It is undecidable. If you could solve this, you could also solve Hilbert's 10th problem.
Suppose we have an algorithm solving your problem for all n. Given a polynomial pinmathbb[x1,dots,xn], let's decide whether it has integer solutions. If p is constant, this is trivial. Otherwise we can find z0inmathbbZn such that |p(x)|>1 for all xinz0+[0,1]n. Let's work with a polynomial f(x)=p(x+z0) rather than p. It satisfies |f(x)|>1 for all xin[0,1]n.
Let g(x)=x1(x1−1)x2(x2−1)dotsxn(xn−1). Apply our algorithm to the inequality
r(x):=(f(x)2−1)cdotg(x)<0.
If it says that S+mathbbZn=mathbbRn, then we know that S contains a point from mathbbZn, and this point must a root of f. If it says that S+mathbbZnnemathbbRn, then we know that there is cinmathbbRn such that r(c+z)ge0 for all zinmathbbZn.
This c must belong to mathbbZn. Indeed, suppose that e.g. c1notinmathbbZ. We may assume that all coordinates of c are positive and 0<c1<1. Substitute z=(0,z2,dots,zn) where z2,dots,zn are arbitrary positive integers and conclude that |f(c+z)|le1 for all such z. It follows that f is constant on the hyperplane x1=c1, and the modulus of this constant no greater than 1. This contradicts
the fact that |f|>1 on [0,1]n.
Thus we know that cinmathbbZn, or, equivalently, that r(z)ge0 for all zinmathbbZn. This means that f does not have integer roots except possibly at points where one of the coordinates is 0 or 1. Thus we reduced the problem to the case of n−1 variables and can solve it by induction.
No comments:
Post a Comment