ICLR 2022 Challenge for Computational Geometry and Topology: Design and Results

TL;DR

This paper details the design and outcomes of a challenge on geometric and topological machine learning algorithms, emphasizing implementations respecting specific software APIs, with seven teams participating.

Contribution

It introduces a novel challenge framework for evaluating ML algorithms on manifolds, fostering development aligned with Geomstats and Scikit-Learn standards.

Findings

Seven teams participated in the challenge.

The challenge revealed diverse approaches to geometric ML.

Insights into implementation challenges and best practices.

Abstract

This paper presents the computational challenge on differential geometry and topology that was hosted within the ICLR 2022 workshop ``Geometric and Topological Representation Learning". The competition asked participants to provide implementations of machine learning algorithms on manifolds that would respect the API of the open-source software Geomstats (manifold part) and Scikit-Learn (machine learning part) or PyTorch. The challenge attracted seven teams in its two month duration. This paper describes the design of the challenge and summarizes its main findings.