VACE: Learning Geometrically Structured Representations for Time Series Anomaly Detection

• URL: http://arxiv.org/abs/2605.23504v1
• Date: Fri, 22 May 2026 11:07:09 GMT
• Title: VACE: Learning Geometrically Structured Representations for Time Series Anomaly Detection
• Abstract summary: VACE is a self-supervised anomaly detection method.<n>It represents a compact, directionally coherent region in the embedding space.<n>It significantly outperforms more complex methods trained on substantially larger budgets.
• Score: 0.788278500923222
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Anomaly detection in multivariate time series is a critical task across a wide range of real-world applications, where abnormal behaviour is rare, labels are unavailable, and the cost of a miss is high. The central challenge is learning a characterisation of normality precise enough to flag deviations. Representation self-supervised learning, typically through contrastive approaches, addresses this by embedding temporal patches into a latent space where normality occupies a well-defined region, with anomalies detected by geometric deviation. However, contrastive approaches shape this space indirectly through pair-sampling heuristics, providing no explicit control over the geometric structure that distance-based scoring requires. This means how tightly normal representations are grouped, and whether distances are directionally meaningful. We present VACE (Velocity-Aligned Channel Embeddings), a self-supervised anomaly detection method that represents normality as a compact, directionally coherent region in the embedding space. To this end, VACE trains a channel-aware encoder through a velocity-consistency objective, with no negatives and no synthetic anomalies, so that normal trajectories are locally smooth and aligned. At test time, a Mahalanobis positional score and a velocity-bank directional score are combined multiplicatively, flagging points that are simultaneously off-distribution and dynamically atypical. Despite its simplicity, VACE achieves state-of-the-art performance on TSB-AD-M under rigorous evaluation, significantly outperforming more complex methods trained on substantially larger budgets.

Related papers

• FreqAnchorAD: Language-Free Zero-Shot Anomaly Detection via Frequency-Deviation Anchoring [11.908079655919849]
  We propose FreqAnchorAD, a framework that organizes frequency-enhanced responses for anchor-relative anomaly discrimination.<n>FreqAnchorAD achieves state-of-the-art mean performance in image-level anomaly recognition and pixel-level defect localization.
  arXiv  Detail & Related papers  (2026-08-01T14:48:19Z)
• TPA-AD: A Two-Stage Pseudo Anomaly-Guided Method for Bearing Time-Series Anomaly Detection [15.271061413934332]
  Two-stage pseudo anomaly-guided anomaly detection method is proposed.<n>Method generates pseudo-anomalous windows near the normal boundary.<n>It learns anomaly-sensitive representations through contrastive learning between normal and pseudo-anomalous windows.
  arXiv  Detail & Related papers  (2026-06-02T15:39:10Z)
• SphereVAD: Training-Free Video Anomaly Detection via Geodesic Inference on the Unit Hypersphere [90.85001795656633]
  Video anomaly detection (VAD) aims to automatically identify events that deviate from normal patterns in untrimmed surveillance videos.<n>We propose SphereVAD, a fully training-free, zero-shot VAD framework that recasts anomaly discrimination as von Mises-Fisher (vMF) likelihood-ratio geodesic inference on the unit hypersphere.
  arXiv  Detail & Related papers  (2026-05-08T16:57:38Z)
• Stochastic Voronoi Ensembles for Anomaly Detection [6.508284558109273]
  We propose SVEAD (Stochastic Voronoi Ensembles Anomaly Detector), which constructs ensemble random Voronoi diagrams and scores points by normalized cell-relative distances weighted by local scale.<n> Experiments on 45 datasets demonstrate that SVEAD outperforms 12 state-of-the-art approaches.
  arXiv  Detail & Related papers  (2026-01-07T07:37:38Z)
• CEDL: Centre-Enhanced Discriminative Learning for Anomaly Detection [34.133298397463754]
  Centre-Enhanced Discriminative Learning (CEDL) is a novel supervised anomaly detection framework.<n>It embeds geometric normality directly into the discriminative objective.<n>CEDL achieves competitive and balanced performance across diverse real-world anomaly detection tasks.
  arXiv  Detail & Related papers  (2025-11-15T23:44:18Z)
• Correcting False Alarms from Unseen: Adapting Graph Anomaly Detectors at Test Time [60.341117019125214]
  We propose a lightweight and plug-and-play Test-time adaptation framework for correcting Unseen Normal pattErns in graph anomaly detection (GAD)<n>To address semantic confusion, a graph aligner is employed to align the shifted data to the original one at the graph attribute level.<n>Extensive experiments on 10 real-world datasets demonstrate that TUNE significantly enhances the generalizability of pre-trained GAD models to both synthetic and real unseen normal patterns.
  arXiv  Detail & Related papers  (2025-11-10T12:10:05Z)
• Isolation-based Spherical Ensemble Representations for Anomaly Detection [60.989157958972356]
  Anomaly detection is a critical task in data mining and management with applications spanning fraud detection, network security, and log monitoring.<n>Existing unsupervised anomaly detection methods face fundamental challenges including conflicting distributional assumptions, computational inefficiency, and difficulty handling different anomaly types.<n>We propose ISER (Isolation-based Spherical Ensemble Representations) that extends existing isolation-based methods by using hypersphere radii as proxies for local density characteristics while maintaining linear time and constant space complexity.
  arXiv  Detail & Related papers  (2025-10-15T09:00:05Z)
• Detecting Anomalies in Dynamic Graphs via Memory enhanced Normality [39.476378833827184]
  Anomaly detection in dynamic graphs presents a significant challenge due to the temporal evolution of graph structures and attributes. We introduce a novel spatial- temporal memories-enhanced graph autoencoder (STRIPE) STRIPE significantly outperforms existing methods with 5.8% improvement in AUC scores and 4.62X faster in training time.
  arXiv  Detail & Related papers  (2024-03-14T02:26:10Z)
• Physics-Guided Abnormal Trajectory Gap Detection [2.813613899641924]
  We propose a Space Time-Aware Gap Detection (STAGD) approach to leverage space-time indexing and merging trajectory gaps. We also incorporate a Dynamic Region-based Merge (DRM) approach to efficiently compute gap abnormality scores.
  arXiv  Detail & Related papers  (2024-03-10T17:07:28Z)
• Unsupervised Distance Metric Learning for Anomaly Detection Over Multivariate Time Series [0.0]
  We propose FCM-wDTW, an unsupervised distance metric learning method for anomaly detection over time series. FCM-wDTW encodes raw data into latent space and reveals normal dimension relationships through cluster centers. Experiments with 11 different types of benchmarks demonstrate our method's competitive accuracy and efficiency.
  arXiv  Detail & Related papers  (2024-03-04T09:55:16Z)
• Video Anomaly Detection via Spatio-Temporal Pseudo-Anomaly Generation : A Unified Approach [49.995833831087175]
  This work proposes a novel method for generating generic Video-temporal PAs by inpainting a masked out region of an image. In addition, we present a simple unified framework to detect real-world anomalies under the OCC setting. Our method performs on par with other existing state-of-the-art PAs generation and reconstruction based methods under the OCC setting.
  arXiv  Detail & Related papers  (2023-11-27T13:14:06Z)
• Learning Signed Hyper Surfaces for Oriented Point Cloud Normal Estimation [53.19926259132379]
  We propose a novel method called SHS-Net for oriented normal estimation of point clouds by learning signed hyper surfaces. The signed hyper surfaces are implicitly learned in a high-dimensional feature space where the local and global information is aggregated. An attention-weighted normal prediction module is proposed as a decoder, which takes the local and global latent codes as input to predict oriented normals.
  arXiv  Detail & Related papers  (2023-05-10T03:40:25Z)
• Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy [68.86835407617778]
  Anomaly Transformer achieves state-of-the-art performance on six unsupervised time series anomaly detection benchmarks. Anomaly Transformer achieves state-of-the-art performance on six unsupervised time series anomaly detection benchmarks.
  arXiv  Detail & Related papers  (2021-10-06T10:33:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.