Homomorphic Secret Sharing: Secure Computation and Beyond

ongoing

"Secure computation is a powerful cryptographic tool that enables distrusting parties to jointly emulate the correctness and privacy guarantees of an incorruptible trusted third party. The past decades have seen tremendous progress in secure computation protocol design, yet there remain significant gaps in our understanding. On one hand, ""classical"" approaches require excessive communication between parties; on the other hand, approaches based on fully homomorphic encryption (FHE) require heavy computation as well as significant concrete communication costs. These shortcomings are particularly limiting as the scale of secure computations grows to match the modern regime of big data. The central objective of Project HSS, is to understand and develop the capabilities of secure computation design and surrounding technology via a new methodology of homomorphic secret sharing (HSS). Recently introduced by the PI, HSS is a useful relaxation of FHE where homomorphic evaluation may be distributed across parties who do not interact. Initial results indicate that HSS yields a new paradigm for secure computation with analogous advantages as FHE, while opening the door to simpler, more efficient solutions. HSS-based techniques demonstrate exciting potential also as a means for greatly advancing supporting technologies within all paradigms, and even bridging to other disciplines, with developments and implications reaching beyond the realm of secure computation. The focus of Project HSS will be in exploring these vast new horizons. "