The inverse sieve problem for algebraic varieties over global fields (Q1982550): Difference between revisions

Let \(K\) be a global field of degree \(d_K\) and \(\mathcal O_K\) its ring of integers. Writing \(H\) for the absolute multiplicative height function, let \(S\) be a subset of the finite set \[\left\{ x \in \mathcal O_K : H(x) \leq N \right\}^d,\] where \(N > 0\) and \(d > 0\) are integers. If such a subset \(S\) occupies few residue classes modulo different prime ideals of \(\mathcal O_K\), one can expect that \(S\) is either small or it possesses some strong algebraic structure. The paper under review carries out a detailed quantitative analysis of this observation, which is known as the ``inverse sieve problem''. The authors' first result in this direction asserts the existence of a constant \(C = C(d,k,\varepsilon,\alpha,\eta, K)\) for \(0 \leq k < d\) and \(\varepsilon, \alpha,\eta > 0\) such that if \(S\) occupies less than \(\alpha \left| \mathcal O_K/\mathfrak p \right|^k\) residue classes for every prime \(\mathfrak p\) in \(\mathcal O_K\), then at least one of the following two conditions holds: \begin{itemize} \item[1.] (\(S\) is small) \(|S| \lesssim_{d,k,\varepsilon,\alpha,K} N^{d_K (k-1) + \varepsilon}\), \item[2.] (\(S\) is strongly algebraic) There exists a nonzero polynomial of degree at most \(C\) in \(\mathcal O_K[X_1,\dots,X_d]\) and coefficients of height no bigger than \(N^C\) vanishing at more than \((1-\eta) |S|\) points of \(S\). \end{itemize} Further, the authors produce two additional versions of their result with an allowed restriction to an affine or to a projective variety~\(Z\) over~\(K\): that is, \(S\) is assumed to be a subset of~\(Z\) occupying fewer residue classes modulo primes than \(Z\). Then the size of~\(S\) and the constant~\(C\) in the theorem above also depend on the dimension of~\(Z\). The authors' results extend and generalize previous work of \textit{M. Walsh} [Duke Math. J. 161, No. 10, 2001--2022 (2012; Zbl 1357.11090)], as do their methods of proof.