Tarski-Seidenberg theorem


Theorem (Tarski-Seidenberg).

That is, if A⊂ℝn×ℝm is a semialgebraic set, and if π is the projection onto the first n coordinates, then π⁢(A) is also semialgebraic.

Łojasiewicz generalized this theorem further. For this we need a bit of notation.

Let U⊂ℝn. Suppose 𝒜⁢(U) is any ring of real valued functions on U. Define 𝒮⁢(𝒜⁢(U)) to be the smallest set of subsets of U, which contain the sets {x∈U⁢∣f⁢(x)>⁢0} for all f∈𝒜⁢(U), and is closed under finite union, finite intersectionMathworldPlanetmath and complementMathworldPlanetmath. Let 𝒜⁢(U)⁢[t] denote the ring of polynomials in t∈ℝm with coefficients in 𝒜⁢(U).

Theorem (Tarski-Seidenberg-Łojasiewicz).

Suppose that V⊂U×Rm⊂Rn+m, is such that V∈S⁢(A⁢(U)⁢[t]). Then the projection of V onto the first n variables is in S⁢(A⁢(U)).

References

  • 1 Edward Bierstone and Pierre D. Milman, Semianalytic and subanalytic sets, Inst. Hautes Études Sci. Publ. Math. (1988), no. 67, 5–42. http://www.ams.org/mathscinet-getitem?mr=89k:32011MR 89k:32011
Title Tarski-Seidenberg theorem
Canonical name TarskiSeidenbergTheorem
Date of creation 2013-03-22 16:46:13
Last modified on 2013-03-22 16:46:13
Owner jirka (4157)
Last modified by jirka (4157)
Numerical id 5
Author jirka (4157)
Entry type Theorem
Classification msc 14P15
Classification msc 14P10
Related topic SemialgebraicSet
Related topic SubanalyticSet
Defines Tarski-Seidenberg-Łojasiewicz theorem