LARA: A Light and Anti-overfitting Retraining Approach for Unsupervised Anomaly Detection

Feiyi Chen, Zhen Qin, Mengchu Zhou, Yingying ZHANG, Shuiguang Deng, Lunting Fan, Guansong Pang, Qingsong Wen

Keywords: Anomaly detection, Time series, Light overhead, Anti-overfitting, few-shot learning

TL;DR: We propose a light and anti-overfitting retraining framework to update model, when normal pattern is highly dynamic.

Abstract:

Most of current anomaly detection models assume that the normal pattern remains the same all the time. However, the normal patterns of web services can change dramatically and frequently over time. The model trained on old-distribution data becomes outdated and ineffective after such changes. Retraining the whole model whenever the pattern is changed is computationally expensive.
Further, at the beginning of normal pattern changes, there is not enough observation data from the new distribution. Retraining a large neural network model with limited data is vulnerable to overfitting. Thus, we propose a Light Anti-overfitting Retraining Approach (LARA) based on deep variational auto-encoders for time series anomaly detection.
In LARA we make the following three major contributions: 1) the retraining process is designed as a convex problem and can prevent overfitting as well as converge at a fast rate; 2) a novel ruminate block is introduced, which can leverage the historical data without the need to store them; 3) we mathematically and experimentally prove that when fine-tuning the latent vector and reconstructed data, the linear formations can achieve the least adjusting errors between the ground truths and the fine-tuned ones.
Moreover, we have performed many experiments to verify that retraining LARA with even limited amount of data from new distribution can achieve competitive F1 Score in comparison with the state-of-the-art anomaly detection models trained with sufficient data. Besides, we verify its light computational overhead.

Track: Web Mining and Content Analysis

Submission Guidelines Scope: Yes

Submission Guidelines Blind: Yes

Submission Guidelines Format: Yes

Submission Guidelines Limit: Yes

Submission Guidelines Authorship: Yes

Student Author: Yes

Submission Number: 970