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

Deep models trained with noisy labels are prone to over-fitting and strugglein generalization. Most existing solutions are based on an ideal assumptionthat the label noise is class-conditional, i.e., instances of the same classshare the same noise model, and are independent of features. While in practice,the real-world noise patterns are usually more fine-grained asinstance-dependent ones, which poses a big challenge, especially in thepresence of inter-class imbalance. In this paper, we propose a two-stage cleansamples identification method to address the aforementioned challenge. First,we employ a class-level feature clustering procedure for the earlyidentification of clean samples that are near the class-wise predictioncenters. Notably, we address the class imbalance problem by aggregating rareclasses according to their prediction entropy. Second, for the remaining cleansamples that are close to the ground truth class boundary (usually mixed withthe samples with instance-dependent noises), we propose a novelconsistency-based classification method that identifies them using theconsistency of two classifier heads: the higher the consistency, the larger theprobability that a sample is clean. Extensive experiments on severalchallenging benchmarks demonstrate the superior performance of our methodagainst the state-of-the-art.

Centrality and Consistency: Two-Stage Clean Samples Identification for Learning with Instance-Dependent Noisy Labels