English

Efficient Data Shapley for Weighted Nearest Neighbor Algorithms

Data Structures and Algorithms 2024-01-23 v1 Machine Learning Machine Learning

Abstract

This work aims to address an open problem in data valuation literature concerning the efficient computation of Data Shapley for weighted KK nearest neighbor algorithm (WKNN-Shapley). By considering the accuracy of hard-label KNN with discretized weights as the utility function, we reframe the computation of WKNN-Shapley into a counting problem and introduce a quadratic-time algorithm, presenting a notable improvement from O(NK)O(N^K), the best result from existing literature. We develop a deterministic approximation algorithm that further improves computational efficiency while maintaining the key fairness properties of the Shapley value. Through extensive experiments, we demonstrate WKNN-Shapley's computational efficiency and its superior performance in discerning data quality compared to its unweighted counterpart.

Keywords

Cite

@article{arxiv.2401.11103,
  title  = {Efficient Data Shapley for Weighted Nearest Neighbor Algorithms},
  author = {Jiachen T. Wang and Prateek Mittal and Ruoxi Jia},
  journal= {arXiv preprint arXiv:2401.11103},
  year   = {2024}
}

Comments

AISTATS 2024 Oral

R2 v1 2026-06-28T14:22:16.543Z