Generalization Guarantees for Sparse Kernel Approximation with Entropic Optimal Features

Authors: Liang Ding, Rui Tuo, Shahin Shahrampour

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our numerical experiments on benchmark datasets verify the superiority of EOF over the state-of-the-art in kernel approximation.
Researcher Affiliation Academia The authors are with Wm Michael Barnes 64 Department of Industrial and Systems Engineering, Texas A&M University, College Station, TX, USA.
Pseudocode Yes Algorithm 1 Entropic Optimal Features (EOF)
Open Source Code No The paper does not explicitly state that source code for the described methodology is being released or provide a link to a code repository.
Open Datasets Yes Benchmark Algorithm: We now compare EOF with the following benchmark algorithms on several datasets from the UCI Machine Learning Repository:
Dataset Splits Yes In Table 1, we report the number of training samples Ntrain and test samples Ntest used for each dataset.
Hardware Specification Yes The run time is obtained on a Macbook Pro with a 4-core, 3.3 GHz Intel Core i5 CPU and 8 GB of RAM (2133Mhz).
Software Dependencies No The paper mentions 'Matlab' in the context of efficient matrix operations but does not provide specific version numbers for any software dependencies.
Experiment Setup No The paper describes data standardization and selection of feature sets, but does not provide specific hyperparameter values for model training, such as learning rates, batch sizes, or optimizer settings.