Federated Learning Optimization with Homomorphic Encryption

This case study explores WeBank's innovative use of the FATE framework to enhance federated learning efficiency through partial homomorphic encryption. By partnering with Intel, WeBank accelerates modular exponentiation operations, improving computational efficiency while ensuring data security. The collaboration addresses the challenges of multi-source data sharing in AI applications, particularly in the financial sector, ultimately reducing Total Cost of Ownership (TCO) and enhancing AI model