The need for unsupervised outlier model selection: A review and evaluation of internal evaluation strategies

MQ Ma, Y Zhao, X Zhang, L Akoglu - ACM SIGKDD Explorations …, 2023 - dl.acm.org
Given an unsupervised outlier detection task, how should one select i) a detection algorithm,
and ii) associated hyperparameter values (jointly called a model)? E ective outlier model …

Bond: Benchmarking unsupervised outlier node detection on static attributed graphs

K Liu, Y Dou, Y Zhao, X Ding, X Hu… - Advances in …, 2022 - proceedings.neurips.cc
Detecting which nodes in graphs are outliers is a relatively new machine learning task with
numerous applications. Despite the proliferation of algorithms developed in recent years for …

Ensemble of Autoencoders for Anomaly Detection in Biomedical Data: A Narrative Review

A Nawaz, SS Khan, A Ahmad - IEEE Access, 2024 - ieeexplore.ieee.org
In the context of biomedical data, an anomaly could refer to a rare or new type of disease, an
aberration from normal behavior, or an unexpected observation requiring immediate …

Admoe: Anomaly detection with mixture-of-experts from noisy labels

Y Zhao, G Zheng, S Mukherjee, R McCann… - Proceedings of the …, 2023 - ojs.aaai.org
Existing works on anomaly detection (AD) rely on clean labels from human annotators that
are expensive to acquire in practice. In this work, we propose a method to leverage …

Entropystop: Unsupervised deep outlier detection with loss entropy

Y Huang, Y Zhang, L Wang, F Zhang… - Proceedings of the 30th …, 2024 - dl.acm.org
Unsupervised Outlier Detection (UOD) is an important data mining task. With the advance of
deep learning, deep Outlier Detection (OD) has received broad interest. Most deep UOD …

Learning more effective cell representations efficiently

JX Dou, M Jia, N Zaslavsky, M Ebeid, R Bao… - … 2022 Workshop on …, 2022 - openreview.net
Capturing similarity among cells is at the core of many tasks in single-cell transcriptomics,
such as the identification of cell types and cell states. This problem can be formulated in a …

Robust one-class classification with signed distance function using 1-lipschitz neural networks

L Béthune, P Novello, T Boissin, G Coiffier… - arXiv preprint arXiv …, 2023 - arxiv.org
We propose a new method, dubbed One Class Signed Distance Function (OCSDF), to
perform One Class Classification (OCC) by provably learning the Signed Distance Function …

Data augmentation is a hyperparameter: Cherry-picked self-supervision for unsupervised anomaly detection is creating the illusion of success

J Yoo, T Zhao, L Akoglu - arXiv preprint arXiv:2208.07734, 2022 - arxiv.org
Self-supervised learning (SSL) has emerged as a promising alternative to create
supervisory signals to real-world problems, avoiding the extensive cost of manual labeling …

Model Selection of Anomaly Detectors in the Absence of Labeled Validation Data

C Fung, C Qiu, A Li, M Rudolph - arXiv preprint arXiv:2310.10461, 2023 - arxiv.org
Anomaly detection requires detecting abnormal samples in large unlabeled datasets. While
progress in deep learning and the advent of foundation models has produced powerful …

Outlier Detection Bias Busted: Understanding Sources of Algorithmic Bias through Data-centric Factors

X Ding, R Xi, L Akoglu - Proceedings of the AAAI/ACM Conference on AI …, 2024 - ojs.aaai.org
The astonishing successes of ML have raised growing concern for the fairness of modern
methods when deployed in real world settings. However, studies on fairness have mostly …