AI Paper Atlas

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

The Utility of “Even if” Semifactual Explanation to Optimise Positive Outcomes

Eoin M. Kenny, Weipeng Fuzzy Huang

NeurIPS 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Counterfactual explanations and recourse · 2 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
Account for causal dependencies when computing gain (ablation)S-GEN · Adultgain: $0.055 \pm 0.001$ v. $0.063 \pm 0.001$Improved
S-GEN vs Karimi et al. on diversityS-GEN · causal experimentdiversity: $0.84 \pm 0.09$ v. $0.43 \pm 0.03$Improved
S-GEN vs Domineguez et al. on diversityS-GEN · causal experimentdiversity: $0.84 \pm 0.09$ v. $0.34 \pm 0.03$Improved
S-GEN vs Domineguez et al. on robustnessS-GEN · causal experimentrobustness success: $87\%$ success v. $54\% $ successImproved
S-GEN vs Karimi et al. on robustnessS-GEN · causal experimentrobustness success: $7.2\%$Improved
Semifactual (S-GEN) instead of counterfactual (DiCE) explanationS-GEN · User study, loan acceptedusefulness rating: 3.60$\pm$0.27 v. DiCE=2.33$\pm$0.34Improved
Semifactual (S-GEN) instead of counterfactual (DiCE) explanationS-GEN · User study, loan rejectedusefulness rating: 2.6$\pm$0.32 v. DiCE=4.53$\pm$0.17Hurt
S-GEN vs DiCE* and PIECE* baselines on non-causal testsS-GEN · non-causal, averaged datasetsgain, diversity, robustness etc.Improved
Traditional counterfactual approach DiCE* for gain maximizationDiCE* · non-causalgainHurt
S-GEN vs baselines on individual datasetsS-GEN · 16 individual dataset testsmultiple: significantly better on 11/16 testsMixed

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