homology
Simplicial Homology
You should know: singular homology, chain complexes, linear independence
Overview
Simplicial homology is a computational approach to homology theory defined on triangulated spaces called simplicial complexes (or delta complexes). By decomposing a space into simplices -- points, edges, triangles, tetrahedra, and their higher-dimensional analogues -- one can define chain groups and boundary maps explicitly with integer matrices, making homology algorithmically computable. Simplicial homology agrees with singular homology for triangulable spaces.
Intuition
Break a shape into building blocks: vertices (0-simplices), edges (1-simplices), triangles (2-simplices), tetrahedra (3-simplices). Each edge has two boundary vertices; each triangle has three boundary edges; each tetrahedron has four boundary triangular faces. Summing boundaries with signs gives a 'boundary operator'. A cycle is a collection of simplices with no boundary; a boundary is a cycle that itself bounds something higher-dimensional. Homology = cycles modulo boundaries, measuring the 'holes' that are not filled in.
Formal Definition
A simplicial complex K is a collection of simplices closed under taking faces. The n-th chain group C_n(K) is the free abelian group on the oriented n-simplices. The boundary operator ∂_n: C_n → C_{n-1} sends each n-simplex [v₀,...,vₙ] to the alternating sum of its faces. The simplicial homology groups are H_n(K) = ker(∂_n) / im(∂_{n+1}).
Notation
| Notation | Meaning |
|---|---|
| n-th simplicial chain group of complex K | |
| Boundary homomorphism C_n → C_{n-1} | |
| Oriented n-simplex with vertices v₀,...,vₙ | |
| n-th simplicial homology group of K | |
| n-th Betti number: rank of H_n(K) |
Theorems
Worked Examples
- 1
Label vertices v₀, v₁, v₂ and edges e₀=[v₁,v₂], e₁=[v₀,v₂], e₂=[v₀,v₁] (no 2-simplex).
- 2
C₀ = Z³ with basis {v₀,v₁,v₂}, C₁ = Z³ with basis {e₀,e₁,e₂}, C₂ = 0.
- 3
Boundary: ∂₁(e₀)=v₂-v₁, ∂₁(e₁)=v₂-v₀, ∂₁(e₂)=v₁-v₀.
- 4
H₀ = Z (one connected component). ker(∂₁) has rank 1 (the cycle e₀-e₁+e₂). im(∂₂)=0. So H₁ = Z.
✓ Answer
H₀(S^1) = Z, H₁(S^1) = Z, H_n(S^1) = 0 for n ≥ 2.
Practice Problems
Prove that ∂_{n-1} ∘ ∂_n = 0 for simplicial chain complexes.
A simplicial complex K has 5 vertices, 8 edges, and 4 triangles. Compute χ(K) and the Betti numbers assuming K is connected and simply connected.
Common Mistakes
Orientation of simplices does not matter in simplicial homology.
Orientation is essential. Changing the orientation of a simplex negates it as a chain. However, the homology groups H_n(K) are well-defined independent of the choice of orientations.
Simplicial homology computes something different from singular homology.
For triangulable spaces, simplicial and singular homology agree. Simplicial homology is computationally more tractable, while singular homology is defined for all topological spaces.
Quiz
Historical Background
Simplicial homology was the first rigorous formulation of homology, growing out of Poincaré's use of polyhedra in Analysis Situs (1895). Poincaré introduced the Betti numbers (ranks of homology groups) and the Euler characteristic. Veblen, Alexander, and Lefschetz systematized the theory in the 1920s--1930s. The comparison with singular homology was established by Eilenberg and Steenrod's axiomatic approach.
- 1895
Poincaré defines Betti numbers and homology for polyhedra
Poincaré
- 1923
Veblen publishes a rigorous treatment of simplicial homology
Veblen
- 1952
Eilenberg and Steenrod prove simplicial = singular homology via axioms
Eilenberg, Steenrod
Summary
- Simplicial homology decomposes a triangulated space into simplices and defines chain groups, boundary operators, and homology groups H_n = ker(∂_n)/im(∂_{n+1}).
- The boundary operator satisfies ∂∘∂ = 0, which is the key algebraic identity enabling the definition of homology.
- Simplicial homology agrees with singular homology for triangulable spaces.
- The Euler characteristic equals both the alternating sum of simplex counts and the alternating sum of Betti numbers.
- Computing simplicial homology reduces to linear algebra: finding kernels and images of integer matrices.
References
- BookHatcher, A. Algebraic Topology. Cambridge University Press, 2002. Section 2.1.
- BookMunkres, J.R. Elements of Algebraic Topology. Addison-Wesley, 1984.
Mathematics