Statistical Inference for Persistence Diagrams via Landmark Embeddings: Minimax Theory and Finite Approximation
Abstract
Hilbert-space embeddings enable inference for populations of persistence diagrams, but separation between individual diagrams need not survive population averaging. We develop a framework for inference on population mean embeddings, with particular attention to the additive landmark representations PLACE and PALACE. Treating each diagram as one independent observation, we apply Hilbert-space limit theory to obtain covariance estimators, two-sample tests, and confidence balls under suitable moment conditions, without requiring a lower-distortion bound. For additive embeddings, we identify the population mean as an embedding of the mean counting measure and show that geometric separation of these measures alone cannot guarantee uniform testing power. We then introduce a model with latent template diagrams, missing features, and location perturbations. Under common or feature-specific prevalence conditions, a diagram-level lower-distortion certificate yields explicit lower bounds on population mean separation. These margins provide finite-sample uniform power guarantees, and an additional information-divergence comparison gives matching sample-complexity bounds over restricted scale ranges. Confidence sets yield lower bounds on transport separation of population mean measures and exclusion guarantees for specified structured alternatives. We also quantify how orthogonal truncation changes the certified signal and the approximation allowance needed for confidence sets targeting the full embedding, relating sample size, retained coordinates, and template separation. Simulations examine calibration, power, and coverage, and an analysis of resting-state connectivity from the Autism Brain Imaging Data Exchange illustrates the procedures.
Related
- Where Do Two Populations of Persistence Diagrams Differ? Calibrated Local Inference at a Fixed Budget
- A Statistical Inference Framework for PMI Estimation and SGNS Word Embeddings
- Persistence diagrams of finite topological spaces
- A result of continuous stability for generalized persistence diagrams
- On the Spectral Synthesis of Lipschitz Vectorizations of Persistence Diagrams
BibTeX
@misc{bagchi2026statistical,
title = {{Statistical Inference for Persistence Diagrams via Landmark Embeddings: Minimax Theory and Finite Approximation}},
author = {Pramita Bagchi and Sushovan Majhi and Atish Mitra and Žiga Virk},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.07691},
eprint = {2609.07691},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.07691},
}
RIS
TY - UNPB
TI - Statistical Inference for Persistence Diagrams via Landmark Embeddings: Minimax Theory and Finite Approximation
AU - Pramita Bagchi
AU - Sushovan Majhi
AU - Atish Mitra
AU - Žiga Virk
PY - 2026
DA - 2026/09/07
PB - arXiv
DO - 10.48550/arxiv.2609.07691
UR - https://arxiv.org/abs/2609.07691
AB - Hilbert-space embeddings enable inference for populations of persistence diagrams, but separation between individual diagrams need not survive population averaging. We develop a framework for inference on population mean embeddings, with particular attention to the additive landmark representations PLACE and PALACE. Treating each diagram as one independent observation, we apply Hilbert-space limit theory to obtain covariance estimators, two-sample tests, and confidence balls under suitable moment conditions, without requiring a lower-distortion bound. For additive embeddings, we identify the population mean as an embedding of the mean counting measure and show that geometric separation of these measures alone cannot guarantee uniform testing power. We then introduce a model with latent template diagrams, missing features, and location perturbations. Under common or feature-specific prevalence conditions, a diagram-level lower-distortion certificate yields explicit lower bounds on population mean separation. These margins provide finite-sample uniform power guarantees, and an additional information-divergence comparison gives matching sample-complexity bounds over restricted scale ranges. Confidence sets yield lower bounds on transport separation of population mean measures and exclusion guarantees for specified structured alternatives. We also quantify how orthogonal truncation changes the certified signal and the approximation allowance needed for confidence sets targeting the full embedding, relating sample size, retained coordinates, and template separation. Simulations examine calibration, power, and coverage, and an analysis of resting-state connectivity from the Autism Brain Imaging Data Exchange illustrates the procedures.
ER -