Privacy and distribution preserving generative adversarial networks with sample balancing

Differential privacy (DP) generative adversarial networks (GANs) can generate protected synthetic samples from downstream analysis. However, training on unbalanced datasets can bias the network towards majority classes, leading minority undertrained. Meanwhile, gradient perturbation in DP has no gua...

Celý popis

Podrobná bibliografie

Vydáno v:	Expert systems with applications Ročník 258; s. 125181
Hlavní autoři:	Sun, Haoran, Tang, Jinchuan, Dang, Shuping, Chen, Gaojie
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Elsevier Ltd 15.12.2024
Témata:	Data balancing Data-wise differential privacy Feature holding Generative adversarial networks Sample perturbation Sample perturbation Generative adversarial networks Data-wise differential privacy Data balancing Feature holding
ISSN:	0957-4174
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!