2026-07-22T14:14:25Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:25886 2026-06-23T13:00:14Z ddc:004 open_access

D-GRIL: End-To-End Topological Learning with 2-Parameter Persistence Mukherjee, Soham Samaga, Shreyas N. Xin, Cheng Oudot, Steve Dey, Tamal K. Topological Data Analysis Persistent Homology Multiparameter Persistence Graph Learning Graph Neural Networks 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 theory for gradient descent on Gril producing D-Gril. We show that D-Gril can be used to learn a bifiltration function on benchmark graph datasets. Further, we exhibit that this framework can be applied in the context of bio-activity prediction in drug discovery. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Soham Mukherjee and Shreyas N. Samaga and Cheng Xin and Steve Oudot and Tamal K. Dey 2026 Is Part Of LIPIcs, Volume 367, 42nd International Symposium on Computational Geometry (SoCG 2026) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.SoCG.2026.79 urn:nbn:de:0030-drops-258865 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.SoCG.2026.79 eng https://creativecommons.org/licenses/by/4.0/legalcode