Asterisk: Super-fast MPC with a Friend
Banashri Karmakar, Nishat Koti, Arpita Patra, Sikhar Patranabis, Protik Paul, Divya Ravi
IEEE Symposium on Security and Privacy 2024 · Day 1 · Continental Ballroom 6
The talk "Asterisk: Super-fast MPC with a Friend" introduces a novel approach to **Multi-Party Computation (MPC)** that significantly enhances efficiency and security guarantees by incorporating a semi-honest "helper" party. Presented by a team of researchers including Banashri Karmakar, Nishat Koti, Arpita Patra, Sikhar Patranabis, Protik Paul, and Divya Ravi, this work addresses long-standing challenges in balancing the computational cost, communication overhead, and security resilience of MPC protocols. The core innovation lies in a new threat model that bridges the gap between traditional honest-majority and dishonest-majority MPC paradigms, offering the best features of both.
AI review
Asterisk introduces a highly novel MPC protocol, leveraging a semi-honest helper to achieve fairness and near-dishonest majority resilience with unprecedented efficiency. This foundational work includes an impossibility proof and demonstrates massive performance gains, making advanced privacy-preserving computation practical for sensitive applications like dark pools.