Research map / Learning theory and algorithms / Kernels and regression theory
The Geometry of Nonlinear Embeddings in Kernel Discriminant Analysis
Jiae Kim, Yoonkyung Lee, Zhiyu Liang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use odd-degree homogeneous polynomial kernel (ablation) | kernel discriminant analysis · Scenario 1 (different means) | discriminant direction | Improved |
| use even-degree homogeneous polynomial kernel (ablation) | kernel discriminant analysis · Scenario 1 (different means) | discriminant direction | Hurt |
| use even-degree homogeneous polynomial kernel (ablation) | kernel discriminant analysis · Scenario 2 (different covariances) | discriminant | Improved |
| use odd-degree homogeneous polynomial kernel (ablation) | kernel discriminant analysis · Scenario 2 (different covariances) | discriminant | Hurt |
| use inhomogeneous instead of homogeneous polynomial kernel (ablation) | kernel discriminant analysis · Scenario 1 (different means) | discriminant direction | Improved |
| increase truncation degree N of Gaussian kernel expansion (ablation) | Gaussian discriminant · Scenarios 1 and 2 | eigenvalue lambda_N (class separation) | Improved |
| use random Fourier features instead of polynomial features (ablation) | Gaussian discriminant approximation · Scenarios 1 and 2 | eigenvalue (class separation) | Hurt |
| increase number of random Fourier features D from 2 to 40 | Gaussian discriminant approximation · Scenarios 1 and 2 | approximation of optimal boundary | Improved |
Closest papers by meaning
- Discriminant Feature Extraction by Generalized Difference Subspace (TPAMI 2022 · similarity 0.87)
- Neural Fisher Discriminant Analysis: Optimal Neural Network Embeddings in Polynomial Time (ICML 2022 · similarity 0.87)
- Learning to Embed Distributions via Maximum Kernel Entropy (NeurIPS 2024 · similarity 0.85)
- Learning with Exact Invariances in Polynomial Time (ICML 2025 · similarity 0.85)
- Polynomial-Time Robust Multiclass Linear Classification under Gaussian Marginals (NeurIPS 2026 · similarity 0.84)
- Learning in Feature Spaces via Coupled Covariances: Asymmetric Kernel SVD and Nyström method (ICML 2024 · similarity 0.84)
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.
