aiqbench analysis
This paper introduces ISER, a novel unsupervised tabular anomaly detection framework that improves upon Isolation Forest by using hypersphere radii to encode local density and address axis-parallel bias. The method maintains linear time complexity while demonstrating competitive performance across 20 real-world datasets for tasks ranging from network security to quality control. It represents a significant step forward in handling conflicting distributional assumptions within machine learning anomaly detection pipelines.
Isolation-based Spherical Ensemble Representations for Tabular Anomaly Detection
arXiv:2510.13311v2 Announce Type: replace Abstract: Unsupervised tabular anomaly detection is a critical task with applications spanning offensive language detection, network security, and quality control. Despite extensive research, existing unsupervised anomaly detection methods still face fundamental challenges including conflicting distributional assumptions, computational inefficiency, and difficulty handling different anomaly types. To address these problems, we propose ISER (Isolation-based Spherical Ensemble Representations) that extends existing isolation-based methods by using hypersphere radii as a monotonic transformation of local density characteristics while maintaining linear time and constant space complexity w.r.t. the dataset size. ISER constructs ensemble representations where hypersphere radii encode local sparsity through a monotonic transformation of density: smaller radii correspond to dense regions while larger radii correspond to sparse regions. We introduce a novel similarity-based scoring method that measures pattern consistency by comparing ensemble representations against a theoretical anomaly reference pattern. Additionally, we enhance the performance of Isolation Forest by using ISER and adapting the scoring function to address axis-parallel bias and local anomaly detection limitations. Comprehensive experiments on 20 real-world datasets demonstrate ISER's competitive performance over 12 SOTA methods.
Abstract:Unsupervised tabular anomaly detection is a critical task with applications spanning offensive language detection, network security, and quality control. Despite extensive research, existing unsupervised anomaly detection methods still face fundamental challenges including conflicting distributional assumptions, computational inefficiency, and difficulty handling different anomaly types. To address these problems, we propose ISER (Isolation-based Spherical Ensemble Representations) that extends existing isolation-based methods by using hypersphere radii as a monotonic transformation of local density characteristics while maintaining linear time and constant space complexity w.r.t. the dataset size. ISER constructs ensemble representations where hypersphere radii encode local sparsity through a monotonic transformation of density: smaller radii correspond to dense regions while larger radii correspond to sparse regions. We introduce a novel similarity-based scoring method that measures pattern consistency by comparing ensemble representations against a theoretical anomaly reference pattern. Additionally, we enhance the performance of Isolation Forest by using ISER and adapting the scoring function to address axis-parallel bias and local anomaly detection limitations. Comprehensive experiments on 20 real-world datasets demonstrate ISER's competitive performance over 12 SOTA methods.
Machine Learning (cs.LG)
arXiv:2510.13311 [cs.LG]
(or arXiv:2510.13311v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2510.13311 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 264 words
Related coverage
RAD: Rule-Augmented Relational Anomaly Detection
arXiv · Aug 25, 2026
Revisiting Energy-based Tabular Anomaly Detection: Energy and Reconstruction are Complementary
arXiv · Aug 17, 2026
Handling Imbalanced Classification: What Works Better Than SMOTE
Analytics Vidhya · Jul 12, 2026
Correlation-Aware Reconstruction for Multivariate Time Series Anomaly Detection
arXiv · Oct 14, 2026