UniOD: A Universal Model for Outlier Detection across Diverse Domains
This addresses the inconvenience and computational cost of outlier detection for practitioners by enabling a single model to generalize across domains, though it is incremental as it builds on existing graph-based and transfer learning methods.
The paper tackles the problem of outlier detection requiring dataset-specific tuning and training by proposing UniOD, a universal model that trains on labeled datasets to detect outliers across diverse domains without per-dataset tuning, achieving effectiveness on 15 benchmark datasets against 15 baselines.
Outlier detection (OD) seeks to distinguish inliers and outliers in completely unlabeled datasets and plays a vital role in science and engineering. Most existing OD methods require troublesome dataset-specific hyperparameter tuning and costly model training before they can be deployed to identify outliers. In this work, we propose UniOD, a universal OD framework that leverages labeled datasets to train a single model capable of detecting outliers of datasets from diverse domains. Specifically, UniOD converts each dataset into multiple graphs, produces consistent node features, and frames outlier detection as a node-classification task, and is able to generalize to unseen domains. As a result, UniOD avoids effort on model selection and hyperparameter tuning, reduces computational cost, and effectively utilizes the knowledge from historical datasets, which improves the convenience and accuracy in real applications. We evaluate UniOD on 15 benchmark OD datasets against 15 state-of-the-art baselines, demonstrating its effectiveness.