Training unbiased diffusion models from biased dataset

Y Kim, B Na, M Park, JH Jang, D Kim… - The Twelfth …, 2024 - openreview.net
With significant advancements in diffusion models, addressing the potential risks of dataset
bias becomes increasingly important. Since generated outputs directly suffer from dataset …

Fair Transition Loss: From label noise robustness to bias mitigation

Y Canalli, F Braida, L Alvim, G Zimbrão - Knowledge-Based Systems, 2024 - Elsevier
The Machine learning widespread adoption has inadvertently led to the amplification of
societal biases and discrimination, with many consequential decisions now influenced by …

[PDF][PDF] ROBUST LOSS AND PENALTY FOR FAIR MACHINE LEARNING

Y de Mello Canalli - 2024 - cos.ufrj.br
Brisha Borden was running late to pick up her god-sister from school. She and her friend,
both 18-year-old girls, took an unlocked kid's bicycle and scooter and ride for a while. They …