One-Nearest-Neighbor Search is All You Need for Minimax Optimal Regression and Classification

This is an accompanying code for the paper "One-Nearest-Neighbor Search is All You Need for Minimax Optimal Regression and Classification" (arXiv:2202.02464).

Requirements

The experiments were run under the following environment:

python==3.8.1
matplotlib==3.3.4
numpy==1.20.0
pandas==1.2.1
py-cpuinfo==7.0.0
scikit-learn==0.24.1
scipy==1.6.0

Datasets

• To run each experiment with a real-world dataset, find the link from Table 2 of the paper and donwload the dataset under data/dataset-name/.

To replicate

• For the synthetic experiment (Section 5.1): run

python main_synthetic.py

• For the real-world dataset experiment (Section 5.2): run, e.g.,

python main.py --parallel True --test-size 0.05 --n-folds 10 --n-trials 10 --algorithm kd_tree --dataset SUSY

To be implemented

• Support node-level parallel computation.

Acknowledgments

• A cross validation code snippet was adapted from that of this repository.