Database Paper Browser

Back to papers

P-Shapley: Shapley Values on Probabilistic Classifiers

Summary: Introduces P-Shapley, a Shapley-value formulation for classification that uses model-predicted probabilities (with convex calibration functions) as utility instead of binary accuracy. Theoretically and empirically yields more stable, more discriminative data valuations. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID
13413
Venue
VLDB
Year
2024
Pagerank
4.6089395e-05
Overall Rank
7,935 | 44.86%
DOI
10.14778/3654621.3654638

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank Citing Paper Year Venue Pagerank
10,403 Shapley Value Estimation Based on Differential Matrix 2025 SIGMOD 4.1905499e-05
10,663 A Comprehensive Study of Shapley Value in Data Analytics 2025 VLDB 4.1905499e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 14 of 14 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers