English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Unsupervised outlier detection, which predicts if a test sample is an outlieror not using only the information from unlabelled inlier data, is an importantbut challenging task. Recently, methods based on the two-stage frameworkachieve state-of-the-art performance on this task. The framework leveragesself-supervised representation learning algorithms to train a feature extractoron inlier data, and applies a simple outlier detector in the feature space. Inthis paper, we explore the possibility of avoiding the high cost of training adistinct representation for each outlier detection task, and instead using asingle pre-trained network as the universal feature extractor regardless of thesource of in-domain data. In particular, we replace the task-specific featureextractor by one network pre-trained on ImageNet with a self-supervised loss.In experiments, we demonstrate competitive or better performance on a varietyof outlier detection benchmarks compared with previous two-stage methods,suggesting that learning representations from in-domain data may be unnecessaryfor outlier detection.

Do We Really Need to Learn Representations from In-domain Data for Outlier Detection?