KAN-AD: Time Series Anomaly Detection with Kolmogorov-Arnold Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhou, Quan, Pei, Changhua, Sun, Fei, Han, Jing, Gao, Zhengwei, Pei, Dan, Zhang, Haiming, Xie, Gaogang, Li, Jianhui
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917541080727552
author Zhou, Quan
Pei, Changhua
Sun, Fei
Han, Jing
Gao, Zhengwei
Pei, Dan
Zhang, Haiming
Xie, Gaogang
Li, Jianhui
author_facet Zhou, Quan
Pei, Changhua
Sun, Fei
Han, Jing
Gao, Zhengwei
Pei, Dan
Zhang, Haiming
Xie, Gaogang
Li, Jianhui
contents Time series anomaly detection (TSAD) underpins real-time monitoring in cloud services and web systems, allowing rapid identification of anomalies to prevent costly failures. Most TSAD methods driven by forecasting models tend to overfit by emphasizing minor fluctuations. Our analysis reveals that effective TSAD should focus on modeling "normal" behavior through smooth local patterns. To achieve this, we reformulate time series modeling as approximating the series with smooth univariate functions. The local smoothness of each univariate function ensures that the fitted time series remains resilient against local disturbances. However, a direct KAN implementation proves susceptible to these disturbances due to the inherently localized characteristics of B-spline functions. We thus propose KAN-AD, replacing B-splines with truncated Fourier expansions and introducing a novel lightweight learning mechanism that emphasizes global patterns while staying robust to local disturbances. On four popular TSAD benchmarks, KAN-AD achieves an average 15% improvement in detection accuracy (with peaks exceeding 27%) over state-of-the-art baselines. Remarkably, it requires fewer than 1,000 trainable parameters, resulting in a 50% faster inference speed compared to the original KAN, demonstrating the approach's efficiency and practical viability.
format Preprint
id arxiv_https___arxiv_org_abs_2411_00278
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle KAN-AD: Time Series Anomaly Detection with Kolmogorov-Arnold Networks
Zhou, Quan
Pei, Changhua
Sun, Fei
Han, Jing
Gao, Zhengwei
Pei, Dan
Zhang, Haiming
Xie, Gaogang
Li, Jianhui
Machine Learning
Time series anomaly detection (TSAD) underpins real-time monitoring in cloud services and web systems, allowing rapid identification of anomalies to prevent costly failures. Most TSAD methods driven by forecasting models tend to overfit by emphasizing minor fluctuations. Our analysis reveals that effective TSAD should focus on modeling "normal" behavior through smooth local patterns. To achieve this, we reformulate time series modeling as approximating the series with smooth univariate functions. The local smoothness of each univariate function ensures that the fitted time series remains resilient against local disturbances. However, a direct KAN implementation proves susceptible to these disturbances due to the inherently localized characteristics of B-spline functions. We thus propose KAN-AD, replacing B-splines with truncated Fourier expansions and introducing a novel lightweight learning mechanism that emphasizes global patterns while staying robust to local disturbances. On four popular TSAD benchmarks, KAN-AD achieves an average 15% improvement in detection accuracy (with peaks exceeding 27%) over state-of-the-art baselines. Remarkably, it requires fewer than 1,000 trainable parameters, resulting in a 50% faster inference speed compared to the original KAN, demonstrating the approach's efficiency and practical viability.
title KAN-AD: Time Series Anomaly Detection with Kolmogorov-Arnold Networks
topic Machine Learning
url https://arxiv.org/abs/2411.00278