Spherical Structured Feature Maps for Kernel Approximation

Authors: Yueming Lyu

ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirically, SSF maps achieve superior performance compared with other methods.
Researcher Affiliation Academia 1Department of Computer Science, City University of Hong Kong, Tat Chee Avenue, Hong Kong .
Pseudocode Yes Algorithm 1
Open Source Code No The paper does not provide explicit statements or links to open-source code for the methodology described.
Open Datasets Yes We evaluate reconstruction error of Gaussian kernel, zeroorder arc-cosine kernel and first-order arc-cosine kernel on CIFAR10 (Krizhevsky & Hinton, 2009), MNIST (Le Cun & Cortes, 2010), usps and dna dataset.
Dataset Splits No The paper mentions randomly selecting 2,000 samples for constructing the Gram matrix but does not specify clear training, validation, or test dataset splits or percentages.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments are provided.
Software Dependencies No The paper mentions using 'MATLAB' but does not specify any version numbers for MATLAB or any specific software dependencies.
Experiment Setup Yes In all the experiments, we fix M = 1 (the number of one-dimensional QMC points) for SSF maps.