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

We live in an era of data floods, and deep neural networks play a pivotalrole in this moment. Natural data inherently exhibits several challenges suchas long-tailed distribution and model fairness, where data imbalance is at thecenter of fundamental issues. This imbalance poses a risk of deep neuralnetworks producing biased predictions, leading to potentially severe ethicaland social problems. To address these problems, we leverage the recentgenerative models advanced in generating high-quality images. In this work, wepropose SYNAuG, which utilizes synthetic data to uniformize the given imbalancedistribution followed by a simple post-calibration step considering the domaingap between real and synthetic data. This straightforward approach yieldsimpressive performance on datasets for distinctive data imbalance problems suchas CIFAR100-LT, ImageNet100-LT, UTKFace, and Waterbirds, surpassing theperformance of existing task-specific methods. While we do not claim that ourapproach serves as a complete solution to the problem of data imbalance, weargue that supplementing the existing data with synthetic data proves to be aneffective and crucial step in addressing data imbalance concerns.

SYNAuG: Exploiting Synthetic Data for Data Imbalance Problems