AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

On the Impact of the Utility in Semivalue-based Data Valuation

Mélissa Tamine, Benjamin Heymann, Maxime Vono, Patrick Loiseau

ICLR 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Adaptive testing and experiments · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Banzhaf weights instead of Shapley when switching utility from accuracy to F1 (ablation)semivalue data valuation · TitanicKendall rank correlation: -0.19 -> 0.94Improved
use Banzhaf weights instead of Shapley (ablation)semivalue data valuation · CreditKendall rank correlation: -0.47 -> 0.87Improved
use Banzhaf weights instead of Shapley (ablation)semivalue data valuation · PolKendall rank correlation: 0.67 -> 0.40Hurt
use Banzhaf weights instead of Shapley or Beta Shapley (ablation)semivalue data valuation · binary classification datasets, p-robustness R_pR_pImproved
use Banzhaf weights in utility trade-off scenario (ablation)semivalue data valuation · regression and multiclass datasetsR_pImproved
use (4,1)-Beta Shapley instead of Shapley (ablation)semivalue data valuation · TitanicKendall rank correlation: -0.19 -> -0.17No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy