MPI-INF Logo
Campus Event Calendar

Event Entry

What and Who

Effectivity Issues and Results for Hilbert's 17th Problem

Marie-Francoise Roy
Université de Rennes 1
Talk
AG 1, AG 2, AG 3, AG 4, AG 5, RG1, SWS, MMCI  
MPI Audience
English

Date, Time and Location

Thursday, 25 February 2016
16:15
60 Minutes
E1 4
024
Saarbrücken

Abstract

If a polynomial is everywhere non negative, it is a sum of squares of rational fractions (which is the positive solution of Hilbert's 17th problem). This is an example of a certificate for positivity (more precisely non-negativity), i.e. an algebraic identify certifying that the polynomial is non-negative.


But how to construct this sum of squares from a proof of the non negativity?

Artin's initial proof of Hilbert's 17th problem in 1927 was very indirect, and did not produce explicitly the sum of squares. Effective methods have been designed since then (starting from Kreisel and his students) but the degree estimate of the corresponding constructions were primitive recursive functions.

In this talk, we recall the initial proof of Artin and present a new constructive proof for Hilbert's 17th problem, and show that following this construction, the degree of the polynomials in the identity is bounded by an elementary recursive function in the number of variables, and the degree of the polynomial (more precisely a tower of five exponentials). We end with a discussion on what could be hoped for.

Based on joint work with Henri Lombardi and Daniel Perrucci.

Contact

Michael Sagraloff
--email hidden
passcode not visible
logged in users only

Michael Sagraloff, 02/12/2016 10:37
Michael Sagraloff, 02/05/2016 12:07 -- Created document.