Lundberg and Lee: a unified approach to interpreting model predictions

What changed

Defined the SHAP framework for local feature attribution, identified a class of additive feature-attribution methods and gave uniqueness results under the paper’s stated properties.

Scientific scope

The record is the exact NIPS 2017 two-author paper. Shapley’s game-theoretic value, LIME, DeepLIFT, later TreeSHAP implementations and claims about causal explanation or universal faithfulness remain separate.

Evidence

The linked sources identify this contribution and its scope. WRS has not independently reconstructed the scientific claims. The record is the exact NIPS 2017 two-author paper. Shapley’s game-theoretic value, LIME, DeepLIFT, later TreeSHAP implementations and claims about causal explanation or universal faithfulness remain separate.

Contributors

• Scott M. Lundberg · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.
• Su-In Lee · SCIENCE: Named source author of the scoped work; individual effort shares are not measured.

• NeurIPS paper for the SHAP framework
• NeurIPS record for the SHAP framework

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction