SYSTEM ENABLING DATA INTERCONNECTION BETWEEN SECURE MULTI-PARTY COMPUTATION AND FEDERATED LEARNING

• Main Authors: XIE, Xiang, LI, Shenglin, DU, Jifeng, SUN, Lilin
• Format: Patent
• Language: Chinese; English; French
• Published: 19.10.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; PHYSICS; TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHICCOMMUNICATION

Disclosed is a system enabling data interconnection between secure multi-party computation and federated learning, which comprises a federated learning system and a secret sharing module imported into the federated learning system, the secret sharing module being in communication connection with the secure multi-party computation system. The secret sharing module is used for realizing secret sharing of computing nodes in the secure multi-party computation system. According to the present invention, the secret sharing module is imported into the federated learning system, and is used for realizing secret sharing of computing nodes in the secure multi-party computation system, so as to realize data interconnection between the secure multi-party computation system and the federated learning system. The secret sharing module can be imported in a plug-in mode, is lightweight and easy to use, has a short response period, and can meet real-world requirements at a low cost. L'invention divulgue un système permettant