Anomaly Detection Requires Better Representations

Reiss, Tal; Cohen, Niv; Horwitz, Eliahu; Abutbul, Ron; Hoshen, Yedid

doi:10.1007/978-3-031-25069-9_4

Cited by 9 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such knowledge may consist of statistical assumptions regarding the distribution of the normal data that can be used to adapt the representation. This was done in the cases of previous methods: DeepSVDD ( 48 ), PANDA ( 49 ), Mahalanobis ( 50 ) and OOD-no-labels ( 29 ). Prior knowledge may also come in the form of additional auxiliary features ( 18 ) or labels.…”

Section: Discussionmentioning

confidence: 99%

Detecting anomalous proteins using deep representations

Michael-Pitschaze,

Cohen,

Ofer

et al. 2024

NAR Genomics and Bioinformatics

Self Cite

View full text Add to dashboard Cite

Many advances in biomedicine can be attributed to identifying unusual proteins and genes. Many of these proteins’ unique properties were discovered by manual inspection, which is becoming infeasible at the scale of modern protein datasets. Here, we propose to tackle this challenge using anomaly detection methods that automatically identify unexpected properties. We adopt a state-of-the-art anomaly detection paradigm from computer vision, to highlight unusual proteins. We generate meaningful representations without labeled inputs, using pretrained deep neural network models. We apply these protein language models (pLM) to detect anomalies in function, phylogenetic families, and segmentation tasks. We compute protein anomaly scores to highlight human prion-like proteins, distinguish viral proteins from their host proteome, and mark non-classical ion/metal binding proteins and enzymes. Other tasks concern segmentation of protein sequences into folded and unstructured regions. We provide candidates for rare functionality (e.g. prion proteins). Additionally, we show the anomaly score is useful in 3D folding-related segmentation. Our novel method shows improved performance over strong baselines and has objectively high performance across a variety of tasks. We conclude that the combination of pLM and anomaly detection techniques is a valid method for discovering a range of global and local protein characteristics.

show abstract

Section: Discussionmentioning

confidence: 99%

Detecting anomalous proteins using deep representations

Michael-Pitschaze,

Cohen,

Ofer

et al. 2024

NAR Genomics and Bioinformatics

Self Cite

View full text Add to dashboard Cite

show abstract

ECS-STPM: An Efficient Model for Tunnel Fire Anomaly Detection

Song,

Wen,

Song

et al. 2024

Web and Big Data

View full text Add to dashboard Cite

Understanding the limitations of self-supervised learning for tabular anomaly detection

Mai,

Davies,

Griffin

2024

Pattern Anal Applic

View full text Add to dashboard Cite

While self-supervised learning has improved anomaly detection in computer vision and natural language processing, it is unclear whether tabular data can benefit from it. This paper explores the limitations of self-supervision for tabular anomaly detection. We conduct several experiments spanning various pretext tasks on 26 benchmark datasets to understand why this is the case. Our results confirm representations derived from self-supervision do not improve tabular anomaly detection performance compared to using the raw representations of the data. We show this is due to neural networks introducing irrelevant features, which reduces the effectiveness of anomaly detectors. However, we demonstrate that using a subspace of the neural network’s representation can recover performance.

show abstract

Anomaly Detection Requires Better Representations

Cited by 9 publications

References 20 publications

Detecting anomalous proteins using deep representations

Detecting anomalous proteins using deep representations

ECS-STPM: An Efficient Model for Tunnel Fire Anomaly Detection

Understanding the limitations of self-supervised learning for tabular anomaly detection

Contact Info

Product

Resources

About