D-GRIL: End-to-End Topological Learning with 2-parameter Persistence
Machine Learning
2025-02-25 v3 Artificial Intelligence
Algebraic Topology
Abstract
End-to-end topological learning using 1-parameter persistence is well-known. We show that the framework can be enhanced using 2-parameter persistence by adopting a recently introduced 2-parameter persistence based vectorization technique called GRIL. We establish a theoretical foundation of differentiating GRIL producing D-GRIL. We show that D-GRIL can be used to learn a bifiltration function on standard benchmark graph datasets. Further, we exhibit that this framework can be applied in the context of bio-activity prediction in drug discovery.
Cite
@article{arxiv.2406.07100,
title = {D-GRIL: End-to-End Topological Learning with 2-parameter Persistence},
author = {Soham Mukherjee and Shreyas N. Samaga and Cheng Xin and Steve Oudot and Tamal K. Dey},
journal= {arXiv preprint arXiv:2406.07100},
year = {2025}
}