mlsanityData to 7 Oct 2026

Graph Mixture of Experts and Memory-augmented Routers for Multivariate Time Series Anomaly Detection

arXiv 2412.19108 ↗First posted 26 Dec 2024cs.LG cs.AIcited by 19

On the map: Mixture-of-experts → MoE for time series and graphs

Abstract (arXiv)

Multivariate time series (MTS) anomaly detection is a critical task that involves identifying abnormal patterns or events in data that consist of multiple interrelated time series. In order to better model the complex interdependence between entities and the various inherent characteristics of each entity, the GNN based methods are widely adopted by existing methods. In each layer of GNN, node features aggregate information from their neighboring nodes to update their information. In doing so, from shallow layer to deep layer in GNN, original individual node features continue to be weakened and more structural information,i.e., from short-distance neighborhood to long-distance neighborhood, continues to be enhanced. However, research to date has largely ignored the understanding of how hierarchical graph information is represented and their characteristics that can benefit anomaly detection. Existing methods simply leverage the output from the last layer of GNN for anomaly estimation while neglecting the essential information contained in the intermediate GNN layers. To address such limitations, in this paper, we propose a Graph Mixture of Experts (Graph-MoE) network for multivariate time series anomaly detection, which incorporates the mixture of experts (MoE) module to adaptively represent and integrate hierarchical multi-layer graph information into entity representations. It is worth noting that our Graph-MoE can be integrated into any GNN-based MTS anomaly detection method in a plug-and-play manner. In addition, the memory-augmented routers are proposed in this paper to capture the correlation temporal information in terms of the global historical features of MTS to adaptively weigh the obtained entity representations to achieve successful anomaly estimation. Extensive experiments on five challenging datasets prove the superiority of our approach and each proposed module.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Placed in MoE for time series and graphs, one of 12 directions in Mixture-of-experts. Builds on 27 of its 52 references that are in the map. First posted on arXiv 26 Dec 2024.

Builds on

27 of 52 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Graph Mixture of Experts: Learning on Large-Scale Graphs with Explicit Diversity Modeling6 Apr

Other references · 26

  1. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  2. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  3. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,601
  4. 2018Representation Learning on Graphs with Jumping Knowledge Networkscited by 875
  5. 2019Graph WaveNet for Deep Spatial-Temporal Graph Modelingcited by 820
  6. 2020Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networkscited by 563
  7. 2018Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholdingcited by 407
  8. 2020Generating Radiology Reports via Memory-driven Transformercited by 363
  9. 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 320
  10. 2021BASE Layers: Simplifying Training of Large, Sparse Modelscited by 315
  11. 2021Graph Neural Network-Based Anomaly Detection in Multivariate Time Seriescited by 300
  12. 2019Deep Semi-Supervised Anomaly Detectioncited by 290
14 more
  1. 2020Multivariate Time-series Anomaly Detection via Graph Attention Networkcited by 165
  2. 2019Time-Series Anomaly Detection Service at Microsoftcited by 163
  3. 2020Multivariate Probabilistic Time Series Forecasting via Conditioned Normalizing Flowscited by 142
  4. 2020DROCC: Deep Robust One-Class Classificationcited by 121
  5. 2020Structural Temporal Graph Neural Networks for Anomaly Detection in Dynamic Graphscited by 55
  6. 2019Memory Augmented Graph Neural Networks for Sequential Recommendationcited by 46
  7. 2023ADA-GAD: Anomaly-Denoised Autoencoders for Graph Anomaly Detectioncited by 23
  8. 2022Detecting Multivariate Time Series Anomalies with Zero Known Labelcited by 21
  9. 2021Dynamic Memory based Attention Network for Sequential Recommendationcited by 16
  10. 2022A Tale of HodgeRank and Spectral Method: Target Attack Against Rank Aggregation Is the Fixed Point of Adversarial Gamecited by 16
  11. 2021Poisoning Attack against Estimating from Pairwise Comparisonscited by 14
  12. 2021Neural Relational Inference with Efficient Message Passing Mechanismscited by 10
  13. 2024Sequential Manipulation Against Rank Aggregation: Theory and Algorithmcited by 7
  14. 2024Heterogeneous Interaction Modeling With Reduced Accumulated Error for Multi-Agent Trajectory Predictioncited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 8; 2026: 11.

19 papers.

Sort:

By area:

  1. 2025Can Multimodal LLMs Perform Time Series Anomaly Detection?cited by 11
  2. 2025Learning Soft Sparse Shapes for Efficient Time-Series Classificationcited by 10
  3. 2025EmoVerse: A MLLMs-Driven Emotion Representation Dataset for Interpretable Visual Emotion Analysiscited by 6
  4. 2025Mixture-of-Experts for Personalized and Semantic-Aware Next Location Predictioncited by 2
  5. 2025Variational Mixture of Graph Neural Experts for Alzheimer's Disease Recognition across Frequency Bands in EEG Brain Networkscited by 2
  6. 2026Towards Accurate Emotion-Attributed Video Captioning via Fine-grained Emotion-Cause Pair Extractioncited by 2
  7. 2026Audio-Visual Exchange-Aware Token Pruning for Efficient Audio-Visual Captioningcited by 2
  8. 2026EmoStyle: Affective Conditioning of Style-Specialist Experts for Emotional Image Generationcited by 2
  9. 2025Cluster-Aware Causal Mixer for Online Anomaly Detection in Multivariate Time Seriescited by 1
  10. 2026Modality as Heterogeneity: Node Splitting and Graph Rewiring for Multimodal Graph Learningcited by 1
  11. 2025Dynamic Adaptive Parsing of Temporal and Cross-Variable Patterns for Network State Classificationcited by 0
  12. 2025Explainable Heterogeneous Anomaly Detection in Financial Networks via Adaptive Expert Routingcited by 0
7 more
  1. 2026Routing-Aware Explanations for Mixture of Experts Graph Models in Malware Detectioncited by 0
  2. 2026Affinity Is Not Enough: Recovering the Free Energy Principle in Mixture-of-Expertscited by 0
  3. 2026Geometry-aware Gaussian Prior and Axial Attention for Cervical Cytology Image Classificationcited by 0
  4. 2026Rationale-Guided Knowledge Distillation for Cross-Lingual Stance Detectioncited by 0
  5. 2026DSPO: Diversity-aware Subjective Policy Optimization for Robust Emotional Reasoningcited by 0
  6. 2026Causal-EVC: Breaking Emotional Spurious Causality via Spatiotemporal Grounding and Counterfactual Interventioncited by 0
  7. 2026Decoding Affective Nuances: Enhancing MLLMs via Hierarchical Emotion Reasoning and Contrastive Discriminative Pruningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023MIM-GAN-based Anomaly Detection for Multivariate Time Series Data
  2. 2026Evolutionary Router Feature Generation for Zero-Shot Graph Anomaly Detection with Mixture-of-Experts
  3. 2026Graph-Spectral Flow Matching for Multivariate Time Series Anomaly Detection
  4. 2025Task-Aware Mixture-of-Experts for Time Series Analysis
  5. 2025Time Tracker: Mixture-of-Experts-Enhanced Foundation Time Series Forecasting Model with Decoupled Training Pipelines
  6. 2024Counterfactual Data Augmentation with Denoising Diffusion for Graph Anomaly Detection
  7. 2024Interpretable mixture of experts for time series prediction under recurrent and non-recurrent conditions
  8. 2024Mixture of Experts for Node Classification