AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Model Shapley: Equitable Model Valuation with Black-box Access

Xinyi Xu, Thanh Lam, Chuan-Sheng Foo, Bryan Kian Hsiang Low

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Counterfactual and statistical explanations · 0 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
learn on alpha_i features instead of average prediction h_i (ablation)GPR model appraiser · MNIST, 20% trainMSE: 8.36e−5 -> 1.59e−6Improved
learn on alpha_i features instead of average prediction h_i (ablation)GPR model appraiser · MNIST, 20% trainMaxE: 1.59e−2 -> 3.53e−3Improved
learn on alpha_i features instead of average prediction h_i (ablation)GPR model appraiser · CIFAR-10, 20% trainMSE: 3.05e−4 -> 1.79e−5Improved
learn on alpha_i features instead of average prediction h_i (ablation)GPR model appraiser · CIFAR-10, 20% trainMaxE: 3.23e−2 -> 9.05e−3Improved
raise training ratio of model-MSV pairs from 5% to 50%GPR model appraiser · MNIST and CIFAR-10test errorImproved
use CNN instead of MLP or LRMSV · MNIST misclassified queriesMSVImproved
multiply top-class probability by 1, 5, 10 (higher predictive certainty)MSV · MNIST CNN modelsMSVImproved
use class-specific Dirichlet abstractions instead of unpartitioned query set (ablation)MSV · KDD99 imbalancedconsistency with F1Improved
buy models in highest-MSV-first order instead of lowest-firstrandom forest of decision trees · Breast Cancertest accuracy vs number of modelsImproved
buy models in highest-MSV-first order instead of lowest-firstvoting ensemble of LeNets · CIFAR-10test accuracy vs number of modelsImproved

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