AI Paper Atlas

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

Convex Representation Learning for Generalized Invariance in Semi-Inner-Product Space

Yingyi Ma, Vignesh Ganapathiraman, Yaoliang Yu, Xinhua Zhang

ICML 2020 · Learning theory and algorithms › Kernels and regression theory · approach: Kernel regression under shift · 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
Use sipsf representations instead of vanilla mixupGaussian RKHS mixup classifier · MNIST, n=500, p=4ntest accuracy: 91.40\scriptstyle\pm1.04 -> 92.11\scriptstyle\pm1.01Improved
Use sipsf representations instead of vanilla mixupGaussian RKHS mixup classifier · USPS, n=1000, p=4ntest accuracy: 95.32\scriptstyle\pm1.13 -> 95.67\scriptstyle\pm0.96Improved
Use sipsf representations instead of vanilla mixupGaussian RKHS mixup classifier · Fashion MNIST, n=500, p=4ntest accuracy: 81.72\scriptstyle\pm1.96 -> 82.52\scriptstyle\pm1.49Improved
Use sipsf representations instead of vanilla mixupGaussian RKHS mixup classifier · Fashion MNIST, n=1000, p=4ntest accuracy: 83.69\scriptstyle\pm1.31 -> 84.34\scriptstyle\pm1.31Improved
Use sipsf with label-logic regularization instead of the hierarchical SVMStructured multilabel prediction · Enron, 100/100 train/testaccuracy: 93.1 -> 96.2Improved
Use sipsf with label-logic regularization instead of the hierarchical SVMStructured multilabel prediction · Reuters, 200/200 train/testaccuracy: 97.3 -> 97.2Hurt
Use sipsf with label-logic regularization instead of the hierarchical SVMStructured multilabel prediction · WIPO, 500/500 train/testaccuracy: 98.7 -> 98.4Hurt
Use Euclidean embedding with the max transformation-derivative penalty instead of ordinary SVMMNIST binary classification · 4 vs 9test accuracy: 97.1 -> 97.8Improved
Use Euclidean embedding with the max transformation-derivative penalty instead of ordinary SVMMNIST binary classification · 2 vs 3test accuracy: 98.4 -> 98.9Improved

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