Obfuscation I
summary
Proposes a concrete candidate for indistinguishability obfuscation that encodes matrix branching programs with Kilian randomization and multilinear-map “jigsaw” encodings, rotates extra primed levels to foil the specific input-mixing attack, and multiplies each matrix by a random scalar. Claims the construction preserves functionality (NC1 via Barrington), yields no known algebraic attacks and gives IO (not VBB) under multilinear assumptions; notes the curious fact that IO would trivially exist if P = NP.
AI-written summary · about
chapters (15)
- 0:00Intro
- 4:00Secure Multiparty Computation
- 7:40Fundamental Break
- 8:15Hiding Secrets
- 10:20Selfeating Programs
- 14:30Positive Results
- 20:35Philosophical Question
- 23:20Interesting Fact
- 27:10Why Im here
- 28:15What will I talk about
- 33:40The formal setting
- 39:43The construction
- 44:27Killian randomization
- 48:33Multilinear maps
- 56:42Obfuscation
related talks
01h 06mObfuscation: Past, Present, and Possible Futures
01h 04mHow to Encrypt a Functionality
01h 04mTutorial on obfuscation
01h 26mObfuscation 1
01h 04mCryptanalysis of Candidate Program Obfuscators
51mObfuscation via GGH15 Encodings, 10 Years Later
31mObfuscation of Probabilistic Circuits and Applications
57mIndistinguishability Obfuscation from Well-Founded Assumptions
