LogAnMeta: Log Anomaly Detection Using Meta Learning

TL;DR

LogAnMeta introduces a meta-learning approach for effective log anomaly detection in telecom systems, especially when only limited anomaly samples are available, outperforming traditional supervised methods.

Contribution

This work presents a novel meta-learning framework for log anomaly detection that handles few-shot scenarios, improving detection of unseen anomalies.

Findings

Effective in few-shot anomaly detection

Outperforms traditional supervised methods

Demonstrates robustness on telecom log datasets

Abstract

Modern telecom systems are monitored with performance and system logs from multiple application layers and components. Detecting anomalous events from these logs is key to identify security breaches, resource over-utilization, critical/fatal errors, etc. Current supervised log anomaly detection frameworks tend to perform poorly on new types or signatures of anomalies with few or unseen samples in the training data. In this work, we propose a meta-learning-based log anomaly detection framework (LogAnMeta) for detecting anomalies from sequence of log events with few samples. LoganMeta train a hybrid few-shot classifier in an episodic manner. The experimental results demonstrate the efficacy of our proposed method