Abstract
We prove a multiple coloured Tverberg theorem and a balanced coloured Tverberg theorem, applying different methods, tools and ideas. The proof of the first theorem uses a multiple chessboard complex (as configuration space) and the Eilenberg–Krasnoselskii theory of degrees of equivariant maps for non-free group actions. The proof of the second result relies on the high connectivity of the configuration space, established by using discrete Morse theory.
Export citation and abstract BibTeX RIS
Proposition 4.2 was supported by the Russian Science Foundation, grant no. 16-11-10039. R. Živaljević has been supported by the Ministry of Education, Science and Technology of the Serbian Republic (a grant of the Mathematical Institute of the Serbian Academy of Sciences). |
§ 1. Introduction
We begin with two very early predecessors of the results discussed in this paper.
It is well known that the sphere is non-embeddable in .
The topological Radon theorem refines this result by stating that for any continuous map of a tetrahedron boundary to the plane, there are two disjoint faces of the tetrahedron whose images intersect. Here a face is defined as the intersection of the tetrahedron with a support plane. More precisely,
- 1)either the images of two opposite edges intersect,
- 2)or the image of a vertex belongs to the image of the opposite face.
The complete graph with five vertices (regarded as the -skeleton of the four-dimensional simplex) is non-embeddable in .
The Van Kampen–Flores theorem strengthens this result by stating that for every continuous map , there are two disjoint edges whose images intersect.
For several decades the generalization of these results to higher dimensions and to intersections of different multiplicity has been one of the central research themes in topological combinatorics; see [1], [2] and § 2 for an introduction and an overview.
These generalizations are called Tverberg type theorems and generalized Van Kampen–Flores theorems. They are usually stated as assertions about continuous maps or, more generally, about maps , where is a subcomplex. (Throughout the paper, we write for the -dimensional simplex with vertices .)
For example, the topological Tverberg theorem [3]–[5] for the plane and intersection multiplicity asserts that for every continuous map , there are four pairwise disjoint faces of whose images have a common point. This result can be restated in terms of maps , where is the complete graph with vertices; see [6], [7] and the survey [8], Theorem 2.3.2.
Another predecessor of our first new result (Theorem 1.1) is the following. Consider a -dimensional simplex whose vertex set is coloured by five colours: the vertices labeled by and have the same colour. A coloured topological Tverberg-type theorem (see [9], [1], § 6.5, and [10]) asserts that for every continuous map from to the plane, there are four pairwise disjoint faces such that their images have a common point and each is rainbow-like (that is, contains at most one vertex from each of the pairs , ).
The following multiple coloured Tverberg theorem is our first main result. A new feature of Theorem 1.1 is that some vertices can appear twice as vertices of different faces .
As above, it is instructive to visualize the set of vertices as coloured by four colours: are red, are blue, are green, and the last vertex is white.
Theorem 1.1. Let be a -dimensional simplex with vertex set .
Then for every continuous map , there are four faces , , , of such that
1. their images intersect:
2. every face is rainbow, that is, contains at most one vertex in each pair , ;
3. each vertex of occurs in the faces at most once when is odd and at most twice when is even.
Our second main result is the balanced coloured Tverberg theorem (Theorem 1.2). It is an extension of the coloured Tverberg theorem of type B [11], [12] and a coloured analogue of the balanced Van Kampen–Flores theorem, Theorem 1.2 in [13] (see also [14] for a short proof).
Theorem 1.2. Assume that is a prime power and . Define integers and by the condition
or, more explicitly,
Put and consider a simplex , assuming that there is a partition of its set of vertices into colour classes: , where for each . Then for every continuous map
there are disjoint faces of such that and
The paper is organized as follows. In § 2 we describe the above two theorems in the context of recent progress concerning Tverberg-type results. This is followed by the proofs.
The proof of Theorem 1.1 (§ 3) is based on the Eilenberg–Krasnoselskii theory of degrees of equivariant maps for non-free group actions; see the monograph [15] for a detailed presentation of this theory.
The proof of Theorem 1.2 (§ 4) is based on the high connectivity of the configuration space (Proposition 4.2). This connectivity result is established by using discrete Morse theory and the methods from our papers [16]–[18].
For the reader's convenience, we briefly outline the basic facts and ideas of discrete Morse theory in § 5. A more detailed presentation can be found in [19]. The fundamental comparison principle for equivariant maps between spaces with non-free group actions is stated in § 6.
§ 2. A brief overview of the Tverberg theorem and related results
The following result is known as the topological Tverberg theorem.
Theorem 2.1 (see [3], [4]). Assume that is a prime power. Then for every continuous map
there are disjoint faces such that
Let be a geometric realization of a finite simplicial complex. Following [20]–[24], we say that a continuous map is an almost -embedding if for every -tuple of pairwise disjoint faces of . If there are no almost -embeddings of in , we say that is not almost -embeddable in . In these terms, Theorem 2.1 asserts that
The following four assertions illustrate results of coloured Tverberg type (see [2], [25] for more details and references):
By definition, is the complete multipartite simplicial complex obtained as a joint of zero-dimensional complexes (finite sets). For example, is the complete bipartite graph such that each of the 'red vertices' is connected with each of the 'blue vertices'. In (7) we recognize a statement closely related to the non-planarity of , while (8) says that the -dimensional complex admits no affine almost -embedding in .
By a colouring of vertices of a simplex by colours we understand a partition into monochromatic subsets . A subset is called a rainbow simplex or a rainbow face if for all . If the cardinality of is , then is precisely the subcomplex of all rainbow faces in . A coloured Tverberg theorem is any statement of the form
and the problem is to specify the values of , , and for which (11) holds.
We refer the reader to [11], [12], [26]–[28] and [1], [34], [8], [2] for more general results, proofs, history and applications of monochromatic and coloured Tverberg theorems.
Following [2], § 21.4, we classify coloured Tverberg theorems as theorems of type A, B, or C depending on whether , , or , where is the number of colours and is the dimension of the target space.
The main difference between types A and B is that must satisfy the inequality in case B, while there are no a priori restrictions in case A.
In terms of this classification, (8) and (10) are topological Tverberg theorems of type A, while (7) and (9) are topological Tverberg theorems of type B.
The following results (Theorems 2.2 and 2.3) are the main representatives of these two classes of coloured Tverberg theorems. In particular, (7), (9) and (10) are their simple corollaries.
Note that when is divisible by , our second main result (Theorem 1.2) becomes a coloured Tverberg theorem of type B (Theorem 2.3).
Theorem 2.2 (type A; see [26]). Suppose that is a prime and . Then the complex , which is the join of copies of the zero-dimensional complex and a singleton, is not almost -embeddable in .
Theorem 2.3 (type B; see [12], [34]). Suppose that is a prime power, , and is an integer such that . Then the complex , which is the join of copies of the zero-dimensional complex , is not almost -embeddable in .
Remark 2.4. To simplify the notation and presentation, we do not distinguish between the -dimensional (geometric) simplex and the abstract simplicial complex spanned by vertices (). Therefore, subsets are interpreted as faces of . For , we have , where is the cardinality of .
2.1. The multiple Tverberg theorem
Assertion (8) was obtained by Bárány and Larman [29]. It says that every 9-tuple of points on the plane evenly coloured by three colours can be partitioned into three 'rainbow triangles' with non-empty intersection.
It is currently unknown whether or not the following non-linear (topological) version of (8) holds:
Assertion (12) clearly follows from the stronger assertion
However, is also unknown whether or not (13) is true, and we suspect that this is not the case.
The following multiple coloured Tverberg theorem is a restatement of Theorem 1.1. It says that (13) holds for all continuous maps satisfying an additional (3-to-2)-constraint.
Definition 2.5. A function that glues together the last two points of is called a (3-to-2)-map. More generally, a simplicial map is a (3-to-2)-map if it glues together two points in each copy of the -element set .
Theorem 2.6. Let be a -dimensional simplicial complex with ten vertices divided into four colour classes, and let be a map admitting a factorization for some , where
is a (3-to-2)-map in the sense of Definition 2.5. Then there are four pairwise disjoint simplices (four pairwise disjoint rainbow simplices) , , , in such that
In other words, for any map , the composite is not an almost -embedding of in .
Therefore (12) holds for a special class of non-linear maps.
Corollary 2.7. Assume that is a continuous map admitting a factorization
for some , where is a (3-to-2)-map. Then there are three disjoint triangles , , in such that
2.2. The balanced coloured Tverberg theorem
Our balanced coloured Tverberg theorem (Theorem 1.2) can be regarded as an extension of the coloured Tverberg theorem of type B (Theorem 2.3) to the following theorem, which is referred to as the balanced extension of the generalized Van Kampen–Flores theorem.
Theorem 2.8 ([13], Theorem 1.2). Suppose that is a prime power, , and for some integers and , . Then for every continuous map , there are pairwise disjoint faces of such that , where for and for .
When is divisible by , that is, and for all , Theorem 2.8 becomes the generalized Van Kampen–Flores theorem; see [30]–[32].
The balanced coloured Tverberg theorem (Theorem 1.2) can now be described as a relative of Theorem 2.8 and a balanced extension of Theorem 2.3.
§ 3. Proof of the multiple coloured Tverberg theorem
In accordance with the configuration space/test map scheme [2], [1], [33], [34], the first step of the proof of Theorem 2.6 is a standard reduction to a problem of equivariant topology.
Beginning with a continuous map , we define the associated configuration space as a deleted join
where is the standard chessboard complex of all arrangements of mutually non-attacking rooks on a -chessboard.
The test map which tests whether or not a simplex satisfies (14) is defined as a -equivariant map
where is the diagonal (-dimensional) subspace and is the standard -dimensional representation of . (Throughout the paper, denotes the symmetric group.)
Thus, the existence of a 4-tuple satisfying (14) is equivalent to the existence of zeros of the -equivariant map (16).
For the next step, we need to use a multiple chessboard complex defined as the complex of all rook placements on a -chessboard with at most two rooks in the second column and at most one rook in any row and in the first column. (Here we adopt Cartesian notation, that is, the -chessboard is regarded as the Cartesian product with two columns and four rows.)
Multiple chessboard complexes were studied in [35], and our notation follows that paper. In particular, the vectors (resp. ) describe the restrictions on the number of rooks in the rows (resp. columns) of the -chessboard.
Lemma 3.1. Let be a map admitting a factorization for some map , where
is a (3-to-2)-map in the sense of Definition 2.5. Then the equivariant map (16) admits a factorization into -equivariant maps, as shown in the following commutative diagram:
where is the multiple chessboard complex defined above and is an epimorphism.
Proof. The proof is by elementary inspection. Note that the map which induces in the diagram (17) can be described informally as the map that contracts two columns of the -chessboard into one column of the -chessboard.
Summarizing the first two steps, we observe that the proof of Theorem 2.6 will be complete if we show that the -equivariant map always has a zero.
3.1. Equivariant maps from
The -representation under consideration can be described as with the action induced by the symmetries of a regular tetrahedron centred at the origin. If the map has no zeros, then there is a -equivariant map
where is the boundary sphere of the simplex . However, this is ruled out by the following theorem.
Theorem 3.2. Let be the Klein four-group. Let be the multiple chessboard complex based on a -chessboard, where and , and let be the boundary of the simplex spanned by the vertices of . Both and are -spaces, where the group action in the first case permutes the rows of the chessboard , and in the second case it permutes the vertices of the simplex . Then there is no -equivariant map
where the action of on the join is diagonal.
Theorem 3.2 will be proved by arguments using the notion of degree of an equivariant map. These arguments can be traced back to Eilenberg and Krasnoselskii; see [15] for a thorough treatment and § 6 for a statement of one of the main theorems.
Before proving Theorem 3.2, we describe a convenient geometric model of the complex . Recall that the Bier sphere of a simplicial complex is the deleted join of and its Alexander dual ; see [1] for more details.
Lemma 3.3. The multiple chessboard complex is a triangulation of a -sphere. More explicitly, there is an isomorphism , where is the -skeleton of the tetrahedron and is the Bier sphere associated with a simplicial complex .
Proof. This follows directly from the observation that the subcomplexes of generated by the vertices in the second and first columns of the chessboard are and , respectively.
The following lemma describes the structure of as a -space, where is the Klein four-group.
Lemma 3.4. As a -space, the sphere is homeomorphic to the regular octohedral sphere centred at the origin and the generators are the rotations by around the axes connecting the pairs of opposite vertices of the octahedron.
More explicitly, let be the one-dimensional -representation characterized by the conditions and (we also define and in a similar way) and let , , be the corresponding zero-dimensional -spheres. Then the complex is -isomorphic to the -sphere with induced -action.
Remark 3.5. Here is a geometric interpretation (visualization) of the -isomorphism . The complex and its dual can be geometrically realized as the tetrahedron and its polar body . If both tetrahedra are inscribed in the cube , the geometric realization of can be regarded as a triangulation of the boundary of the cube.
Lemma 3.6. As a -space, the boundary sphere of the tetrahedron is isomorphic to the octahedral sphere described in Lemma 3.4. Moreover, there is a radial -isomorphism .
Summarizing, we can see that the -spheres studied in this section have two combinatorial interpretations ( and ) and three equivalent geometric incarnations (the boundary of the cube, the boundary of the tetrahedron and the boundary of the octahedron).
3.2. Completion of the proof of Theorem 3.2
Proposition 3.7. Let be an arbitrary -equivariant map. Then
Proof. It follows from Theorem 6.1 that for any equivariant maps between these spaces. Here we use the fact that is a topological manifold. Note that the inequality (20), which is necessary for Theorem 6.1 to be applicable, becomes an equality in view of the decomposition (19).
Hence it suffices to produce a map of odd degree. We know that and are -isomorphic -dimensional spheres. Taking as the -isomorphism, we obtain .
Proof of Theorem 3.2. We have
Suppose that there is a -equivariant map . Let be the inclusion map and let be the composite of these maps.
The map is homotopically trivial since for every . However, the degree of is odd by Proposition 3.7. Contradiction.
Remark 3.8. It has been pointed out by a referee that an alternative and somewhat shorter proof of Theorem 3.2 can be obtained by using Volovikov's theorem [5], [31] instead of Theorem 6.1 as in the proof of Theorem 1.2 (see § 4). Indeed, the -connectivity of is an immediate consequence of the homeomorphism (Lemma 3.3). Moreover, the action of on and has no fixed points since, by Lemmas 3.4 and 3.6, there are -homeomorphisms
§ 4. Proof of the balanced coloured Tverberg theorem
Following the configuration space/test map scheme [1], [2], we describe the configuration space used in the proof of Theorem 1.2.
Definition 4.1. Put and . The configuration space of all -tuples of disjoint rainbow simplices satisfying the restrictions listed in Theorem 1.2 is the simplicial complex whose simplices are labelled by , where
– is a partition such that ;
– each is a rainbow set (a rainbow simplex) and, in particular, for every ;
– the number of simplices with does not exceed .
Note that the dimension of the simplex is . Moreover, the facets of can be formally obtained by deleting an element of one of the sets and adding this element to .
Proposition 4.2. The configuration space is -connected.
Let us explain briefly how Theorem 1.2 can be deduced from Proposition 4.2. This standard argument was used, for example, in the proof of the topological Tverberg theorem; see [1], § 6, or [2], [5].
Suppose that Theorem 1.2 is false. Then there is a -equivariant map
whose image is disjoint from the diagonal . This contradicts Volovikov's theorem [5], [31] since is -homotopy equivalent to a sphere of dimension while the configuration space is -connected.
Proof of Proposition 4.2. We begin by introducing some useful abbreviations.
A set is said to be -full if it contains a vertex of colour . A simplex is said to be -full if each is -full or, equivalently, if . A simplex is said to be -full if it contains (the maximal allowed number) of -sets among the . A simplex is said to be saturated if it is -full and for every .
Saturated simplices are maximal faces of the configuration space . Their dimension is .
Following discrete Morse theory and Theorem 5.1, we shall define a matching for . Given any simplex , we shall either describe a simplex paired with it, or recognize it as a critical (that is, unmatched) simplex.
This will be done stepwise. We shall have 'big' steps, each of which splits into successive small steps. The big steps treat the sets one-by-one, and the small steps treat the colours one-by-one.
Step 1.
Step 1.1. Assume that the vertices of each colour are enumerated as . We put
and match with whenever both of these simplices belong to .
A simplex of type is not matched if and only if it is equal to
This is a zero-dimensional simplex. It will stay unmatched till the end of the matching process.
If a simplex of type is unmatched, then is either -full, or and is -full.
Step 1.2. Put
and match with whenever both of these simplices belong to and were not matched at Step 1.1.
- –If a simplex of type is unmatched, then is either -full, or and is -full. Such simplices are said to be 'Step 1.2 – Type 1'-unmatched.
- –If a simplex of type is unmatched, then and is -full (these conditions are necessary but not sufficient). The reason is that in this case belongs to , but may have been matched at Step 1.1. Such simplices are said to be 'Step 1.2 – Type 2 '-unmatched.
In what follows, we use similar abbreviations. 'Step – Type 1' means that one cannot move an element coloured by from to . 'Step – Type 2' means that one cannot move an element coloured by from to .
Step 1.3 and subsequent steps (up to Step ) follow by analogy.
Summarizing, we can make the following conclusion.
Lemma 4.3. Except for the unique zero-dimensional unmatched simplex, if a simplex is unmatched after Step 1, then
1) either ,
2) or and is -full.
Proof. This follows directly by analysing the matching algorithm at small steps.
Step 2. We now treat for the simplices that remain unmatched after Step 1.
Step 2.1. We put
and match with whenever both of these simplices belong to and have not been matched at Step 1.
- –If a simplex of type is not matched now, then either and is -full, or is -full. Such simplices are called 'Step 2.1 – Type 1'-simplices.
- –If a simplex of type is unmatched, then it is -full and . Such simplices are called 'Step 2.1 – Type 2'-simplices.
Step 2.2. We put
and match with whenever both of these simplices belong to and were not matched earlier, that is, at Step 1 or Step 2.1.
Step 2.3 and subsequent steps (up to Step ) follow by analogy.
Summarizing, we reach the following conclusion.
Lemma 4.4. Except for the unique zero-dimensional unmatched simplex, if a simplex remains unmatched after Step 2, then it is also unmatched after Step 1 (and satisfies Lemma 4.3). Moreover,
1) either ,
2) or and is -full.
Steps , , and follow by analogy.
Lemma 4.5. The numbers are well defined for all the steps .
Proof. Indeed, for , the set contains at least elements. (Here we use the fact that and for each .) The entries are either not in , or (by construction) they are the smallest consecutive entries in . Their total number is strictly less than .
Special attention should be paid to the last Step .
First of all, we observe that the following is already known (by construction).
Lemma 4.6. Except for the unique zero-dimensional unmatched simplex, if a simplex is unmatched after Step , then
1) either ,
2) or for some , and is -full.
Proof. This follows from Lemma 4.4 and its analogues for Steps .
Step . We now turn our attention to .
Step . We put
The set may be empty for , so that is undefined.
This means that is -full. Such simplices remain unmatched and are said to be 'Step – Type 3'-unmatched.
If is well defined, we proceed in the standard way: match and if these two simplices belong to and have not been matched before.
Step . We put
Once again, if this number is undefined, then is a -full simplex and we leave it to be 'Step – Type 3' unmatched.
Otherwise we proceed in the standard way.
Step and subsequent steps (up to Step ) follow by analogy.
Summarizing, we make the following conclusion.
Lemma 4.7. Except for the unique zero-dimensional unmatched simplex, if a simplex remains unmatched after Step , then it is saturated.
Proof. We have for all by Lemma 4.6.
If a simplex is such that for some , then some colour does not occur in . Let be the smallest index of a missing colour. It follows that this simplex was matched at Step since is well defined and can be added to .
At every Step , the simplex is either of Type 1, or of Type 2, or (this can occur only at Step ) of Type 3. If it was of Type 2 at least once (it does not matter at which step), then the same lemma implies that it is -full, hence saturated.
If the simplex was always of Type 1 at Steps and is not saturated, then for all . Since , it is saturated.
It remains to prove that the matching is acyclic.
Assume that we have a gradient path
For every simplex , we consider the sequence of numbers
These are all the numbers listed in the same order as they appear in the matching algorithm. When is undefined, we let it be .
Lemma 4.8. Along the path is strictly decreasing with respect to the lexicographic order. Hence the matching is acyclic.
Proof. This follows from a case-by-case analysis.
First of all, it suffices to consider only three-step paths:
1. Suppose that means adding a colour to and means removing a colour from . Then
- 1)either is matched with some -dimensional simplex obtained by removing colour from and the path terminates here,
- 2)or is matched before Step .
2. Suppose that means adding a colour to and means removing a colour from . Then is matched before Step .
3. Suppose that means adding a colour to and means removing a colour from with . Then
- 1)either is matched by adding colour to ,
- 2)or is matched before Step .
4. Suppose that means adding a colour to and means removing a colour from with . Then
- 1)either is matched with some -dimensional simplex obtained by removing colour from and the path terminates here,
- 2)or is matched before Step .
This completes the proof of Proposition 4.2 and that of Theorem 1.2.
§ 5. Appendix 1. Discrete Morse theory
By definition [19], a discrete Morse function on a simplicial complex is an acyclic matching on the Hasse diagram of the partially ordered set .
Here are some details. The -dimensional simplices (-simplices for brevity) of a simplicial complex are denoted by , , , , . A discrete vector field is a set of pairs (called a matching) such that
(a) each simplex of the complex occurs in at most one pair;
(b) in each pair , the simplex is a facet of ;
(c) the empty set is not matched, that is, implies that .
The pair can be informally thought of as a vector in the vector field . Therefore it is often denoted by or (in this case and are informally referred to as the beginning and the end of the arrow ).
Let be a discrete vector field. A gradient path in is a sequence of simplices
satisfying the following conditions:
1) belongs to for each ;
2) is a facet of for each ;
3) for each .
A path is closed if . A discrete Morse function (DMF for brevity) is a discrete vector field without closed paths.
The critical simplices of a discrete Morse function are those simplices in the complex that are not matched. The Morse inequality [19] implies that critical simplices cannot be completely avoided.
In the present paper we use the following theorem.
Theorem 5.1 [19]. Assume that a discrete Morse function on a simplicial complex has a single zero-dimensional critical simplex and that all other critical simplices have the same dimension . Then is homotopy equivalent to a wedge of -dimensional spheres.
If all critical simplices, except for , are of dimension at least , then the complex is -connected.
§ 6. Appendix 2. Comparison principle for equivariant maps
The following theorem was proved in [15], § 2, Theorem 2.1. Note that the hypothesis that the -fixed point sets are locally -connected for holds automatically when is a representation sphere. Therefore, in this case it suffices to show that is globally -connected or, equivalently,
Theorem 6.1. Let be a finite group acting on a compact topological manifold and on a sphere of the same dimension, let be a closed invariant subset, and let be the orbit types in . Assume that is a globally and locally -connected set for all , where . Then the following relation holds for every pair of -equivariant maps that are equivariantly homotopic on :
The authors are grateful to A. Skopenkov for valuable remarks.