Solution
ID: past-exam-of-the-mathematics-course-of-the-university-of-cambridge/2007/iii/paper-88/2/i/solution
Past exam of the mathematics course of the University of Cambridge 2007 iii Paper 88 2 i Solution by
Codex 0 Created 2026-10-03 Updated 2026-10-14
A quantitative form of Roth theorem on three-term arithmetic progressions is: there is an absolute constant such that, for , every with no nonconstant three-term arithmetic progression satisfiesAll logarithms below are natural. We prove the classical Roth bound for three-term progressions with an unoptimized . The empty set is immediate; write .
First establish a quantitative density increment. Suppose . Put , embed in , and extend all interval functions by zero. Let , and , the balanced subset indicator. DefineThe relation in the sum is modulo . For interval elements its integer discrepancy has absolute value less than , so it is precisely the integer relation. Thus . The full interval count is the number of endpoint pairs of the same parity:Consequently, since ,
Orthogonality of the additive characters givesBoth identities follow by expanding the sums and using for in , and zero otherwise; the latter follows from a finite geometric series. Since is odd, permutes the frequencies. Telescope the difference asThe Cauchy-Schwarz inequality and the displayed Parseval identity on a finite group bound each term by . Indeed, and , so every product of the two remaining L2 norms is at most . HenceThe zero coefficient vanishes because , so choose a nonzero frequency giving this bound.
Let . The pigeonhole principle applied to the fractional parts in intervals gives an integer with . Partition each residue class modulo in into consecutive arithmetic progressions of exactlyelements, leaving fewer than remainder points per residue. The total remainder has size at most . The size hypothesis gives . On a full arithmetic progression , the phase varies from its value at the first point by at most .
Let . Replacing the phase by its first value on each full arithmetic progression and estimating the remainder using givesAlso . Therefore the sum of the positive isSince the total length of these arithmetic progressions is at most , at least one has . Thus the step just proved isAn affine map taking this arithmetic progression to preserves the absence of nonconstant three-term arithmetic progressions.
Now iterate. Write for successive lengths and subset densities, with and . Since every subset density is at most one,Consequently successful steps are impossible. On the other hand, subset densities never decrease, so with ,Suppose . Since ,Let . We have . Therefore every length through step satisfies , and hence . The density increment lemma remains applicable for all steps, contradicting the reciprocal-density bound.
Thus , proving the boxed theorem. The proof includes the Fourier analysis on a finite abelian group detection, conversion to an integer arithmetic progression, and the length bookkeeping that produces the double logarithm.
New to topics? Read the docs here!