×

The algebraic groups leading to the Roth inequalities. (English. French summary) Zbl 1276.11120

A general ROTH system \(S=(\ell_1,\dots,\ell_n,c(1),\dots,c(n))\) is given by \(n\) linearly independent linear forms \((\ell_1,\dots,\ell_n)\) in \(n\) variables with real algebraic coefficients and \(n\) real numbers \(c(1),\dots,c(n)\) such that \(c(1)+\dots+c(n)=0\) and such that, for each \(\delta>0\), there are only finitely many integral solutions to the system of inequalities \[ |\ell_q|<Q^{-c(q)-\delta}, \quad Q>1, \; 1\leq q\leq n. \] The special case \(n=2\), \(\ell_1(x_1,x_2)=x_1\), \(\ell_2(x_1,x_2)=\alpha x_1-x_2\) where \(\alpha\) is an irrational real algebraic number and \(c(1)=-1\), \(c(2)=1\) is called a classical ROTH system.
To a general ROTH system \(S\) is attached a filtration \(F_S^\cdot V\) over \(\overline{\mathbb Q}\) of the \(\mathbb Q\)-vector space \(V=\mathbb Q x_1+\dots + \mathbb Q x_n\), defined as \[ F_S^{i}V=\sum_{c(q)\geq i}\overline{\mathbb Q} \ell_q \quad (i\in\mathbb R). \] In the case of a classical ROTH system attached to an irrational algebraic number \(\alpha\), this filtered space is denoted by \((\check{V}, F_\alpha^\cdot \check{V})\). The slope of a filtered vector space \((V, F^\cdot V)\) is \[ \mu(V)=\frac{1}{\dim_\mathbb Q V} \sum_{w\in \mathbb R} w \dim_{\overline{\mathbb Q}} {\mathrm{gr}}^w(F^{\cdot} V) \] where \( {\mathrm{gr}}^w(F^{\cdot} V)=F^wV/F^{w+}V\), \(F^{w+}V=\bigcup _{j>w}F^jV\). The filtration is semi stable if \(\mu(W)=\mu(V)\) for any nonzero \(\mathbb Q\)-vector subspace \(W\) of \(V\). The category of finite dimensional vector spaces over \(\mathbb Q\) with semistable filtration of slope zero is denoted by \(C_0^{\mathrm{ss}}(\mathbb Q,\overline{\mathbb Q})\). See Chap. VI, Theorem 2B of [W. M. Schmidt, Diophantine approximation. Berlin etc.: Springer-Verlag (1980; Zbl 0421.10019)]. See also [G. Faltings, Proc. ICM ’94. Vol. I. Basel: Birkhäuser, 648–655 (1995; Zbl 0871.14010)] and [B. Totaro, Duke Math. J. 83, No. 1, 79–104 (1996; Zbl 0873.14019)].
Here is the main result of the paper under review. Let \(\alpha\) be an irrational real algebraic number. If \(\alpha\) is not quadratic, then there exists a fully faithful tensor functor \(\iota\) of the category \({\mathrm{Rep}}_\mathbb Q {\mathrm{SL}}_2\) of finite dimensional representations over \(\mathbb Q\) of the special linear group \({\mathrm{SL}}_2\) of degree \(2\) into the tensor category \(C_0^{\mathrm{ss}}(\mathbb Q,\overline{\mathbb Q})\) such that the functor \(\iota\) commutes with the forgetful tensor functor to the tensor category \({\mathrm{Vec}}_\mathbb Q\) of finite dimensional vector spaces over \(\mathbb Q\) and such that the image of \(\iota\) contains the filtered vector space \((\check{V}, F_\alpha^\cdot \check{V})\) derived from a classical ROTH System.
On the other hand, if \(\alpha\) is quadratic, then there exists a fully faithful tensor functor \(\iota\) of the category \({\mathrm{Rep}}_\mathbb Q T_\alpha\) of finite dimensional representations over \(\mathbb Q\) of a one dimensional anisotropic torus \(T_\alpha\) over \(\mathbb Q\) into the tensor category \(C_0^{\mathrm{ss}}(\mathbb Q,\overline{\mathbb Q})\) such that the group \(T_\alpha(\mathbb Q)\) is isomorphic to the kernel of the norm map of the quadratic number field \(\mathbb Q(\alpha)\) over \(\mathbb Q\), such that the functor \(\iota\) is compatible with the forgetful tensor functor to \({\mathrm{Vec}}_\mathbb Q\) and such that the image of \(\iota\) contains the filtered vector space \((\check{V}, F_\alpha^\cdot \check{V})\).

MSC:

11J68 Approximation to algebraic numbers
11J13 Simultaneous homogeneous approximation, linear forms
14G05 Rational points
18D10 Monoidal, symmetric monoidal and braided categories (MSC2010)
PDFBibTeX XMLCite
Full Text: DOI Numdam

References:

[1] Y. André, Slope filtrations. Confluentes Math. 1 (2009), 1-85 (arXiv:0812.3921v2). · Zbl 1213.14039
[2] A. Borel, Linear Algebraic Groups, Second Enlarged Edition. Graduate Texts in Math. 126, Springer-Verlag, New York, 1991. · Zbl 0726.20030
[3] J.-F. Dat, S. Orlik, and M. Rapoport, Period Domains over Finite and \(p \)-adic Fields. Cambridge Tracts in Math. 183, Cambridge Univ. Press, New York, 2010. · Zbl 1206.14001
[4] P. Deligne and J. S. Milne, Tannakian Categories. In Hodge Cycles, Motives, and Shimura Varieties, Lect. Notes in Math. 900, 101-228, Springer-Verlag, Berlin Heidelberg, 1982. · Zbl 0477.14004
[5] J.-H. Evertse, The subspace theorem and twisted heights. Preprint, 32pp. (http://www.math.leidenuniv.nl/ evertse/publications.shtml) · Zbl 1282.11085
[6] G. Faltings, Mumford-Stabilität in der algebraischen Geometrie. Proceedings of the International Congress of Mathematicians 1994, Zürich, Switzerland, 648-655, Birkhäuser Verlag, Basel, Switzerland, 1995. · Zbl 0871.14010
[7] G. Faltings and G. Wüstholz, Diophantine approximations on projective spaces. Invent. Math. 116 (1994), 109-138. · Zbl 0805.14011
[8] M. Fujimori, On systems of linear inequalities. Bull. Soc. Math. France 131 (2003), 41-57. Corrigenda. ibid. 132 (2004), 613-616. · Zbl 1119.11038
[9] M. Rapoport, Analogien zwischen den Modulräumen von Vektorbündeln und von Flaggen. Jahresber. Deutsch. Math.-Verein. 99 (1997), 164-180. · Zbl 0891.14010
[10] M. Rapoport, Period domains over finite and local fields. In Algebraic Geometry—Santa Cruz 1995, Proc. Sympos. Pure Math. 62, 361-381, Amer. Math. Soc., Providence, RI, 1997. · Zbl 0924.14009
[11] N. S. Rivano, Catégories Tannakiennes. Lect. Notes in Math. 265, Springer-Verlag, Berlin Heidelberg, 1972. · Zbl 0241.14008
[12] W. M. Schmidt, Diophantine Approximation. Lect. Notes in Math. 785, Springer-Verlag, Berlin Heidelberg, 1980. · Zbl 0421.10019
[13] B. Totaro, Tensor products in \(p \)-adic Hodge theory. Duke Math. J. 83 (1996), 79-104. · Zbl 0873.14019
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.