johnantonn / cash-for-unsupervised-ad

Systematic Evaluation of CASH Search Strategies for Unsupervised Anomaly Detection
3 stars 0 forks source link
anomaly-detection automl cash model-selection outlier-detection smac unsupervised-anomaly-detection

Systematic Evaluation of CASH Search Strategies for Unsupervised Anomaly Detection

Repository for the corresponding full-paper accepted at the LIDTA-2022 workshop of the ECML/PKDD 2022.

Note: This repository initially served as the code repo for my thesis in Master of Artificial Intelligence programme at KU Leuven but it was later modified/extended to accommodate the relevant content of the LIDTA-2022 full-paper submission.

Description

The code provides an experimental evaluation of how the structure of the validation set, i.e., its size and label bias, impacts the performance of different CASH search strategies within the context of anomaly detection.

Contents

How to run the code

Provide the experiment parameters in src/config.json:

External links

Name Description Link
Auto-Sklearn Automated machine learning toolkit :link:
PyOD Python library for anomaly detection :link:
Datasets Anomaly detection datasets :link:

License

Copyright © 2022 Ioannis Antoniadis