Time series anomaly detection has been received growing interest in industrial and academic communities due to its substantial theoretical value and practical significance in reality. Recent advanced methods for time series anomaly detection are based on deep learning techniques, since they have shown their superiority in some specific situations. However, most existing deep learning-based anomaly detection methods require predefined, specific tasks of reconstruction or prediction, necessitating task-specific loss functions. Designing such anomaly-aware loss functions poses a significant challenge due to the ambiguity in defining ground-truth anomalies. Moreover, these methods often rely on complex network architectures that tend to lead to over-generalization, resulting in even abnormal data being well reconstructed or fitted. To mitigate this situation, grounded in activation learning theory, we propose a novel unsupervised time series anomaly detection paradigm termed ALAD. ALAD utilizes a straightforward fully connected network architecture, measuring the typicality of input patterns through the sum of the squared output. Despite its simplicity, ALAD achieves competitive performance compared to state-of-the-art models trained using backpropagation. By utilizing various real-world and synthetic datasets, experimental results have confirmed the effectiveness and feasibility of the proposed paradigm. This work also demonstrates that biologically-plausible local learning can sometimes outperform backpropagation in real-world scenarios.

Anomaly detection is an important data-mining task closely related to specific applications. In recent years, spatio-temporal data deriving from various networks has been collected from many real-world scenarios, but how to detect anomalies in such data is still an open problem. In this article, we propose a novel anomaly detection framework, Multi-scale Fuzzy Contrastive Anomaly Detection (MFCAD), by capturing anomalous patterns of data from multiple spatio-temporal scales so as to learn distinguishable feature representations. Different from conventional reconstruction- or prediction-based anomaly detection methods, this approach is not concerned with the consistency of the encoded representation of the implicit layer and the discriminability of the implicit layer of the anomalies. MFCAD implements a more discriminative representation using fuzzy contrastive learning and explicitly performs anomaly detection in the potential space by measuring the distance between outliers and implicit values. This proposed method has been applied for the anomaly detection of 5G mobile network clusters (5G-MNCs) in China Mobile. Furthermore, in order to validate the generalizability of the proposed method, it is further tested on public datasets and experiment results show the promising performance.