Research map / Learning theory and algorithms / Kernels and regression theory
Nearest-Neighbor Sample Compression: Efficiency, Consistency, Infinite Dimensions
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 |
|---|---|---|---|
| Replace classic 1-NN with KSU compression-based 1-NN (gamma-net majority-vote with compression bound for scale selection) | 1-NN multiclass classifier in finite doubling-dimension metric spaces · finite doubling dimension metric spaces (theoretical) | Bayes-consistency (strong) | Helped |
| Use KSU (compression-based 1-NN) instead of k-NN on Preiss's infinite-dimensional construction | KSU vs k-NN · Preiss infinite-dimensional construction (Example 1) | Bayes-consistency (weak) | Helped |
| Use majority-vote gamma-net partitions with vanishing diameter that violate the packing property (ablation) | Voronoi partition classifiers in infinite dimensions · Preiss infinite-dimensional construction | Bayes-consistency | Hurt |
Closest papers by meaning
- Stabilizing Sample Similarity in Representation via Mitigating Random Consistency (ICML 2025 · similarity 0.86)
- Trading off Consistency and Dimensionality of Convex Surrogates for Multiclass Classification (NeurIPS 2024 · similarity 0.85)
- Generalization Bounds via Meta-Learned Model Representations: PAC-Bayes and Sample Compression Hypernetworks (ICML 2025 · similarity 0.84)
- A Universal Growth Rate for Learning with Smooth Surrogate Losses (NeurIPS 2024 · similarity 0.84)
- Leveraging Data Symmetries to Select an Optimal Subset of Training Data under Label Noise (NeurIPS 2026 · similarity 0.84)
- Nearest-Neighbor Radii under Dependent Sampling (NeurIPS 2026 · 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.