Adaptive NAD framework improves online unsupervised anomaly detection in security domains by generating reliable high-confidence pseudo-labels through an interpretable two-layer strategy. An online learning scheme is also introduced to update Adaptive NAD using a novel threshold calculation technique, enabling it to adapt to new threats. Experimental results demonstrate significant improvements over state-of-the-art solutions on various datasets, including CIC-Darknet2020, CIC-DoHBrw-2020, and Edge-IIoTset. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making computer systems that can detect when something suspicious happens in the internet of things. The problem is that current methods aren’t good enough for real-world use because they rely on known data and don’t explain themselves. The researchers created a new system called Adaptive NAD that uses two steps to find suspicious patterns and then updates itself to keep up with new threats. They tested it on several datasets and showed that it works better than existing solutions. |
Keywords
» Artificial intelligence » Anomaly detection » Online learning » Unsupervised
