Secure Multi-party Computation over Noisy Networks

Secure multi-party computation allows a set of mutually distrustful parties to compute a joint function securely, i.e., so that no party or adversary (controlling a coalition of parties) can disrupt the computation or learn private information. The problem has seen a flourish recently and has attracted attention by both theorists and practitioners. However, the vast majority of proposed solutions assumes that the parties can communicate over a reliable network. Obtaining such a network on top of the Internet incurs heavy overheads due to error correction.