Persistent Diagram Estimation of Multivariate Piecewise Hölder-continuous Signals

Hugo Henneuse (1, 2)

1 DATASHAPE - Understanding the Shape of Data
2 LMO - Laboratoire de Mathématiques d'Orsay

To our knowledge, the analysis of convergence rates for persistent diagram estimation from noisy signals had remained limited to lifting signal estimation results through sup norm (or other functional norm) stability theorems. We believe that moving forward from this approach can lead to considerable gains. We illustrate it in the setting of Gaussian white noise model. We examine from a minimax perspective, the inference of persistent diagram (for sublevel sets filtration). We show that for piecewise Hölder-continuous functions, with control over the reach of the discontinuities set, taking the persistent diagram coming from a simple histogram estimator of the signal, permit to achieve the minimax rates known for Hölder-continuous functions.

Mots clés: Topological Data Analysis; Persistent Homology; Non-parametric statistics; Minimax

Domaines: Statistiques [math.ST]; Topologie algébrique [math.AT]

Hugo Henneuse. Persistent Diagram Estimation of Multivariate Piecewise Hölder-continuous Signals. 2024. ⟨hal-04524998⟩