On the Effectiveness of Persistent Homology

Authors: Renata Turkes, Guido F. Montufar, Nina Otter

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments demonstrate that PH is successful in these tasks, outperforming several baselines, including Point Net, an architecture inspired precisely by the properties of point clouds.
Researcher Affiliation Academia Renata Turkeš University of Antwerp renata.turkes@uantwerpen.be Guido Montúfar University of California, Los Angeles montufar@math.ucla.edu Nina Otter Queen Mary University of London n.otter@qmul.ac.uk
Pseudocode No The paper describes the 'PH pipeline' and its steps but does not include any formal pseudocode or algorithm blocks.
Open Source Code Yes All the code for data generation, analysis, and visualization is publicly available at https://github.com/renataturkes/on_the_effectiveness_of_persistent_homology.
Open Datasets Yes We provide data sets that can be directly used as a benchmark for our tasks or other related pointcloud-analysis or classification problems.
Dataset Splits No The paper specifies training and testing splits (e.g., 'train the classifier on 80%... and test on the remaining 20%'), but it does not explicitly provide details for a separate validation split.
Hardware Specification Yes The computational experiments were performed using a computer with 3.5 GHz 6-Core Intel Xeon E5 processor and 64 GB 1866 MHz DDR3 RAM.
Software Dependencies Yes The code is written in Python 3.8.10... The packages used are: gudhi (v.3.5.0), numpy (v.1.20.1), matplotlib (v.3.3.4), ripser (v.0.6.2), scikit-learn (v.0.24.1), scipy (v.1.6.2), sklearn (v.0.0).
Experiment Setup Yes SVMs with radial basis function (RBF) kernels were chosen for regression and classification tasks. The hyperparameters of the SVMs were optimized by grid search over values of C {0.1, 1, 10, 100} and γ {0.001, 0.01, 0.1, 1, 10, 100}.