On the Bootstrap for Persistence Diagrams and Landscapes

Frédéric Chazal (1) , Brittany Terese Fasy (2) , Fabrizio Lecci (3) , Alessandro Rinaldo (3) , Aarti Singh (4) , Larry Wasserman (3)

1 GEOMETRICA - Geometric computing
2 Department of Computer Science [Tulane]
3 Statistics Department, Carnegie Mellon University
4 Machine Learning Department [Carnegie Mellon Univ.]

Persistent homology probes topological properties from point clouds and functions. By looking at multiple scales simultaneously, one can record the births and deaths of topological features as the scale varies. In this paper we use a statistical technique, the empirical bootstrap, to separate topological signal from topological noise. In particular, we derive confidence sets for persistence diagrams and confidence bands for persistence landscapes.

Domaines: Géométrie algorithmique [cs.CG]; Topologie algébrique [math.AT]; Applications [stat.AP]

Frédéric Chazal, Brittany Terese Fasy, Fabrizio Lecci, Alessandro Rinaldo, Aarti Singh, et al.. On the Bootstrap for Persistence Diagrams and Landscapes. 2013. ⟨hal-00879982⟩