New algorithms efficiently value data for kNN classifiers

Researchers have developed new algorithms to efficiently calculate the Banzhaf value, a game-theoretic method for data valuation, specifically for k-nearest neighbors (kNN) classifiers. The study proves the computational hardness of the problem but introduces practical exact algorithms using dynamic programming, achieving pseudo-polynomial time complexity for weighted kNN and linear time complexity for unweighted kNN. Experiments on real-world datasets confirm the efficiency and effectiveness of these novel valuation methods. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Introduces more efficient methods for understanding data contributions, potentially improving model training and interpretability.

RANK_REASON Academic paper detailing a new algorithmic approach to a machine learning problem. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Banzhaf value
• data valuation
• k-nearest neighbors

• paper
• other