AI Paper Atlas

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

PolySHAP: Extending KernelSHAP with Interaction-Informed Polynomial Regression

Fabian Fumagalli, R. Teal Witter, Christopher Musco

ICLR 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Shapley value attribution · 4 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
add pairwise (order-2) interaction terms to the regression (ablation)KernelSHAP (standard sampling) · 15 explanation gamesMSE of Shapley estimateImproved
include only 50% of order-2 interactions (ablation)2-PolySHAP · 15 explanation gamesMSEImproved
switch KernelSHAP to paired sampling vs 2-PolySHAP (ablation)KernelSHAP · 15 explanation gamesMSENo clear effect
add order-3 interactions under paired sampling (ablation)PolySHAP · Housing, Adult, Estate, Forest, CancerMSEImproved
add order-3 interactions in high dimensions (d>=60) (ablation)PolySHAP · CG60, IL60, NHANES, CrimeMSEMixed
add 4th-order terms on top of 3-PolySHAP under paired sampling (ablation)3-PolySHAP · 15 explanation gamesMSENo 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