A Survey of Vectorization Methods in Topological Data Analysis
arXiv:2212.09703 · doi:10.1109/TPAMI.2023.3308391
Abstract
Attempts to incorporate topological information in supervised learning tasks have resulted in the creation of several techniques for vectorizing persistent homology barcodes. In this paper, we study thirteen such methods. Besides describing an organizational framework for these methods, we comprehensively benchmark them against three well-known classification tasks. Surprisingly, we discover that the best-performing method is a simple vectorization, which consists only of a few elementary summary statistics. Finally, we provide a convenient web application which has been designed to facilitate exploration and experimentation with various vectorization methods.
References in corpus (12)
- Scikit-learn: Machine Learning in Python
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- A persistence landscapes toolbox for topological statistics
- On the stability of persistent entropy and new summary functions for TDA
- PersLay: A Neural Network Layer for Persistence Diagrams and New Graph Topological Signatures
- Persistence paths and signature features in topological data analysis
- Tropical Coordinates on the Space of Persistence Barcodes
- Approximating Continuous Functions on Persistence Diagrams Using Template Functions
- A Topological Machine Learning Pipeline for Classification
- Higher interpolation and extension for persistence modules
- Adaptive Partitioning for Template Functions on Persistence Diagrams
- Signatures, Lipschitz-free spaces, and paths of persistence diagrams