Review on rationality problems of algebraic k-tori
Youngjin Bae

TL;DR
This paper reviews the rationality issues of algebraic k-tori, linking them to invariant fields and Noether's Problem, and discusses methods to determine their rationality.
Contribution
It provides a comprehensive overview of rationality problems for algebraic k-tori, including character group analysis and numerical approaches for rationality determination.
Findings
Rationality of k-tori is equivalent to the rationality of their invariant function fields.
Character groups are useful tools in studying k-tori rationality.
Numerical methods can assist in determining the rationality of algebraic k-tori.
Abstract
Rationality problems of algebraic k-tori are closely related to rationality problems of the invariant field, also known as Noether's Problem. We describe how a function field of algebraic k-tori can be identified as an invariant field under a group action and that a k-tori is rational if and only if its function field is rational over k. We also introduce character group of k-tori and numerical approach to determine rationality of k-tori.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPolynomial and algebraic computation · Commutative Algebra and Its Applications · Algebraic Geometry and Number Theory
Review on rationality problems of algebraic k-tori
Youngjin Bae
Abstract
Rationality problems of algebraic are closely related to rationality problems of the invariant field, also known as Noether’s Problem. We describe how a function field of algebraic can be identified as an invariant field under a group action and that a is rational if and only if its function field is rational over . We also introduce character group of and numerical approach to determine rationality of .
Contents
1 Introduction
Let be a field and is a finitely generated field extension of . is called rational over k or k-rational if is isomorphic to where are transcendental over and algebraically independent. There are also relaxed notions of rationality. is called stably k-rational if is for some transcendental and algebraically independent . is called if for some pure transcendental extension .
The Noether’s Problem is the question of rationality of the invariant field under finite group action. For example, if and and acts on as permutation of variables (i.e. fixes , and ), then the invariant field is .
Example 1.1
* and , acting on as permutation of variables. Let , are coprime. We have*
[TABLE]
*By observing that , we have and .
Therefore, are symmetric, field of fractions (quotient field) of . It is easy to see that is isomorphism, where*
[TABLE]
Therefore, and , .
We can also consider case of acting on both of coefficients and variables.
Example 1.2
* and . Suppose acts on by permuting and as complex conjugation on coefficients.
For example, . Then, , is .*
Proof. For , where are coprime, and are also coprime. From , we have and . Thus, is quotient field of where .
Define a map as
[TABLE]
[TABLE]
[TABLE]
The coefficients of are real numbers. This is because, if we let , we have that
[TABLE]
.
Therefore, , . It is easy to see that is actually isomorphism, , and .
Another perspective to view this change of variables is identifying the field with rational function field of algebraic . (see Example 2.5 and Example 2.6)
2 Algebraic
Let be a field. Then is n-dimension affine space over the field , simply with usual vector space structure on it. A subset of is an algebraic k-variety (k-variety in short) if it is a set of zeros of a system of equations with variables over . The ideal of polynomials that vanish on every points of will be denoted by . The coordinate ring of a variety is defined to be the quotient
[TABLE]
Projective varieties can be similarly defined as the set of zeros of a system of homogeneous equations. Projective is defined as set of lines passing the origin in .
If are varieties, a map is called regular if it can be presented as fraction of polynomials , where does not vanishes in . A map is called rational if it is regular on Zariski open dense set. (Formally, a regular map is defined as an equivalence class of pairs where is Zariski open subset of . See [1]) Let be a variety, is the rational function field, or function field in short, the set of rational maps . For example, if is an affine variety over algebraically closed field , is quotient field of .
Example 2.1
*Let be a variety over .
Then, and .*
Two varieties are isomorphic (resp. birationally isomorphic) if there is a bijective regular map (resp. rational map) and its inverse is also regular (resp. rational).
A variety in is an algebraic group if it has a group structure on it, where the group operation and inversions are regular maps. (i.e. and are regular)
Algebraic , or algebraic , is a special type of algebraic group over . We call an algebraic group as when it is isomorphic to some power of multiplicative group over , the algebraic closure of .
Definition 2.1** (Multiplicative Group)**
Let be a field, the multiplicative group is algebraic group in , defined as , with operation of
Example 2.2
* is the curve on the real affine plane. It is isomorphic to as a group. is group isomorphism.*
As field changes, same system of equations can define different varieties. For instance, the equation in previous example defines in , which is different from . If is a field and is its algebraic closure, an irreducible variety over entails the ring of equations, . If happens to be in (ring of polynomials over ), we can define , a variety over defined by equations in . This can be viewed as restriction of scalar. Extension of scalar can be defined similarly.
Definition 2.2** (Algebraic -tori)**
Let be a field with algebraic closure . If is an algebraic group over , it is if and only if
**
for some . The is called dimension of .
Example 2.3
* is one dimensional . This is because .*
From now, let be the one dimensional torus over . There are two one-dimensional -tori, one can be recognized as , the other one can be recognized as as a group.
Example 2.4
The norm one torus is a real algebraic group in , defined by equation , and operation such that
[TABLE]
*Indeed, is isomorphic to as a group.
Also, is isomorphic to as algebraic group. The map *
[TABLE]
is isomorphism. Therefore, is one dimensional real torus.
If is a , is called split over K if it satisfies for some extension and some . For instance, is split over , is not.
It is easy to find split torus such as or , being another torus. Also, for any integer , is -dimensional . Meanwhile, there are also some non-trivial(not a product of low-dimensional torus) torus.
Example 2.5
Let be a real algebraic group in , defined as
[TABLE]
Alternatively,
[TABLE]
and operation such that
[TABLE]
Which is compatible with complex multiplication of
[TABLE]
Moreover, is isomorphic to , by sending
[TABLE]
Therefore, is 2-dimensional .
By tracking the function fields of and , we have the same trick of change of variables as in Example 1.2.
Example 2.6
In the previous example, the coordinate ring of is
[TABLE]
where and . The function field of is
[TABLE]
Let acts on as in Example 1.2. Observe that the coordinate ring of is and the function field of is (note that actions on and are equivalent through the isomorphism). In short, we have that
[TABLE]
Therefore, when action on is given, we can convert the rationality problem to the rationality problem of , the function field of . In this sense, the following definition and theorem are natural.
Definition 2.3** (Rationality of )**
*We say that a variety over is rational if, equivalently,
(1) is birationally isomorphic to for some .
(2)
If is Galois extension, a is if it is rational as a -variety . If is algebraically closed, there is unique -dimension tori . Since the function field of is , thus is -rational.
Theorem 2.1
*The following two problems are equivalent.
(1) The rationality problem of dimensional
*(2) The rationality problem of invariant field
where and .
There is a connection between the action on and , connecting the two rationality problems given in the previous theorem. To be specific, the character group of determines both the action and uniquely.
3 Character group of
Definition 3.1** (Character group of )**
Let be . Then , the character group of is the set of algebraic group homomorphisms(a regular map preserving the group structure) from to , denoted by or .
The character group of has a group structure defined by component-wise multiplication. Also, if is split over for finite Galois extension of base field , acts on . Moreover, it is known that is torsion-free -module(i.e. isomorphic to for some ). Therefore, is a (a free with -action).
Example 3.1
If is multiplicative group of , then is set of regular functions such that for . Since is a rational function, it is a meromorphic function over . Also, we have , which implies 0 is the only point where can have zeros or poles. Therefore, for some . If we write a function as , we have
[TABLE]
as a group. acts trivially on .
In general, if is algebraically closed, the character group of is
Example 3.2
Let be the 2-dimension in Example 2.5. Then, the character group of is
[TABLE]
Let , , then we have the natural extension of to
[TABLE]
Observe that the complex conjugation , exchanges and , thus acting on as matrix .
It is known that when a action (as -linear function) on is given, there exists unique -dimensional which has the given as its character group. Furthermore, there are conditions of corresponding to the rationality conditions of and of invariant fields.
4 Flabby resolution and numerical approach
This section contains many results in [2]. Let be a group and be a ( as group and has -linear action on it). is called a permutation G-lattice if for some subgroups of (equivalently, there exists a -basis of such that acts on as permutation of the basis). is called stably permutation G-lattice if for some permutation and . is called invertible if it is a direct summand of a permutation -lattice, i.e. for some permutation -lattice and .
Definition 4.1** (1st Group Cohomology)**
Let be a group and be a -lattice. For and , let be acting on . The first group cohomology is a group defined as
[TABLE]
where and
simply implies that if satisfies , then there exists such that . is called coflabby if .
Definition 4.2** (-1st Tate Cohomology)**
Let be finite group of order n and be a -lattice. The -1st group cohomology is a group defined as
[TABLE]
where
[TABLE]
,
[TABLE]
Similarly, is called flabby if . It is clear that a is rational if and only if is permutation -lattice. Thus, the rationality problems of and invariant fields can be reduced into problem of finding permutation -lattice(equivalent to find finite subgroup of . However, this problem is not solved yet, even though there are many results in weakened problems.
Let be the category of all -lattices and be the category of all permutation -lattices. Define equivalence relation on by if and only if there exist such that . Let be equivalence class containing under this relation.
Theorem 4.1
(Endo and Miyata [3, Lemma 1.1], Colliot-Thélène and Sansuc [4, Lemma 3])* For any -lattice , there is a short exact sequence of -lattices where is permutation and is flabby.*
In the previous theorem, is called the flabby class of , denoted by .
Theorem 4.2
(Akinari and Aiichi [2, 17pp])* If is stably permutation, then . If is invertible, is invertible.*
It is not difficult to see that
[TABLE]
Furthermore, it is true that
[TABLE]
In [2], they gave the complete list of stably permutation lattices for dimension 4 and 5 by computing for finite subgroup of , which is equivalent to classifying stably rational tori. Thus, the rationality problems for low dimensional can be resolved by finding conditions which can determine a stably permutation is permutation or not.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] Robin Hartshorne Algebraic Geometry . Springer, New York, 24-25, 1977.
- 2[2] Akinari Hoshi, Aiichi Yamasaki Rationality Problem for Algebraic Tori (Memoirs of the American Mathematical Society) American Mathematical Society, 2017.
- 3[3] S.Endo, T.Miyata On a classification of the function fields of algebraic tori Nagoya Math, 85-104, 1975.
- 4[4] J.-L. Colliot-Thélène, J.-J. Sansuc La R-équivalence sur les tores 175-229, 1977.
