Showing cs.LGShow all
2 papers · 1 filter
cs.LG2024
Inference-Time Rule Eraser: Fair Recognition via Distilling and Removing Biased Rules
Yi Zhang, Dongyuan Lu, Jitao Sang
Machine learning models often make predictions based on biased features such as gender, race, and other social attributes, posing significant fairness risks, especially in societal…
cs.LG2020
An Experimental Study of Semantic Continuity for Deep Learning Models
Shangxi Wu, Dongyuan Lu, Xian Zhao +2
Deep learning models suffer from the problem of semantic discontinuity: small perturbations in the input space tend to cause semantic-level interference to the model output. We arg…