Out-of-Distribution Detection without Class Labels

12/14/2021
by   Niv Cohen, et al.
0

Anomaly detection methods identify samples that deviate from the normal behavior of the dataset. It is typically tackled either for training sets containing normal data from multiple labeled classes or a single unlabeled class. Current methods struggle when faced with training data consisting of multiple classes but no labels. In this work, we first discover that classifiers learned by self-supervised image clustering methods provide a strong baseline for anomaly detection on unlabeled multi-class datasets. Perhaps surprisingly, we find that initializing clustering methods with pre-trained features does not improve over their self-supervised counterparts. This is due to the phenomenon of catastrophic forgetting. Instead, we suggest a two stage approach. We first cluster images using self-supervised methods and obtain a cluster label for every image. We use the cluster labels as "pseudo supervision" for out-of-distribution (OOD) methods. Specifically, we finetune pretrained features on the task of classifying images by their cluster labels. We provide extensive analyses of our method and demonstrate the necessity of our two-stage approach. We evaluate it against the state-of-the-art self-supervised and pretrained methods and demonstrate superior performance.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/24/2023

AMAE: Adaptation of Pre-Trained Masked Autoencoder for Dual-Distribution Anomaly Detection in Chest X-Rays

Unsupervised anomaly detection in medical images such as chest radiograp...
research
04/08/2021

The Single-Noun Prior for Image Clustering

Self-supervised clustering methods have achieved increasing accuracy in ...
research
01/13/2022

STEdge: Self-training Edge Detection with Multi-layer Teaching and Regularization

Learning-based edge detection has hereunto been strongly supervised with...
research
03/24/2020

Attention-Based Self-Supervised Feature Learning for Security Data

While applications of machine learning in cyber-security have grown rapi...
research
12/04/2020

Optical Wavelength Guided Self-Supervised Feature Learning For Galaxy Cluster Richness Estimate

Most galaxies in the nearby Universe are gravitationally bound to a clus...
research
09/28/2022

Feature Decoupling in Self-supervised Representation Learning for Open Set Recognition

Assuming unknown classes could be present during classification, the ope...
research
01/17/2022

Self-Supervised Anomaly Detection by Self-Distillation and Negative Sampling

Detecting whether examples belong to a given in-distribution or are Out-...

Please sign up or login with your details

Forgot password? Click here to reset