Research map / Learning theory and algorithms / Kernels and regression theory
Learning Deep Kernels for Non-Parametric Two-Sample Tests
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| use a learned deep kernel instead of an optimized Gaussian kernel | MMD two-sample test · Higgs | test power: 0.376 -> 0.403 | Helped |
| use a learned deep kernel instead of an optimized Gaussian kernel | MMD two-sample test · MNIST vs DCGAN | test power: 0.572 -> 0.910 | Helped |
| use a learned deep kernel instead of an optimized Gaussian kernel | MMD two-sample test · CIFAR-10 vs CIFAR-10.1 | mean rejection rate: 0.316 -> 0.744 | Helped |
| replace the sign-based kernel with a linear classifier kernel (ablation) | classifier two-sample test · Blob, HDGM, Higgs, and MNIST | test power | Mixed |
| replace the linear classifier kernel with a Gaussian feature kernel (ablation) | classifier two-sample test · Blob, HDGM, Higgs, and MNIST | test power | Mixed |
| replace the Gaussian feature kernel with the full deep kernel (ablation) | MMD two-sample test · Blob | test power: 0.966 -> 0.985 | Helped |
| replace the Gaussian feature kernel with the full deep kernel (ablation) | MMD two-sample test · HDGM | test power: 0.604 -> 0.659 | Helped |
| replace the Gaussian feature kernel with the full deep kernel (ablation) | MMD two-sample test · Higgs | test power: 0.364 -> 0.403 | Helped |
| replace the Gaussian feature kernel with the full deep kernel (ablation) | MMD two-sample test · MNIST | test power: 0.976 -> 0.996 | Helped |
| optimize approximate test power instead of cross-entropy (ablation) | deep-kernel and classifier-kernel variants · Blob, HDGM, Higgs, and MNIST | test power | Mixed |
| maximize empirical MMD instead of approximate test power (ablation) | MMD two-sample test · Blob, HDGM, Higgs, and MNIST | test power | Mixed |
Closest papers by meaning
- Meta Two-Sample Testing: Learning Kernels for Testing with Limited Data (NeurIPS 2021 · similarity 0.92)
- DUAL: Learning Diverse Kernels for Aggregated Two-sample and Independence Testing (NeurIPS 2025 · similarity 0.90)
- Kernel-based Maximum-of-difference Test for Two-sample Comparison (ICML 2026 · similarity 0.90)
- Kernel Selection is Model Selection: A Unified Complexity-Penalised Approach for MMD Two-Sample Tests (NeurIPS 2026 · similarity 0.88)
- Stochastic Kernel Regularisation Improves Generalisation in Deep Kernel Machines (NeurIPS 2024 · similarity 0.86)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.85)
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.