Finding Relevant Information via a Discrete Fourier Expansion

Authors: Mohsen Heidari, Jithin Sreedharan, Gil I Shamir, Wojciech Szpankowski

ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Lastly, we present an algorithm based on our measure and verify our findings via numerical experiments on various datasets.
Researcher Affiliation Collaboration 1NSF Center for Science of Information, Purdue University, West Lafayette, USA 2Wadhwani AI, Mumbai, India 3Google Inc., Pittsburgh, USA.
Pseudocode Yes Procedure 1 FOURIER-ORTH
Open Source Code Yes The source codes are available at https://github.com/jithin-k-sreedharan/Fourier feature selection.
Open Datasets Yes The real-world datasets are benchmarks and taken from (Li et al., 2018) and the UCI repository (Dua & Graff, 2017).
Dataset Splits Yes The experiments employ 5-fold cross-validation with feature selection and the support vector machine (SVM) classifier with radial basis function as a kernel.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory specifications) used for running experiments are provided in the paper.
Software Dependencies No The paper mentions using the support vector machine (SVM) classifier and refers to implementations from external sources (Li et al., 2018) for other algorithms, but no specific software versions (e.g., Python 3.x, PyTorch 1.x, scikit-learn 0.x) are listed.
Experiment Setup No The paper states that experiments use 'support vector machine (SVM) classifier with radial basis function as a kernel' but does not provide specific hyperparameter values (e.g., learning rate, batch size, SVM parameters C/gamma) or other detailed training configurations.