Alphabeta Math
TheoremStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck pass
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.

Parity has approximate degree N and requires at least N over 2 queries

Statement

For N≥1, define PARITYN(x)=∣x∣(mod2). For every 0≤ϵ<1/2, deg⁡~ϵ(PARITYN;{0,1}N)=N, and every local bit-query algorithm computing parity with error at most ϵ uses at least ⌈N/2⌉ queries.

Facts & Assumptions

Given: An integer N≥1, an error 0≤ϵ<1/2, and the parity function on the full Boolean cube.

[F1]

Acceptance polynomials and approximate degree on Boolean domains: approximate degree is the least degree bound of a real approximant on the stated domain.

[F2]

Symmetrizing a Boolean-cube polynomial gives a Hamming-weight polynomial: an approximant to a symmetric function can be replaced by a univariate polynomial in Hamming weight without increasing degree.

[F4]

A nonzero real polynomial of degree n has no more than n distinct real roots: a nonzero univariate degree-m polynomial has at most m distinct roots.

[F5]

A T-query bit algorithm has an acceptance polynomial of degree at most 2T: local bit-query complexity is at least the ceiling of half the approximate degree.

[F6]

Polynomial rings in finitely many commuting indeterminates by iteration and Monomials, coefficients, degree in each variable and total degree in F[x1,…,xn]: finite products and sums of real linear polynomials are real polynomials, and the product of N linear factors has total degree at most N.

[F7]

The reals form a totally ordered field: real arithmetic respects a total order, so ϵ<1/2 makes the even and odd approximating values lie strictly on opposite sides of 1/2.

Proof

1.1givenF1F2F7algebra

Choose an ϵ-approximant p of degree at most d=deg⁡~ϵ(PARITYN) by [F1]. By [F2] it yields a univariate polynomial q(t) of degree at most d with ∣q(k)−(k mod 2)∣≤ϵ for every integer 0≤k≤N. Hence q(k)−1/2<0 for even k and q(k)−1/2>0 for odd k, because ϵ<1/2 and [F7].

2.1step 1.1F3F4algebra

The polynomial q(t)−1/2 is continuous by [F3]. For each k=0,…,N−1, its values at k and k+1 have opposite signs by step 1.1. The intermediate value theorem [F3] gives a root in each open interval (k,k+1); the root is interior because neither endpoint value is zero. These N open intervals are disjoint, so the roots are N distinct real numbers. Since q−1/2 is not the zero polynomial (its value at 0 is negative), [F4] implies its degree is at least N, and therefore d≥N.

3.1givenF1F5F6step 2.1algebra∎

The polynomial P(X)=12(1−∏i=0N−1(1−2Xi)) has degree at most N by [F6] and equals parity on the Boolean cube, since each factor 1−2xi is 1 for bit 0 and −1 for bit 1. Thus the approximate degree is at most N; with step 2.1 it equals N. The query lower bound follows from [F5].

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

55 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