Succinct Randomized Encodings from Lattices
summary
Construction gives witness encryption for P and hence succinct randomized encodings with runtime and memory cost polylogarithmic in program time T and memory M, assuming a circular, decomposed LWE variant. It builds homomorphic lattice encodings, a rate-1 lattice encodings booster via bootstrapping, and a Merkle-tree style recursion to evaluate RAM read/write steps and reuse precomputed K_P values; the scheme meets near‑optimal parameters under the assumption but still relies on circularity and heavy lattice/FHE ingredients.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 5:02Lattice-based constructions
- 13:37Talk outline and homomorphic encodings
- 15:01Witness encryption attempt
- 22:53Problems with witness encryption
- 24:19RAM programs with lattice encodings
- 32:46Merkle tree for succinctness
- 41:10Bootstrapping for reduced evaluation time
- 50:21Assumptions and heuristic
- 53:54Applications and open problems
- 1:02:23Q&A
related talks
57mTutorial: Encrypted Computation from Lattices
01h 07mLattices and Encrypted Computation
01h 04mThe Mathematics of Lattices II
46mCandidate Witness Encryption from Lattice Techniques
01h 29mLattices, Multilinear Maps and Program Obfuscation
01h 03mLockable Obfuscation
54mGraph-Induced Multilinear Maps from Lattices
01h 40mHistorical Talk on Lattice-Based Cryptography
