ICML 2023 Topological Deep Learning Challenge : Design and Results
Machine Learning
2024-01-19 v4
Abstract
This paper presents the computational challenge on topological deep learning that was hosted within the ICML 2023 Workshop on Topology and Geometry in Machine Learning. The competition asked participants to provide open-source implementations of topological neural networks from the literature by contributing to the python packages TopoNetX (data processing) and TopoModelX (deep learning). The challenge attracted twenty-eight qualifying submissions in its two-month duration. This paper describes the design of the challenge and summarizes its main findings.
Keywords
Cite
@article{arxiv.2309.15188,
title = {ICML 2023 Topological Deep Learning Challenge : Design and Results},
author = {Mathilde Papillon and Mustafa Hajij and Helen Jenne and Johan Mathe and Audun Myers and Theodore Papamarkou and Tolga Birdal and Tamal Dey and Tim Doster and Tegan Emerson and Gurusankar Gopalakrishnan and Devendra Govil and Aldo Guzmán-Sáenz and Henry Kvinge and Neal Livesay and Soham Mukherjee and Shreyas N. Samaga and Karthikeyan Natesan Ramamurthy and Maneel Reddy Karri and Paul Rosen and Sophia Sanborn and Robin Walters and Jens Agerberg and Sadrodin Barikbin and Claudio Battiloro and Gleb Bazhenov and Guillermo Bernardez and Aiden Brent and Sergio Escalera and Simone Fiorellino and Dmitrii Gavrilev and Mohammed Hassanin and Paul Häusner and Odin Hoff Gardaa and Abdelwahed Khamis and Manuel Lecha and German Magai and Tatiana Malygina and Rubén Ballester and Kalyan Nadimpalli and Alexander Nikitin and Abraham Rabinowitz and Alessandro Salatiello and Simone Scardapane and Luca Scofano and Suraj Singh and Jens Sjölund and Pavel Snopov and Indro Spinelli and Lev Telyatnikov and Lucia Testa and Maosheng Yang and Yixiao Yue and Olga Zaghen and Ali Zia and Nina Miolane},
journal= {arXiv preprint arXiv:2309.15188},
year = {2024}
}