Secure Computation 1
Talkat Cryptography 10 Years Later: Obfuscation, Proof Systems, and Secure Computation Boot Camp 2025 · 2025-05-19..05-23·
summary
Quick tour of modern MPC with concrete trade-offs: honest‑majority uses Shamir secret sharing and degree reduction or Damgård‑Nielsen preprocessing to cut communication to O(n), yielding practical systems like ATLAS that report about 2 million mults/sec for 61‑bit fields with three parties and active security (with abort). Dishonest‑majority relies on Beaver triples and SPDZ‑style authenticated sharings to get active security with small online cost, while one‑time truth tables give OT‑style optimal online bandwidth at exponential preprocessing. Garbled circuits keep two‑round latency; best AND cost is ~1.5λ bits with free XOR and recent loglog circuit‑size improvements.
AI-written summary · about
chapters (6)
- 0:00Introduction to MPC
- 13:55Secret sharing foundations
- 21:40Efficient honest majority MPC
- 28:56Dishonest majority protocols
- 37:36Active security with MACs
- 48:55Garbled circuits
related talks
40mApplied MPC
58mSecure Multiparty Computation I
01h 06mSecure Computation and Low-Complexity Cryptography
45mPrice of Active Security in Multiparty Computation
40mCryptography in Distributed Systems
29mTowards Scalable Constant-Round MPC from Minimal Assumptions via Round Collapsing
01h 10mSecure Multiparty Computation II
35mOblivious RAM for MPC