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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Graph contrastive learning algorithms have demonstrated remarkable success invarious applications such as node classification, link prediction, and graphclustering. However, in unsupervised graph contrastive learning, somecontrastive pairs may contradict the truths in downstream tasks and thus thedecrease of losses on these pairs undesirably harms the performance in thedownstream tasks. To assess the discrepancy between the prediction and theground-truth in the downstream tasks for these contrastive pairs, we adapt theexpected calibration error (ECE) to graph contrastive learning. The analysis ofECE motivates us to propose a novel regularization method, Contrast-Reg, toensure that decreasing the contrastive loss leads to better performance in thedownstream tasks. As a plug-in regularizer, Contrast-Reg effectively improvesthe performance of existing graph contrastive learning algorithms. We provideboth theoretical and empirical results to demonstrate the effectiveness ofContrast-Reg in enhancing the generalizability of the Graph Neural Network(GNN)model and improving the performance of graph contrastive algorithms withdifferent similarity definitions and encoder backbones across variousdownstream tasks.

Calibrating and Improving Graph Contrastive Learning