Study Overview
This research proposes a novel decentralized platform enabling privacy-preserving data collaboration and analytics across organizations with private data sources. The platform integrates advanced cryptographic techniques like secure multi-party computation and federated learning to facilitate secure cross-entity data analysis and machine learning for accurate financial auditing, fraud/laundering detection while ensuring data privacy. Its decentralized approach breaks data silos, empowering collaborative insights extraction without compromising privacy and security. The platform simplifies development of privacy-preserving applications, reduces costs, and promotes regulatory compliance. It has broader applications beyond finance for secure cross-organizational analytics through privacy-preserving decentralized computing.
Study Results
We built HADES, a fully homomorphic encryption (FHE) based private aggregation system for public data that supports point, range predicates, and boolean combinations. Our one-round HADES protocol efficiently generates predicate indicators by leveraging the plaintext form of public data records. It introduces a novel elementwise-mapping operation and an optimized reduction algorithm, achieving latency efficiency within a limited noise budget. Our highly scalable, multi-threaded implementation improves performance over previous one-round FHE solutions by 204x to 6574x on end-to-end TPC-H queries, reducing aggregation time on 1M records from 15 hours to 38 seconds
IBSI Funding Acknowledgement: Lab for Inclusive FinTech (LIFT)