Search:
Match:
1 results

Analysis

This paper introduces a robust version of persistent homology, a topological data analysis technique, designed to be resilient to outliers. The core idea is to use a trimming approach, which is particularly relevant for real-world datasets that often contain noisy or erroneous data points. The theoretical analysis provides guarantees on the stability of the proposed method, and the practical applications in simulated and biological data demonstrate its effectiveness.
Reference

The methodology works when the outliers lie outside the main data cloud as well as inside the data cloud.