Hyperparameter Sensitivity in Deep Outlier Detection: Analysis and a Scalable Hyper-Ensemble Solution

TL;DR

This paper analyzes the hyperparameter sensitivity of deep outlier detection methods and introduces ROBOD, a scalable hyper-ensemble approach that improves robustness and reduces training time while maintaining state-of-the-art performance.

Contribution

It provides the first large-scale analysis of hyperparameter sensitivity in deep OD and proposes ROBOD, a novel hyper-ensemble method with strategies for efficient training.

Findings

Hyperparameter choice is critical in deep outlier detection.

ROBOD achieves robust, state-of-the-art performance.

Training time is reduced to 2-10% of naive ensembles.

Abstract

Outlier detection (OD) literature exhibits numerous algorithms as it applies to diverse domains. However, given a new detection task, it is unclear how to choose an algorithm to use, nor how to set its hyperparameter(s) (HPs) in unsupervised settings. HP tuning is an ever-growing problem with the arrival of many new detectors based on deep learning, which usually come with a long list of HPs. Surprisingly, the issue of model selection in the outlier mining literature has been "the elephant in the room"; a significant factor in unlocking the utmost potential of deep methods, yet little said or done to systematically tackle the issue. In the first part of this paper, we conduct the first large-scale analysis on the HP sensitivity of deep OD methods, and through more than 35,000 trained models, quantitatively demonstrate that model selection is inevitable. Next, we design a HP-robust and scalable deep hyper-ensemble model called ROBOD that assembles models with varying HP configurations, bypassing the choice paralysis. Importantly, we introduce novel strategies to speed up ensemble training, such as parameter sharing, batch/simultaneous training, and data subsampling, that allow us to train fewer models with fewer parameters. Extensive experiments on both image and tabular datasets show that ROBOD achieves and retains robust, state-of-the-art detection performance as compared to its modern counterparts, while taking only $2$-$10$\% of the time by the naive hyper-ensemble with independent training.