Near Input Sparsity Time Kernel Embeddings via Adaptive Sampling

Authors: David Woodruff, Amir Zandieh

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

Reproducibility Variable Result LLM Response
Research Type Experimental Furthermore, we empirically show that in large-scale regression tasks, our algorithm outperforms state-of-the-art kernel approximation methods.
Researcher Affiliation Academia 1Carnegie Mellon University, USA 2Ecole polytechnique federale de Lausanne, Switzerland.
Pseudocode Yes Algorithm 1 RECURSIVE LEVERAGE SCORE SAMPLING and Algorithm 2 ROWSAMPLER FOR POLYNOMIAL KERNEL are provided.
Open Source Code No The paper does not provide any explicit statement about releasing source code or a link to a code repository.
Open Datasets Yes We base our comparison on the four standard large-scale regression datasets evaluated in (Le et al., 2013). The size of the data points is denoted by n and the dimensionality is denoted by d in Table 1. DATASET: WINE INSURANCE CT LOCATION FOREST
Dataset Splits Yes We use the same hyperparameters (kernel bandwidth and regularization parameter) across all kernel approximation methods which were selected via cross-validation on the Fourier features method, as our base-line method.
Hardware Specification No The paper does not provide any specific details about the hardware used for running the experiments (e.g., GPU models, CPU types).
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., library names with versions).
Experiment Setup Yes We use the same hyperparameters (kernel bandwidth and regularization parameter) across all kernel approximation methods which were selected via cross-validation on the Fourier features method, as our base-line method.