Two new research papers published on arXiv propose novel methods for estimating Shapley values, a key technique in machine learning for feature attribution. The first paper introduces FUSHAP, designed to handle multi-site studies with missing data by leveraging auxiliary sites to reduce variance without imputation. The second paper presents a semiparametric inference method for conditional Shapley values, offering uncertainty quantification and addressing feature dependency issues. Both methods demonstrate improved accuracy and bias reduction compared to existing approaches in synthetic and real-world datasets. AI
IMPACT These advancements in Shapley value estimation could lead to more reliable and interpretable machine learning models, particularly in complex, multi-site, or dependent feature scenarios.
RANK_REASON Two academic papers published on arXiv presenting new methodologies.
- Bonferroni
- K-fold cross-fitting
- Sage
- Shapley Values
- UCI Concrete
- U-statistic
- arXiv
- FUSHAP
- Shapley value
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
