Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
DRM-FL: Cross-Silo Federated Learning Á¢±Ù¹ýÀÇ ÇÁ¶óÀ̹ö½Ã º¸È£¸¦ À§ÇÑ ºÐ»êÇü ·£´ýÈ ¸ÞÄ¿´ÏÁò |
¿µ¹®Á¦¸ñ(English Title) |
DRM-FL: A Decentralized and Randomized Mechanism for Privacy Protection in Cross-Silo Federated Learning Approach |
ÀúÀÚ(Author) |
¹«ÇÔ¸¶µå Çʴٿ콺
ÃÊ´À¿¡Áø¶ù
¸¶¸®Áî¾Æ±æ¶ö
ÀÌ°æÇö
Muhammad Firdaus
Cho Nwe Zin Latt
Mariz Aguilar
Kyung-Hyune Rhee
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 29 NO. 01 PP. 0264 ~ 0267 (2022. 05) |
Çѱ۳»¿ë (Korean Abstract) |
Recently, federated learning (FL) has increased prominence as a viable approach for enhancing user privacy and data security by allowing collaborative multi-party model learning without exchanging sensitive data. Despite this, most present FL systems still depend on a centralized aggregator to generate a global model by gathering all submitted models from users, which could expose user privacy and the risk of various threats from malicious users. To solve these issues, we suggested a safe FL framework that employs differential privacy to counter membership inference attacks during the collaborative FL model training process and empowers blockchain to replace the centralized aggregator server. |
¿µ¹®³»¿ë (English Abstract) |
|
Å°¿öµå(Keyword) |
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|