Rational approximation to algebraic varieties and a new exponent of simultaneous approximation (Q521620): Difference between revisions

In this paper, the author proves that if \(\zeta = (\zeta_1, \ldots, \zeta_n)\) is a point on an \(n\)-dimensional algebraic variety \(V\) that is suitably well approximated by a vector of rational numbers \((p_1/q, \ldots, p_n/q)\), then \(\zeta\) necessarily lies in the topological closure of \(V \cap {\mathbb{Q}}^n\). This is essentially a consequence of the ``fundamental theorem of transcendence theory'', that nonzero integers have modulus at least one! From this perspective, the author then rederives a number of classical and not-so-classical results from the theory of simultaneous Diophantine approximation, and introduces a new exponent to measure such approximations, denoted \(\chi_n (\zeta)\). The second half of the paper is devoted to proving properties of \(\chi_n (\zeta)\) and, in particular, to comparing it to classical exponents like \(\omega_n (\zeta)\).