AI Paper Atlas

Research map / Learning theory and algorithms / Kernels and regression theory

Random Feature Maps for the Itemset Kernel

Kyohei Atarashi, Subhransu Maji, Satoshi Oyama

AAAI 2019 · Learning theory and algorithms › Kernels and regression theory · approach: Random feature approximation · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Rademacher instead of Laplace random vectors (ablation)RK feature map · Movielens 100K, second-order ANOVA, D = 2dabsolute error: 8.29e-4 -> 6.53e-4Improved
Rademacher instead of Gaussian random vectors (ablation)RK feature map · Movielens 100K, third-order ANOVA, D = 2dabsolute error: 2.67e-5 -> 2.26e-5Improved
different random vector distribution (ablation)RK feature map · Movielens 100K, all-subsets kernelabsolute error: 4.24e-2 vs 4.15e-2 at D = 2dNo clear effect
increase random feature dimension from 2d to 16dRK (Rademacher) · Movielens 100K, second-order ANOVAabsolute error: 6.53e-4 -> 2.33e-4Improved
increase random feature dimension from 2d to 16dRK (Rademacher) · Movielens 100K, all-subsets kernelabsolute error: 4.24e-2 -> 1.49e-2Improved
signed circulant random feature (SCRK) instead of RKRK feature map · Movielens 100K, third-order ANOVAabsolute error: 2.26e-5 -> 2.29e-5No clear effect
SCRK instead of RK (even order)RK feature map · Movielens 100K, second-order ANOVA, D = 2dabsolute error: 6.53e-4 -> 7.22e-4Hurt
SCRK instead of RK at large original dimension dRK feature map · synthetic data, d up to 4096mapping timeImproved
SCRK instead of RK at d = 512RK feature map · synthetic datamapping timeHurt
linear SVM with RK map (D = 1,248) instead of nonlinear SVM / FMlinear SVM · Movielens 100K binarizedtraining and testing time: 2-5 times lessImproved
linear SVM with RK map (D = 1,248) instead of nonlinear SVM / FMlinear SVM · Movielens 100K binarizedaccuracyNo clear effect
RM map for polynomial kernel instead of RK maplinear SVM · Movielens 100K binarizedaccuracyImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy