Delegating Computation I
summary
Constructs a practical path from multiprover proofs to a single-cloud protocol: a two-message delegation scheme for RAM-style computations whose verifier time does not depend on huge outsourced data and whose prover runs near the RAM runtime, proved under the LWE assumption. Key technical fix is replacing the naive FHE-encrypted two-query trick with MIPs resilient to no-signaling strategies, avoiding the Duco et al. attack and yielding quasi-linear verifier time and near-linear prover overhead.
AI-written summary · about
chapters (11)
- 0:00Introduction and welcome
- 0:30Delegating computation
- 2:55Defining the delegation goal
- 7:00Memory delegation settings
- 10:24Two-message RAM delegation
- 12:53History of delegation schemes
- 19:34Landscape of delegation work
- 27:27Pre-processing and snarks
- 36:14Multi-prover interactive proofs
- 42:31Converting MIP to one prover
- 48:20No-signaling strategies
related talks
01h 09mProofs - Practice 1
01h 00mSecure Computation 3
01h 08mSecure Computation 1
59mDoubly Efficient Private Information Retrieval and Fully Homomorphic RAM Computation from Ring LWE
01h 07mProofs - Practice 2
01h 23mProgram Presentations
57mThe Cryptographic Lens: Visions of our Past and Future
01h 08mTutorial on Proof Systems
