Analyzing Shapley Additive Explanations to Understand Anomaly Detection Algorithm Behaviors and Their Complementarity
- URL: http://arxiv.org/abs/2602.00208v1
- Date: Fri, 30 Jan 2026 13:45:43 GMT
- Title: Analyzing Shapley Additive Explanations to Understand Anomaly Detection Algorithm Behaviors and Their Complementarity
- Authors: Jordan Levy, Paul Saves, Moncef Garouani, Nicolas Verstaevel, Benoit Gaudou,
- Abstract summary: We propose a methodology for characterizing anomaly detectors through their decision mechanisms.<n>We show that detectors with similar explanations tend to produce correlated anomaly scores and identify largely overlapping anomalies.<n>We also demonstrate that diversity alone is insufficient; high individual model performance remains a prerequisite for effective ensembles.
- Score: 0.4077787659104315
- License: http://creativecommons.org/licenses/by-nc-nd/4.0/
- Abstract: Unsupervised anomaly detection is a challenging problem due to the diversity of data distributions and the lack of labels. Ensemble methods are often adopted to mitigate these challenges by combining multiple detectors, which can reduce individual biases and increase robustness. Yet building an ensemble that is genuinely complementary remains challenging, since many detectors rely on similar decision cues and end up producing redundant anomaly scores. As a result, the potential of ensemble learning is often limited by the difficulty of identifying models that truly capture different types of irregularities. To address this, we propose a methodology for characterizing anomaly detectors through their decision mechanisms. Using SHapley Additive exPlanations, we quantify how each model attributes importance to input features, and we use these attribution profiles to measure similarity between detectors. We show that detectors with similar explanations tend to produce correlated anomaly scores and identify largely overlapping anomalies. Conversely, explanation divergence reliably indicates complementary detection behavior. Our results demonstrate that explanation-driven metrics offer a different criterion than raw outputs for selecting models in an ensemble. However, we also demonstrate that diversity alone is insufficient; high individual model performance remains a prerequisite for effective ensembles. By explicitly targeting explanation diversity while maintaining model quality, we are able to construct ensembles that are more diverse, more complementary, and ultimately more effective for unsupervised anomaly detection.
Related papers
- SetAD: Semi-Supervised Anomaly Learning in Contextual Sets [25.628827917857603]
Semi-supervised anomaly detection has shown great promise by effectively leveraging limited labeled data.<n>We propose SetAD, a novel framework that reframes semi-supervised AD as a Set-level Anomaly Detection task.<n>To enhance robustness and score calibration, we propose a context-calibrated anomaly scoring mechanism.
arXiv Detail & Related papers (2025-11-26T13:27:59Z) - Contrastive Representation Modeling for Anomaly Detection [0.21427777919040417]
We propose a structured contrastive objective that redefines positive and negative relationships during training, promoting these properties without requiring explicit anomaly labels.<n>We extend this framework with a patch-based learning and evaluation strategy specifically designed to improve the detection of localized anomalies in industrial settings.
arXiv Detail & Related papers (2025-01-09T10:33:16Z) - Error Diversity Matters: An Error-Resistant Ensemble Method for Unsupervised Dependency Parsing [49.60415088899208]
We propose an efficient ensemble-selection approach that considers error diversity and avoids error accumulation.<n>Results demonstrate that our approach outperforms each individual model as well as previous ensemble techniques.
arXiv Detail & Related papers (2024-12-16T08:23:50Z) - Towards a Unified Framework of Clustering-based Anomaly Detection [18.30208347233284]
Unsupervised Anomaly Detection (UAD) plays a crucial role in identifying abnormal patterns within data without labeled examples.
We propose a novel probabilistic mixture model for anomaly detection to establish a theoretical connection among representation learning, clustering, and anomaly detection.
We have devised an improved anomaly score that more effectively harnesses the combined power of representation learning and clustering.
arXiv Detail & Related papers (2024-06-01T14:30:12Z) - Generating and Reweighting Dense Contrastive Patterns for Unsupervised
Anomaly Detection [59.34318192698142]
We introduce a prior-less anomaly generation paradigm and develop an innovative unsupervised anomaly detection framework named GRAD.
PatchDiff effectively expose various types of anomaly patterns.
experiments on both MVTec AD and MVTec LOCO datasets also support the aforementioned observation.
arXiv Detail & Related papers (2023-12-26T07:08:06Z) - Mitigating Shortcut Learning with Diffusion Counterfactuals and Diverse Ensembles [104.60508550106618]
We propose DiffDiv, an ensemble diversification framework exploiting Diffusion Probabilistic Models (DPMs)<n>We show that DPMs can generate images with novel feature combinations, even when trained on samples displaying correlated input features.<n>We show that DPM-guided diversification is sufficient to remove dependence on shortcut cues, without a need for additional supervised signals.
arXiv Detail & Related papers (2023-11-23T15:47:33Z) - Synthetic outlier generation for anomaly detection in autonomous driving [1.0989593035411862]
Anomaly detection is crucial to identify instances that significantly deviate from established patterns or the majority of data.
In this study, we explore different strategies for training an image semantic segmentation model with an anomaly detection module.
By introducing modifications to the training stage of the state-of-the-art DenseHybrid model, we achieve significant performance improvements in anomaly detection.
arXiv Detail & Related papers (2023-08-04T07:55:32Z) - Diversity-Measurable Anomaly Detection [106.07413438216416]
We propose Diversity-Measurable Anomaly Detection (DMAD) framework to enhance reconstruction diversity.
PDM essentially decouples deformation from embedding and makes the final anomaly score more reliable.
arXiv Detail & Related papers (2023-03-09T05:52:42Z) - Prototypical Residual Networks for Anomaly Detection and Localization [80.5730594002466]
We propose a framework called Prototypical Residual Network (PRN)
PRN learns feature residuals of varying scales and sizes between anomalous and normal patterns to accurately reconstruct the segmentation maps of anomalous regions.
We present a variety of anomaly generation strategies that consider both seen and unseen appearance variance to enlarge and diversify anomalies.
arXiv Detail & Related papers (2022-12-05T05:03:46Z) - Towards Interpretable Anomaly Detection via Invariant Rule Mining [2.538209532048867]
In this work, we pursue highly interpretable anomaly detection via invariant rule mining.
Specifically, we leverage decision tree learning and association rule mining to automatically generate invariant rules.
The generated invariant rules can provide explicit explanation of anomaly detection results and thus are extremely useful for subsequent decision-making.
arXiv Detail & Related papers (2022-11-24T13:03:20Z) - Causality-Based Multivariate Time Series Anomaly Detection [63.799474860969156]
We formulate the anomaly detection problem from a causal perspective and view anomalies as instances that do not follow the regular causal mechanism to generate the multivariate data.
We then propose a causality-based anomaly detection approach, which first learns the causal structure from data and then infers whether an instance is an anomaly relative to the local causal mechanism.
We evaluate our approach with both simulated and public datasets as well as a case study on real-world AIOps applications.
arXiv Detail & Related papers (2022-06-30T06:00:13Z) - Diverse Counterfactual Explanations for Anomaly Detection in Time Series [26.88575131193757]
We propose a model-agnostic algorithm that generates counterfactual ensemble explanations for time series anomaly detection models.
Our method generates a set of diverse counterfactual examples, i.e., multiple versions of the original time series that are not considered anomalous by the detection model.
Our algorithm is applicable to any differentiable anomaly detection model.
arXiv Detail & Related papers (2022-03-21T16:30:34Z)
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.