Difference between revisions of "APS of groups"

From Groupprops
Jump to: navigation, search
m (7 revisions)
(One intermediate revision by the same user not shown)
Line 13: Line 13:
Satisfying  the following compatibility conditions:
Satisfying  the following compatibility conditions:
For <math>g, h, k</math> in <math>G_m, G_n, G_p</math> respectively,
For <math>g, h, k</math> in <math>G_m, G_n, G_p</math> respectively:
<math>\Phi_{m+n,p} (\Phi_{m,n}(g,h),k) = \Phi_{m,n+p} (g, \Phi_{n,p}(h,k))</math>.
<math>\Phi_{m+n,p} (\Phi_{m,n}(g,h),k) = \Phi_{m,n+p} (g, \Phi_{n,p}(h,k))</math>.

Latest revision as of 22:49, 7 May 2008

This article gives a basic definition in the following area: APS theory
View other basic definitions in APS theory |View terms related to APS theory |View facts related to APS theory
This article defines the notion of group object in the category of APSs|View other types of group objects

BEWARE! This term is nonstandard and is being used locally within the wiki. [SHOW MORE]


An APS of groups is an APS over the category of groups. More specifically, an APS (G,\Phi) of groups is the following data:

  • For each natural number n, a group, denoted G_n.
  • For each ordered pair (m,n) of natural numbers, a homomorphism \Phi_{m,n}:G_m \times G_n \to G_{m+n}.

Satisfying the following compatibility conditions:

For g, h, k in G_m, G_n, G_p respectively:

\Phi_{m+n,p} (\Phi_{m,n}(g,h),k) = \Phi_{m,n+p} (g, \Phi_{n,p}(h,k)).

The above condition is termed an associativity condition.

We may assume G_0 as the trivial group and define \Phi_{m,0} and \Phi_{0,n} as trivial paddings.


Members and elements

For an APS (G,\Phi) the member G_n is termed the n^{th} member of the APS. A member of the APS is an object that is the n^{th} member for some n.

An element of the APS is an element of some member of the APS.

The home of an element of the APS is the member in which it lies. The index of a member is the n for which it is the n^{th} member.

Block concatenation map

The maps \Phi_{m,n} are termed block concatenation maps.

Ground member

The ground member of an APS is its first member.

Other notions

Homomorphism of APSes

Further information: APS homomorphism

Given APSes (G,\Phi) and (H,\Psi), a homomorphism h: GH associates, to each natural number n, a map h_n: G_nH_n, such that:

\Psi_{m,n}(h_m(a),h_n(b)) = h_{m+n}(\Phi_{m,n}(a,b))

Sub-APS notion

Further information: sub-APS

Given an APS (G,\Phi), a sub-APS H associates, to each n, a subgroup H_n of G_n, such that the image of H_m X H_n under \Phi_{m,n} lies inside H_{m+n}.

When the APS of groups is injective, any sub-APS is also injective.

Quotient APS notion

Further information: quotient APS

A quotient APS is the image of an APS in a homomorphism that is surjective at each member.

Normal sub-APSes, kernels and images

Given a homomorphism of APSes of groups, the kernels of the individual homomorphisms for a sub-APS of the domain APS, and the images of the individual homomorphism form a sub-APS of the range APS. The image is thus a quotient APS.

Further, we have the following result: a sub-APS of an APS of groups occurs as the kernel of an APS homomorphism if and only if every member of it is a normal subgroup of the corresponding member of the whole APS. A sub-APS satisfying either of these equivalent conditions is termed a normal sub-APS.

This parallels the group theory result that a subgroup of a group occurs as the kernel of a group homomorphism if and only if it is normal.



An APS of groups is termed injective, or an IAPS of groups, if every block concatenation map is injective. For an IAPS of groups, we usually also assume the condition of refinability.


Very few APSes of groups are commutative. Note that a commutative APS cannot also be injective.


Most APSes of groups that we encounter satisfy the condition of being padding-injective.