Persistence Paths and Signature Features in Topological Data Analysis.

Author: 

Chevyrev, I
Nanda, V
Oberhauser, H

Publication Date: 

7 December 2018

Journal: 

IEEE transactions on pattern analysis and machine intelligence

Last Updated: 

2019-05-14T15:10:39.44+01:00

DOI: 

10.1109/tpami.2018.2885516

abstract: 

We introduce a new feature map for barcodes as they arise in persistent homology computation. The main idea is to first realize each barcode as a path in a convenient vector space, and to then compute its path signature which takes values in the tensor algebra of that vector space. The composition of these two operations - barcode to path, path to tensor series - results in a feature map that has several desirable properties for statistical learning, such as universality and characteristicness, and achieves state-of-the-art results on common classification benchmarks.

Symplectic id: 

857812

Submitted to ORA: 

Submitted

Publication Type: 

Journal Article