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

The deep neural network (DNN) has been proven effective in various domains.However, they often struggle to perform well on certain minority groups duringinference, despite showing strong performance on the majority of data groups.This is because over-parameterized models learned \textit{bias attributes} froma large number of \textit{bias-aligned} training samples. These bias attributesare strongly spuriously correlated with the target variable, causing the modelsto be biased towards spurious correlations (i.e., \textit{bias-conflicting}).To tackle this issue, we propose a novel \textbf{re}weighted \textbf{s}parse\textbf{t}raining framework, dubbed as \textit{\textbf{REST}}, which aims toenhance the performance of biased data while improving computation and memoryefficiency. Our proposed REST framework has been experimentally validated onthree datasets, demonstrating its effectiveness in exploring unbiasedsubnetworks. We found that REST reduces the reliance on spuriously correlatedfeatures, leading to better performance across a wider range of data groupswith fewer training and inference resources. We highlight that the\textit{REST} framework represents a promising approach for improving theperformance of DNNs on biased data, while simultaneously improving computationand memory efficiency. By reducing the reliance on spurious correlations, RESThas the potential to enhance the robustness of DNNs and improve theirgeneralization capabilities. Code is released at\url{https://github.com/zhao1402072392/REST}

REST: Enhancing Group Robustness in DNNs through Reweighted Sparse Training