How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
PCP soundness amplification by independent repetition
Statement
Let be a nonadaptive PCP verifier with addressable proof length , randomness bound , query bound , fixed finite proof alphabet, and completeness at least the constant and soundness at most the constant , where . For every fixed integer , repeat on independent random tapes, using the same fixed proof in every run, and accept if and only if all runs accept. The repeated verifier has proof length , randomness bound , query bound , completeness at least , and soundness at most . If , perfect completeness remains perfect. For every fixed target , a fixed can be chosen so that ; if , take . No claim of reaching target is made when .
Facts & Assumptions
Given: A verifier satisfying the fixed-proof completeness and soundness conditions of the statement, and a fixed positive integer .
PCP completeness uses one fixed proof on a yes input, while soundness holds for every fixed proof on a no input; the proof alphabet and resource bounds are fixed for the verifier. (PCP classes with completeness and soundness)
A finite product of finite probability spaces has product outcomes and product weights. (The finite product of finite probability spaces)
In a finite product space, events determined by distinct coordinates are mutually independent. (Product weights normalize, and coordinate events are mutually independent)
Independence of event classes means that every finite choice of one event from each of distinct classes has intersection probability equal to the product of its probabilities. (Independent families of event classes)
Proof
Define to use independent blocks of random bits, run once on each block with the original fixed proof, and accept exactly when every run accepts. Concatenating the query lists gives at most symbol queries, all determined by the input and full random tape before answers are read; the proof length and alphabet are unchanged, and the verifier remains uniform polynomial time for fixed .
Fix an input and proof , let , and let be the event that run accepts. The coin blocks form the product space in [F2], each depends only on coordinate , and [F3] makes them mutually independent in the sense of [F4]. Therefore . This remains valid for zero random bits, where each coordinate space is a singleton.
If is a yes input, [F1] supplies one fixed proof with , so step 1.2 gives acceptance . If is a no input, every fixed proof has , so every repeated run on that same proof has acceptance . In particular gives completeness one.
For , put . For each , , hence . Choosing a fixed integer gives ; when , step 2.1 already gives soundness zero with . Because are constants independent of , this is fixed, so multiplying and by preserves logarithmic randomness and constant query bounds.
Depends on
Used by
Dependency tree · two levels
12 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.
Sources
- Arora and Barak, Computational Complexity: A Modern Approach, §18.1, Note 3 to Theorem 18.2, printed p. 354 (standard reference, not scraped)