Alessio Borgi

Alessio Borgi

PhD in Graph Neural Networks and Generative AI 👨🏻‍💻🤖

The Elder Rule and Pairing Uniqueness in Persistent Homology

2 minute read

Published:

TL;DR: The elder rule says that when a new simplex creates a merge of two connected components, the younger component (the one born later) is killed, not the older one. More generally, the persistence pairing assigns each "negative" simplex (that destroys a class) to the youngest "positive" simplex (that created a class) it can kill. This pairing is unique and well-defined regardless of the reduction algorithm.

Positive and Negative Simplices

In the standard persistence algorithm, each simplex \(\sigma_i\) added to the filtration is either:

  • Positive: its addition creates a new homology class (increases \(\beta_n\) by 1).
  • Negative: its addition destroys a homology class (decreases \(\beta_n\) by 1).

The algorithm pairs each negative simplex with the positive simplex whose homology class it kills.

The Elder Rule (\(H_0\) Example)

In \(H_0\) (connected components), the rule is simplest. When an edge \(e = \{u,v\}\) merges two components \(C_u\) (containing \(u\)) and \(C_v\) (containing \(v\)):

  • One component was born earlier (older, born at filtration time \(t_1\)).
  • One was born later (younger, born at time \(t_2 > t_1\)).

The elder rule: the younger component dies. Its birth vertex is paired with the merge edge.

The older component persists; the younger one is absorbed and its class dies.

This is analogous to a family tree rule: when two genealogical lines merge, the shorter/younger line is considered to end.

General Statement

For any homological dimension \(n\): when a negative simplex \(\sigma^-\) kills a class, it is paired with the positive simplex \(\sigma^+\) that created the youngest (most recently born) class that \(\sigma^-\) can kill.

Formally, after reducing the boundary matrix \(R\), we get the pairing:

\[\mathrm{pivot}(R_j) = i \implies \sigma_j \text{ is paired with } \sigma_i\]

Pairing Uniqueness Theorem

Theorem (Edelsbrunner et al. 2002): The pairing defined by the elder rule is unique: it does not depend on the choices made during the reduction algorithm (e.g., which pivot column to reduce first).

This is non-trivial: there are many valid sequences of column operations that reduce the boundary matrix, but they all produce the same set of pairs \((i, j)\).

Key Insight: Pairing uniqueness is what makes persistent homology a well-defined invariant of the filtration, not just an artifact of the algorithm. Different software implementations (Ripser, GUDHI, Javaplex, Dionysus) may use different reduction strategies and even different algorithmic paradigms, but they all output the same persistence pairs — because the pairing is canonically determined by the filtration itself.

References

  • H. Edelsbrunner, D. Letscher, A. Zomorodian, “Topological Persistence and Simplification,” Discrete & Computational Geometry, 2002.
  • H. Edelsbrunner & J. Harer, Computational Topology, Chapter VII.

You May Also Enjoy

GAPE: Remember to Forget — Gated Adaptive Positional Encoding

4 minute read

Published:

GAPE is a drop-in RoPE augmentation that adds content-aware attention logit biases: a query-gate suppresses irrelevant distant context while a key-gate preserves salient distant tokens. Provably sharper attention and improved long-context robustness — no architecture changes needed.

PolyNSD: Polynomial Neural Sheaf Diffusion

4 minute read

Published:

PolyNSD replaces the NSD propagation operator with a degree-K Chebyshev polynomial in the normalised sheaf Laplacian, achieving SOTA on homo- and heterophilic benchmarks with only diagonal restriction maps and dramatically lower memory usage.

TDA in Materials Science: Topology of Structure and Phase

3 minute read

Published:

Persistent homology characterises the multi-scale structure of materials — from pore geometry in catalysts to glass transition in amorphous materials. H₀ captures connectivity, H₁ captures channels and rings, H₂ captures enclosed voids. These topological descriptors predict mechanical properties, adsorption capacity, and phase transitions more accurately than geometric averages.

TDA in Drug Discovery: Molecular Topology

3 minute read

Published:

Persistent homology provides multi-scale topological fingerprints of molecular structures — capturing rings, cavities, and protein pockets that traditional cheminformatics descriptors miss. Applications include binding site detection, ADMET property prediction, and protein-ligand interaction modelling.