The zero locus of a nonzero multivariate polynomial is null #
A nonzero polynomial in finitely many variables over a measurable integral domain vanishes on a null set of any finite product of σ-finite measures without atoms; in particular, a nonzero real polynomial vanishes on a Lebesgue-null set. Determinants and minors are polynomials in the entries of a matrix, so this shows that generic polynomial conditions, such as nonsingularity, hold almost everywhere.
Main declarations #
MvPolynomial.measurable_eval— evaluation is measurable in the coordinates.MvPolynomial.measurableSet_setOfPred_eval_eq_zero— the zero locus of a multivariate polynomial is measurable.MvPolynomial.measure_pi_setOfPred_eval_eq_zero— the zero locus of a nonzero multivariate polynomial is null for a finite product of σ-finite measures without atoms.MvPolynomial.volume_setOfPred_eval_eq_zero— the same for the productvolume.
Evaluating a multivariate polynomial is measurable in the coordinates: a polynomial is built from finitely many coordinate projections by addition and multiplication.
The zero locus of a multivariate polynomial is measurable.
The zero locus of a nonzero multivariate polynomial is null for a finite product of σ-finite measures without atoms.
The zero locus of a nonzero multivariate polynomial is volume-null when volume is a
σ-finite measure without atoms; for instance, for real polynomials and Lebesgue measure.