Toward a model theory for transseries (Q372613): Difference between revisions
From MaRDI portal
Latest revision as of 22:10, 6 July 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Toward a model theory for transseries |
scientific article |
Statements
Toward a model theory for transseries (English)
0 references
9 October 2013
0 references
This is a very lucid and enlightening survey of the state of the art in the model theory of the field of transseries, by three of the main researchers in this fascinating topic. The differential field \(\mathbb T\) of transseries is an ordered field extensions of the real field \(\mathbb R\) and is a kind of universal domain for asymptotic real differential algebra. It has appeared in many different and rich contexts, e.g. Hilbert's sixteenth problem. The main conjecture is that the first-order theory of the ordered differential field \(\mathbb T\) is model-complete and the model-companion of the theory of \(H\)-fields with small derivation. Since the appearance of this paper, the authors have announced further results pointing to the conjecture. I will only mention a quantifier elimination similar to the one conjectured at the end of the present paper, with unary predicates \(\Omega(a), \Lambda(a)\) defined as follows. We have as elements of \(\mathbb T\) the iterated logarithms \(\ell_n\), with \(\ell_0=x\), \(\ell_{n+1}=\log \ell_n\). With these, define \(\Lambda(a) \iff a < \ell_0^{-1} +\ell_0^{-1}\ell_1^{-1}+\ldots+\ell_0^{-1}\ell_1^{-1}\ldots \ell_n^{-1}\) for some natural number \(n\), and \(\Omega(a) \iff a < \ell_0^{-2} +\ell_0^{-2}\ell_1^{-2}+\ldots+\ell_0^{-2}\ell_1^{-2}\ldots \ell_n^{-2}\) for some natural number \(n\), and it turns out that \(\Omega(a) \iff 4y''+ay=0\) for some \(y\neq 0\). The authors take us on a tour of the subject, starting with the field of Laurent series \(\mathbb R((x^{-1}))\) in the variable \(x^{-1}\) as a differential ordered field and aiming at removing its defects of having neither an antiderivative for \(x^{-1}\) (viz. \(\log x\)) nor a reasonable exponentiation function \(f\mapsto \exp(f)\), to the current state of the ambitious conjecture above. The text is accessible to nonspecialists. The authors have succeeded in their aim that only rudimentary knowledge of model theory, valuations and differential fields is necessary. The paper is enhanced with insightful examples, and succeeds to convey the richness of the subject.
0 references
transseries
0 references
Hardy fields
0 references
differential fields
0 references
model-completeness
0 references
NIP
0 references
0 references