Summary of A New Perspective on Time Series Anomaly Detection: Faster Patch-based Broad Learning System, by Pengyu Li et al.
A New Perspective on Time Series Anomaly Detection: Faster Patch-based Broad Learning System
by Pengyu Li, Zhijie Zhong, Tong Zhang, Zhiwen Yu, C.L. Philip Chen, Kaixiang Yang
First submitted to arxiv on: 7 Dec 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 This paper proposes a new approach to time series anomaly detection (TSAD) that leverages the Broad Learning System (BLS), a shallow network framework that excels in optimization speed. The Contrastive Patch-based Broad Learning System (CPatchBLS) combines patching technique and BLS, offering a novel perspective on TSAD. By integrating Simple Kernel Perturbation (SKP) and contrastive learning, CPatchBLS captures differences between normal and abnormal data under various representations. To mitigate temporal semantic loss caused by patching, the authors introduce model-level integration, utilizing BLS’s fast feature extraction to improve detection. Experimental results on five real-world TSAD datasets demonstrate CPatchBLS’s efficacy, outperforming previous deep learning and machine learning methods while maintaining high computing efficiency. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about finding unusual patterns in time series data. It shows that a new approach called Contrastive Patch-based Broad Learning System (CPatchBLS) can be better than using deep learning or machine learning alone. CPatchBLS is like building blocks that fit together to find the unusual parts. It’s fast and works well, even when dealing with lots of data. |
Keywords
» Artificial intelligence » Anomaly detection » Deep learning » Feature extraction » Machine learning » Optimization » Time series
