From Few-Label to Zero-Label: An Approach for Cross-System Log-Based Anomaly Detection with Meta-Learning

Xinlong Zhao, Tong Jia, Minghua He +3

Log anomaly detection plays a critical role in ensuring the stability and reliability of software systems. However, existing approaches rely on large amounts of labeled log data, w…