Mean prediction errors of a BiLSTM trained on real, synthetic, and augmented data for a downstream prediction task. The numbers in parentheses represent the standard deviation

Mean prediction errors of a BiLSTM trained on real, synthetic, and augmented data for a downstream prediction task. The numbers in parentheses represent the standard deviation

<p>Mean prediction errors of a BiLSTM trained on real, synthetic, and augmented data for a downstream prediction task. The numbers in parentheses represent the standard deviation</p>

Saved in:

Bibliographic Details
Main Author:	Raffaele Marchesi (21387979) (author)
Other Authors:	Nicolo Micheletti (21387982) (author), Nicholas I-Hsien Kuo (21387985) (author), Sebastiano Barbieri (410055) (author), Giuseppe Jurman (122705) (author), Venet Osmani (9079892) (author)
Published:	2025
Subjects:	Genetics Immunology Biological Sciences not elsewhere classified specific ethnic backgrounds mitigate representation bias improves model fairness generative adversarial networks avoiding mode collapse world clinical datasets simple resampling methods recent approaches based serious gan failure gan outperforms state synthetic data generated several approaches clinical discoveries art methods gan ). unfair decisions underrepresented subpopulations two diverse synthesises authentic significant challenge research findings quantitative evaluation original distribution novel architecture minority class improved performance health data female patients faithfully maintaining evaluated separately dimensional time black patients benefit equally
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!

Cannot write session to /tmp/vufind_sessions/sess_cdcbnve4pqhepg7p5betdhhh66