Learning on Persistence Diagrams as Radon Measures
This work provides a theoretical framework for applying persistence diagrams in machine learning, but it appears incremental as it builds on existing concepts of Radon measures and optimal transport.
The paper tackles the problem of approximating continuous functions on the space of Radon measures derived from persistence diagrams for use in supervised learning tasks, showing that such functions can be approximated arbitrarily well using polynomial combinations of template-based features.
Persistence diagrams are common descriptors of the topological structure of data appearing in various classification and regression tasks. They can be generalized to Radon measures supported on the birth-death plane and endowed with an optimal transport distance. Examples of such measures are expectations of probability distributions on the space of persistence diagrams. In this paper, we develop methods for approximating continuous functions on the space of Radon measures supported on the birth-death plane, as well as their utilization in supervised learning tasks. Indeed, we show that any continuous function defined on a compact subset of the space of such measures (e.g., a classifier or regressor) can be approximated arbitrarily well by polynomial combinations of features computed using a continuous compactly supported function on the birth-death plane (a template). We provide insights into the structure of relatively compact subsets of the space of Radon measures, and test our approximation methodology on various data sets and supervised learning tasks.