AI Paper Atlas

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

Alternating Circulant Random Features for Semigroup Kernels

Yusuke Mukuta, Yoshitaka Ushiku, Tatsuya Harada

AAAI 2018 · Learning theory and algorithms › Kernels and regression theory · approach: Random feature approximation · 0 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
remove random sign flipping (circ) instead of alternating mixing (ablation)altcirc · CUB BoVW, D=daccuracy (%): 11.26 -> 9.37Hurt
remove random sign flipping (circ) instead of alternating mixing (ablation)altcirc · Stanford Dogs BoVW, D=daccuracy (%): 16.83 -> 12.50Hurt
remove random sign flipping (circ) instead of alternating mixing (ablation)altcirc · Caltech256 VGG-16 softmax, D=daccuracy (%): 69.78 -> 67.35Hurt
increase mixed matrices from m = 2 to m = log2 d (ablation)altcirc · CUB VGG-16 softmax, D=daccuracy (%): 33.76 -> 36.05Improved
increase mixed matrices from m = 2 to m = log2 d (ablation)altcirc · Caltech256 VGG-16 softmax, D=daccuracy (%): 68.54 -> 69.78Improved
increase mixed matrices from m = 2 to m = log2 d (ablation)altcirc · Caltech256 BoVW, D=16daccuracy (%): 36.30 -> 36.09No clear effect
replace random Laplace features with altcirc (m = log2 d)semigroup kernel features · CUB VGG-16 softmax, D=16daccuracy (%): 42.55 -> 42.46No clear effect
replace Gaussian kernel with exponential-semigroup kernel (random Laplace)kernel choice · CUB VGG-16 softmax, D=16daccuracy (%): 23.44 -> 42.55Improved
replace Gaussian kernel with exponential-semigroup kernel (random Laplace)kernel choice · Caltech256 VGG-16 softmax, D=16daccuracy (%): 58.69 -> 72.74Improved
replace Gaussian kernel with semigroup kernel on last-activation featureskernel choice · CUB VGG-16 last activation, D=daccuracy (%): 56.88 -> 55.00Hurt
replace Gaussian kernel with altcirc (m = 2) as dimension growskernel choice · CUB VGG-16 last activation, D=16daccuracy (%): 57.87 -> 58.94Improved
increase output dimension from D = d to D = 16daltcirc (m = log2 d) · Caltech256 BoVWaccuracy (%): 30.06 -> 36.09Improved

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