Hyperparameter Optimization for Unsupervised Outlier Detection

TL;DR

This paper introduces HPOD, a meta-learning based method for optimizing hyperparameters in unsupervised outlier detection without labels, significantly improving performance across various algorithms and datasets.

Contribution

The paper presents the first systematic approach, HPOD, for hyperparameter optimization in unsupervised outlier detection using meta-learning and efficient sampling techniques.

Findings

HPOD outperforms baseline methods with 58-66% performance gains.

Works effectively on both deep and shallow OD algorithms.

Applicable to discrete and continuous hyperparameter spaces.

Abstract

Given an unsupervised outlier detection (OD) algorithm, how can we optimize its hyperparameter(s) (HP) on a new dataset, without any labels? In this work, we address this challenging hyperparameter optimization for unsupervised OD problem, and propose the first systematic approach called HPOD that is based on meta-learning. HPOD capitalizes on the prior performance of a large collection of HPs on existing OD benchmark datasets, and transfers this information to enable HP evaluation on a new dataset without labels. Moreover, HPOD adapts a prominent sampling paradigm to identify promising HPs efficiently. Extensive experiments show that HPOD works with both deep (e.g., Robust AutoEncoder) and shallow (e.g., Local Outlier Factor (LOF) and Isolation Forest (iForest)) OD algorithms on discrete and continuous HP spaces, and outperforms a wide range of baselines with on average 58% and 66% performance improvement over the default HPs of LOF and iForest.