1. Introduction
In the early 20th century, the American mathematician Leonard Eugene Dickson wondered what structure would arise if one axiom in the list of axioms of a division ring was removed. Dickson discovered that there are near-fields which fulfill all axioms of a division ring except one distributive law. In this essay we will be using right near-fields.
Near-vector spaces are less linear than traditional vector spaces. The ones we are interested in were first introduced by André in the paper [1]. There are a few different notions of near-vector spaces, those studied by André, Beidleman [2] and Karzel [3]. André used right near-fields to study near-vector spaces. It was later shown by van der Walt, that an arbitrary finite-dimensional near-vector space can be constructed using a finite number of near-fields [1]. This construction has been used in order to characterize all finite-dimensional near-vector spaces by taking copies of
, for p a prime [4].
In this paper we discuss near-vector spaces constructed over
in terms of their quasi-kernels and regularity. Regularity is a central notion in the study of near-vector spaces. In [1] the Decomposition Theorem states that every near-vector space can be written as a direct sum of maximal regular subspaces. Thus regular subspaces are considered the building blocks for near-vector spaces. The importance of the Decomposition Theorem is in the decomposition of complex structures into small, simpler structures which are convenient to work with.
André’s near-vector spaces have various applications in Finite Linear Games and Cryptography to mention a few.
This paper is organized in three chapters. In Chapter 2 we give basic definitions and examples which lead to the concept of a near-vector space. In Chapter 3 we show how near-vector spaces are constructed over
using a result by van der Walt. We determine the quasi-kernels
of our near-vector spaces and decompose them into maximal regular subspaces using the Decomposition Theorem. We give necessary conditions needed to show when two near-vector spaces are isomorphic (Lemma 3.8). This is something new that has not been provided in any literature before. We observed that if
is a near-vector space of dimension n, i.e.
, then the number of
’s in the Decomposition Theorem is less or equal to n (Lemma 3.11). A proof was provided. In Chapter 4 we look at an application of near-vector spaces over
to finite linear games and consider some examples.
2. Preliminary Material and Examples
This chapter focuses on the fundamentals of near-vector spaces.
2.1. Basic Definitions
Recall the following definitions.
Definition 2.1. [5] A set V is said to be a right vector space over a division ring F, if
is an abelian group and, if for each
and
, there is a unique element
such that the following conditions hold for all
and all
:
a)
,
b)
,
c)
,
d)
.
The elements of V are called vectors, whereas elements of F are called scalars. And the operation that acts on (or combines) a scalar
and a vector v to form the vector
is called scalar multiplication.
Remark 2.2.
a) A vector space V is said to be closed under both operations, by this we mean:
i) Closed under addition: If
then
.
ii) Closed under multiplication: If
then
.
b) A division ring F can be regarded as a set of endomorphisms of V by defining
for all
and
.
c) If
is a vector space, then for every
and for each
, there exists
such that
.
d) If the scalars are real, then V is called a real vector space while if the scalars are complex, V is called a complex vector space.
The following are examples of (real) vector spaces.
Example 2.3. For
,
, is a vector space with usual (point-wise) addition and multiplication.
Example 2.4. For
,
, the set of polynomials of degree at most n is a vector space.
Example 2.5. The set
of all real-valued functions defined on the real line with operations
and
,
for all
, is a vector space over
.
Definition 2.6. [6] Let V be a vector space over F. Let V' be a subset of V. Then V' is called a subspace of V if it is itself a vector space with the induced operations of V.
Before we define what a near-vector space is, we need the following.
Definition 2.7. [7] A right near-ring is a triple
which satisfies:
a)
is a group;
b)
is a semigroup;
c)
for all
.
N is a near-field if
is also a group.
2.2. Near-Vector Spaces
Definition 2.8. [8] A pair
is called a near-vector space if:
a)
is a group and F is a set of endomorphisms of V.
b) F contains the endomorphisms
and
.
c)
is a subgroup of the group of automorphisms of
.
d) If
with
and
, then
or
, i.e. F acts fixed point free (fpf) on V.
e) The quasi-kernel
of V, generates V additively as a group. Here
(1)
Remark 2.9.
a) Since
,
is an abelian group, i.e.
.
b) If
then
and
since
is an endomorphism of V.
c) Every vector space is a near-vector space.
Lemma 2.10. [1] Let
be a near-vector space. Then the quasi-kernel
has the following properties:
a)
.
b) For
,
in Equation(1)is uniquely determined by
and
.
c) If
, then
, i.e.,
.
d) If
and
, for
, then
for some
and for all positive numbers n.
e) If
and
, then there exists a
such that
.
Proof.
a) Let
and take any
. Then
. Hence,
.
b) Let
and
. Then there exist
such that
. That is,
. So by Definition 2.8-(d),
or
. But
. So,
.
c) Let
and
. Suppose
. Then
(by (a) above). Now suppose
. Let
. Then
. Since
, then there exist
such that
, since
.
So
. That is,
. Hence,
.
d) Let
and
. Want to show that
. We prove this by induction. That is,
i) Base case: For
, we have
(by (c) above).
ii) Inductive step: Assume
. We need to show that
. That is,
Therefore, by induction we conclude that
for some
and for all n, where n is a positive number.
e) Let
and
. Then by Remark 2.9-(b),
since F is a set of endomorphisms of V. So there exist
such that
, i.e.
.
□
Example 2.11. Every near-field is a near-vector space over itself.
Example 2.12. Let
. Then
is a near-vector space with scalar multiplication defined by
for all
. It is not a vector space over
. To see that it is not a vector space over
, let
. Then for
, we get the following two equations
and
In general,
. Thus, the distributive law for scalars does not hold in general, so
is not a vector space over
.
Definition 2.13. [9] Let
be a near-vector space. A subset
of
is said to be linearly independent if for all
and
,
implies that
A subset
is a generating system of
if for all
there exist
and
such that
A basis for
is defined to be a basis for V and the cardinality of the basis is the dimension of V.
Remark 2.14. A near-vector space
is said to be a finite-dimensional near-vector space if it is of finite dimension.
Lemma 2.15. [9] Let V be a near-vector space and let
be a basis of
. Then each
is a unique linear combination of elements of B, i.e. there exists
, with
for at most a finite number of
, which are uniquely determined by v and B, such that
Proof. See [9]. □
Example 2.16. The near-vector space from Example 2.12 has basis
. Therefore,
is a 2-dimensional near-vector space.
Definition 2.17. [10] If
is a near-vector space and
is such that
is the subgroup of
generated additively by
where X is a independent subset of
, then we say that
is a subspace of
, or simply
is a subspace of V if F is clear from the context.
Remark 2.18.
a) If
is a subspace of
, then X is a basis of
.
b)
is a subspace of
if and only if
is closed under addition and multiplication.
Example 2.19. Every near-vector space is a subspace of itself.
Next we define what it is meant when two near-vector spaces are said to be isomorphic.
Definition 2.20. [6] Two near-vector spaces
and
are said to be isomorphic if there are group isomorphisms
and
such that
for all
and
. We write
.
To further investigate the structure of near-vector spaces, we need to understand what is meant when two non-zero elements of the quasi-kernel
are compatible and when a near-vector space is regular.
Definition 2.21. [9] The elements u and
are compatible (u cp v), if there exists a
such that
Definition 2.22. [1] Let
be a near-vector space such that
or
. Let
. Define the operation
on F by
for all
.
Lemma 2.23. [1] Two non-zero elements of the quasi-kernel
are compatible if and only if there exists a
such that
Proof. See [9]. □
Theorem 2.24. [9] The compatibility relation is an equivalence relation on
.
Proof.
a) Reflexivity: Let
. Then by Lemma 2.10-(d),
for
. Thus u cp u.
b) Symmetry: Suppose u and v are compatible, then u cp v, with
. Then there exists
such that
. We need to show that v cp u. Since u and v are elements of
,
for
. Thus, v and u are compatible,i.e. v cp u.
c) Transitivity: Let
and suppose u cp v and v cp w. Then by Lemma 2.23
and
for
. We need to show that
for some
. Firstly, define
. So,
Hence, v cp w.
Therefore, the compatibility relation is an equivalence relation on
.
□
Theorem 2.25. [9] Let
and
. Then
a) u cp v, and
b) u cp
.
Proof.
a) Since
, by Definition 2.21 u and v are compatible with
.
b) Set
for
. Then u cp
since
for
(by Lemma 2.10-(d)).
□
Definition 2.26. [9] A near-vector space V is called a regular near-vector space if any two vectors of
are compatible.
Theorem 2.27. [9] A near-vector space V is regular if and only if there exists a basis which consists of mutually pairwise compatible vectors.
Proof. Suppose a near-vector space V is regular. Then, by Definition 2.26, any two non-zero elements of the quasi-kernel
are compatible. Therefore, every basis of
(also a basis of V) consists of mutually pairwise compatible vectors.
Conversely, let V be a near-vector space. Let B be a basis of V consisting of mutually pairwise compatible vectors. Let
. Then u can be written as a unique linear combination of elements of B (Lemma 2.15). Suppose
where
and
for all
. Let
Then
. Since
then for every
, there exists a
such that
But
. Thus, by uniqueness,
and
, which implies that
.
It still remains to check if u and
are compatible.
If
then
and by Lemma 2.10-(d),
. Thus
cp
. If
then by Theorem 2.25,
cp
since
and
are elements of
. But
cp
by Lemma 2.10-(d). Therefore, by transitivity (Theorem 2.24),
cp u. But by assumption
is compatible with every element of B. Therefore, u is compatible with every element of B by transitivity (Theorem 2.24). Thus if
then v cp
and w cp
with any
since
was chosen arbitrarily. Hence again, by transitivity (Theorem 2.24), v cp w. Thus every two non-zero elements of
are compatible. Therefore, a near-vector space V is regular. □
Definition 2.28. [5] The near-vector space
is said to be the direct sum of the subspaces
, symbolised by
if and only if
a)
, and
b)
for each i.
We then have the following important theorem by André.
Theorem 2.29. (The Decomposition Theorem) [11] Every near-vector space V is the direct sum of regular near-vector spaces
such that each
lies in precisely one direct summand
. The subspaces
are maximal regular near-vector spaces.
We note that if V is regular, it is the only maximal regular subspace. The Decomposition Theorem means that every non-regular near-vector space
is a direct sum of disjoint maximal regular near-vector subspaces
, for
, such that
.
We will not prove this theorem here. A proof can be found in [11]. The following is the procedure described in the proof on how V is decomposed into its maximal regular near-vector subspaces:
a) Partition
into sets
of mutually pairwise compatible vectors.
b) Let
be a basis of V and let
.
c) Let
be the subspace of V generated by
. Then each
is a maximal regular subspaces of V and V is a direct sum of those
’s.
As a result of Theorem 2.29, we have the following theorem.
Theorem 2.30. (The Uniqueness Theorem) [9] There exists only one direct decomposition of a near-vector space into maximal regular near subspaces.
Definition 2.31. [9] The uniquely determined direct decomposition of a near-vector space V into maximal regular subspaces, is called the canonical direct decomposition of V.
3. Construction of Near-Vector Spaces over
, for p a Prime
In this chapter we characterize all finite-dimensional near-vector spaces over
where p is prime and we use Van der Walt’s Theorem, also known as the Construction Theorem, to construct these near-vector spaces.
3.1. Automorphisms of
Before we get into constructing near-vector spaces, we start by understanding the basic structure of
, for p a prime. We know that
is a group under addition. For a prime p, let
Then
is a field under addition and multiplication modulo p. That is,
is a field with additive identity zero, multiplicative identity one, additive inverse
and multiplicative inverse
for every
. The inverse of
is an element, denoted by
, satisfying
All
are invertible. Thus,
is a set of invertible elements in
and the inverse of its elements can be computed using
We know that
is a cyclic group under multiplication. That is, there exists
such that
Such an element x is called a generator of
. Being cyclic implies that
is an abelian group.
Lemma 3.1. [9] Let q be a positive integer. Each element of
has a q-th root in
if and only if
.
Proof. Suppose that every element of
,
, has a q-th root in
. Let
, non-zero elements of
, be q-th roots of
, respectively. That is,
(2)
where
is a non-zero elements of
, in a particular order, for
. Suppose that
, where
. We know that if p is a prime number and
, then the congruence,
has exactly d solutions [12]. That is, there exists distinct
, such that
(3)
From Equation (3), we get
for some integer r. If
, then
for some integer t. That is,
. Using the binomial formula, we get
That is,
where m is an integer. Thus,
. This contradicts the fact that there is only one element of
,
, with
(from Equation (2)). Hence, if
and
, then
. Therefore,
.
Conversely, suppose that
. Then by [ [12], Theorem 2.3],
for some
. Let
(note that
has a q-th root being itself). Note that for every element of
,
,
. Since
, by Fermat’s Theorem,
Thus
which implies that
That is,
Hence, x has a q-th root in
, namely
. Therefore, each element of
has a q-th root in
if and only if
. □
The above lemma leads to the following lemma which makes use of the Galois Field
of order
.
Lemma 3.2. [13] The mapping
is an automorphism of the group
if and only if there exists
, with
and
, such that
for all
.
Proof. Since
is a finite field,
is a cyclic group. That is, there exists
such that
. Let
, for some
. Note that we cannot have
since
for all
. That is,
won’t be surjective. Then
for some
. But
, so is
, since
is bijective on
. This means that every element of
has a k-th root in
and by Lemma 3.1,
. Take
.
Conversely, suppose that there exists
, with
and
, such that
for all
. We know that
is a cyclic group and let a be its generator. That is,
is also a generator of
since
and
. Since
is a generator of
, then for any
, there exists
, such that
. Thus,
is surjective. Let
, such that
. Then there exists
, such that
and
. That is,
which implies that
and
. Hence,
is injective. Thus,
is bijective. Also for any
, we have
. Therefore,
is an automorphism of
. □
3.2. Van Der Walt’s Theorem
The following theorem describes how arbitrary finite-dimensional near-vector spaces can be constructed.
Theorem 3.3. [8] Let
be a group and let
, where D is a fixed point free group of automorphisms of V. Then
is a finite-dimensional near-vector space if and only if there exist a finite number of near-fields
, semigroup isomorphism
, and an additive group isomorphism
such that if
, then
for all
.
We will not prove the theorem here, see [14] for more details. We only show how it is used in constructing near-vector spaces over
, for p a prime. The following is a brief step by step approach on how to construct near-vector spaces over
using Theorem 3.3 and making use of Lemma 3.2.
Let
be a field with p a prime. Then
1). Set
where n is the dimension of our near-vector space. That is,
2). List all
such that
.
3). For each i, list all automorphisms,
, where
and
for
.
4). Define scalar multiplication by
for all
and
.
Lemma 3.4. [13] Let
and
(n-copies) be a near-vector space with scalar multiplication defined for all
by
where the
’s are automorphisms of
and they can be equal. Then V is regular if and only if for all
and
,
, for some
.
Proof. See [13]. □
Since
, i.e.
, we have that the maximal regular subspaces of V are those for which the
’s coincide. That is, if we have a near-vector space say,
, with scalar multiplication
. Then the canonical direct decomposition of V is
where
and
. Each
, for
, contains vectors with same type of action. By the Decomposition Theorem, those
’s are the maximal regular subspaces of V.
The following lemma describes the decomposition of the quasi-kernel of a near-vector space constructed using copies of
.
Lemma 3.5. [15] Suppose that V is a n-dimensional near-vector space over
with
and
is the canonical decomposition of V. Then
where
for each
.
Proof. See [15]. □
Consider the following example where we show that a partition across the quasi-kernel gives a near-vector space which is not regular.
Example 3.6. Let
be a near-vector space with scalar multiplication
Then the quasi-kernel is
. Let
. Then
is regular if u and v are compatible, u cp v. That is, there exist a
such that
. Take
and
. Then
Thus,
is not a regular near-vector space.
The number of
’s satisfying Lemma 3.2 is
, where
is the Euler’s totient function. That is, for each i, there are
distinct possibilities for
.
Theorem 3.7. [9] A n-dimensional near-vector space
over
is a vector space if and only if
.
Proof. If
for all i, then from Theorem 3.3 and Lemma 3.2, we get scalar multiplication as
which is a vector space for all
and
. For any other choices of
’s, with
, non-isomorphic near-vector spaces are created. □
We also have
Lemma 3.8. Let
be a near-vector space. For all
and
, let
with scalar multiplication defined by
for all
and
. Let
with scalar multiplication defined by
for all
and
, where all
and
is a positive integer. Then
.
Proof. To show that
. Define
by
and
by
for all
. We need to check the following.
a) If
and
are well-defined:
Let
. Then
Let
. Then
.
b) Checking if
and
are bijective and,
and
:
Let
and
. Suppose
. Then
. That is,
which implies that
(after multiplying by
, recall that
is a set of invertible elements in
). Thus,
is injective.
Let
. Then there exist
such that
. Take
for
. Then
. Thus
is surjective. Therefore,
is bijective. Furthermore,
Suppose also that
. Then
. Thus
is injective. Let
. Then there exist
such that
. That is,
. Thus
is surjective. Therefore,
is bijective. Also
.
c) Checking if
for all
and
:
Therefore
is isomorphic to
as a near-vector space i.e.
. □
3.3. Examples
Here we construct non-isomorphic near-vector spaces by following the procedure stipulated above in constructing near-vector spaces. Recall that for all
,
we get a vector space by Theorem 3.7 and by Lemma 3.8 we get isomorphic near-vector spaces. The quasi-kernel
of each near-vector space will be investigated and we will decompose V into its maximal regular near-vector subspaces by the procedure described in the proof of Theorem 2.29.
Example 3.9. Let
be a field. Let
. We are looking for
such that
. The candidates are
. We can take
and
and the scalar multiplication can be defined as follows
for all
and
. Hence
is a near-vector space.
The quasi-kernel
consists of elements of V,
, such that for every
there exists a
for which
.
i) Let us consider
. For
,
Hence
for each
.
ii) Consider
. For
,
Hence
for
. Note that
.
iii) Consider
. For
,
for
since in general,
. Hence
.
Note also that
for
.
Therefore, the quasi-kernel
is
Now we decompose V into its maximal regular near-vector subspaces. It is not difficult to verify that
is a basis of the near-vector space
. Let
. Then
Partitioning
, we get
and
Now we get that
and
Let
be a near-vector space generated by
. Then
Similarly, let
be a near-vector space generated by
. Then
Since
and
are arbitrary elements in F, we can take
as
and
respectively. So
Therefore by the Decomposition Theorem,
and
are maximal regular near-vector spaces and the canonical direct decomposition of V is
Example 3.10. Consider the field
. Let
. We want
such that
. The candidates for
are
. Therefore, we can take
,
and
. We define the scalar multiplication as follows;
for all
and
. Thus,
is a near-vector space.
We now investigate the quasi-kernel
. The quasi-kernel is found to be
We now decompose V into its maximal regular near-vector subspaces. It is not difficult to verify that
is a basis of the near-vector space
. Let
. Then
Partitioning
, we get
Now we get that
Let
and
be near-vector spaces generated by
and
respectively. Then
Therefore by the Decomposition Theorem,
and
are maximal regular near-vector spaces and the canonical direct decomposition of V is then
From the above examples, we get the following lemma.
Lemma 3.11. If
is a near-vector space of dimension n, i.e.
, then the number of
’s in the Decomposition Theorem is less or equal to n.
Proof. Suppose
is a near-vector space of dimension n and the number of
's in the Decomposition Theorem is greater than n. Then there exist
such that
is one of the cells of partitions of
. By construction in the proof of the Decomposition Theorem, we have
, where B is the basis of V, and
generates
. That is,
. Also V is a direct sum of all
’s,
, thus
. But
and
, which is a contradiction since
. Therefore, the number of
’s in the Decomposition Theorem is less or equal to n. □
4. An Application to Finite Linear Games
There are several applications of near-vector spaces: in Finite Linear Games and Cryptography, to mention a few. In this chapter we look at the application of near-vector spaces to finite or discrete linear games.
Most of the content here is taken from [15].
A finite linear game is a problem where a physical object has a finite number of states of which can be altered by applying certain processes. By so doing, they produce finitely many outcomes. Digital systems in computer science are often of this type. Since there are only a finite number of states, we can use elements of
, for p a prime, to represent the various states.
We generalize the idea by restricting the number of finite states to
for p a prime. But before we do that, consider the following example.
Recall that every vector space is a near-vector space. For consistency in the essay, we write scalars on the right of vectors.
Example 4.1. Five switches control five light bulbs in a row, changing the state, on or off, of the light bulb directly above it and the state of light bulbs adjacent to the left or right. Consider the figure below.
If the first and the third light bulbs are on as in Figure 1(a), then pushing switch A changes the state of the system to the state in Figure 1(b). If we next push switch C, the state changes to the state in Figure 1(c).
We are looking for an order in which we can push the switches so that only the first, third and fifth light bulbs will be on. Assume that all light bulbs are initially off. We make use of
over
. The vectors in
represent the action of each switch where the elements of
represent the action done
Figure 1. A row of five light bulbs controlled by five switches.
to the light bulb by a switch. Let 1 and 0 represent the on and off state of the light bulb, where
. Then the five switches can be represented by
Let
and
represent the initial and final state respectively. That is
To see how we can reach the final state
(target configuration), we need to determine whether there are scalars
such that
. That is,
We get the augmented matrix as
Applying row operations over
we get
Thus
is a free variable and we have two possible solutions
When
and
we get the following solutions respectively
Therefore, we push switch
once,
once and
once to reach the target configuration.
Note that there is no way we can push the switches in such a way that only the first light bulb will be on. When a particular switch is pushed n-times, where n is an even number, the light bulb goes back to its original state. Also for any switch say
,
(n times).
From the above example, we can see that given n light bulbs with p possible states and a switch associated with each light bulb, the initial and final configuration or state of the system is
respectively. For the switches we have
which captures the changes in the state of the light bulb above them. Each light bulb changes its state sequentially. The final state
corresponds to the vector addition of the initial state
and the number of times each switch is pressed (in some order). That is,
where
. The scalar multiplication is defined for all switches
and all
as follows
where q is a positive integer. Since we are working with near-vector spaces constructed over
using a result by van der Walt (Theorem 3.3), then the number of times,
, a particular switch is pressed is given by the following scalar multiplication
where
, for
, satisfies Lemma 3.2.
Note that in Example 4.1 we used the usual multiplication (linear) where
for all
.
Now we consider the case where the scalar multiplication is not linear.
Example 4.2. A row of four light bulbs is controlled by four switches. Each light bulb can have five possible states, 0, 1, 2, 3 or 4, where each number represents a different state. Each switch changes the states, assuming all light bulbs were initially off, of particular light bulbs as follows
Note that we are working over
since there are five possible states. The initial state is
and suppose the final state is
. To reach the final state, we need to determine whether there are scalars
such that
For the case where we use the usual multiplication, suitable sequence
, we get scalars to be
,
,
and
. We know that for suitable sequences
and
the resulting two near-vector spaces are isomorphic. But for a suitable sequence
, the number of times we press switches is different from how we press them with sequence
. With sequence
, we get
,
,
and
. It is no longer the case that for any switch
, for
,
(n-times). Take switch
for example,
whereas
This is all as a result of Lemma 3.2. Note that for the sequence
,
. That is, V is regular.
Example 4.3. Suppose from Example 4.2 we use suitable sequence
to define the scalar multiplication. This yields the following system of equations
(4)
(5)
(6)
(7)
Thus, we get two systems of equations, each has the same type of action.
(8)
(9)
and
(10)
(11)
The first system is linear and the second is non-linear. Keeping in mind that
is a field and every element of
has an inverse in
, by doing so we get
,
,
and
. Note that for Equation (5) and Equation (6) we used the fact that every element of
has a q-th root in
by Lemma 3.1. Note also that V is not regular, and observe that the quasi-kernel
is
Example 4.4. Suppose we have five possible states as in Example 4.2 with suitable sequence
but this time with the following switches.
The initial state is
and suppose the final state is
. We solve the following system of equations
(12)
(13)
(14)
(15)
We solve the same way we solved for Example 4.2 to get
,
,
and
. Note again that the quasi-kernel is
Also note that the switches belong to either
or
.
In a nutshell, we have considered the following cases:
a) Case where
, i.e. V is regular. Here the scalar multiplication is linear and is of the form
for any switch
and
with q satisfying Lemma 3.2. It’s easy to solve for scalars if
. If
, then Lemma 3.1 is of assistance in solving for scalars together with simple substitution.
b) Case where
, i.e. V is not regular. By Theorem 2.29 we can write V into its maximal regular subspaces, i.e.
so as its quasi-kernel
by Lemma 3.5. That is,
.
c) Case where there will be more than one system to solve. The system that needs be solved is solved simultaneously in conjunction with Lemma 3.1.
d) Case where the switches belong to
for some
. Here the rows of the coefficient matrix can be rearranged so that the system can be easier to solve.
5. Conclusion
The aim of this paper was to discuss the construction of near-vector spaces over
, for p a prime. We showed how to construct near-vector spaces using finite (right) near-fields and an important result (theorem) by van der Walt (also known as the Construction Theorem). We then used this theory to describe all finite-dimensional near-vector spaces over
, for p a prime, up to isomorphism. We also looked at their quasi-kernels and regularity. The Decomposition Theorem by André was applied on those near-vector spaces in order to get their maximal regular subspaces. We showed an application of near-vector spaces over
to finite linear games.
NOTES
*Co-first author.