Research map / Adversarial robustness and security / Certified and empirical robustness
Formalising the Robustness of Counterfactual Explanations for Neural Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Apply Δ-robustness-guided search to Wachter et al. | Wachter et al. · diabetes, no2, SBA | vm2: 0% -> 100%; 32% -> 100%; 92% -> 92% | Mixed |
| Apply Δ-robustness-guided search to Proto | Proto · diabetes, no2, SBA, credit | vm2: 18% -> 96%; 32% -> 100%; 6% -> 88%; 22% -> 30% | Helped |
| Apply Δ-robustness-guided search to MILP | MILP · diabetes, no2, SBA, credit | vm2: 0% -> 100%; 32% -> 100%; 4% -> 100%; 74% -> 100% | Helped |
| Use Δ-robustness-guided CFX search instead of the base method | Wachter et al., Proto, MILP · diabetes, no2, SBA, credit | ℓ1: diabetes: 0.051 -> 0.122, 0.063 -> 0.104, 0.049 -> 0.212; no2: 0.035 -> 0.084, 0.036 -> 0.069, 0.032 -> 0.059; SBA: 0.01 | Mixed |
| Use Δ-robustness-guided CFX search rather than ROAR | Wachter et.-R, Proto-R, MILP-R · diabetes, no2, SBA, credit | vm2: diabetes: 100% vs 14%, 96% vs 14%, 100% vs 14%; no2: 100% vs 34%; SBA: 92% vs 78%, 88% vs 78%, 100% vs 78%; credit: 30% | Mixed |
| Require Δ-robustness during the search, increasing allowable counterfactual distance until verification succeeds | Wachter et.-R, Proto-R, MILP-R · four datasets | Δ-validity | Helped |
Closest papers by meaning
- Promoting Counterfactual Robustness through Diversity (AAAI 2024 · similarity 0.89)
- CF-OPT: Counterfactual Explanations for Structured Prediction (ICML 2024 · similarity 0.87)
- Faithful Model Explanations through Energy-Constrained Conformal Counterfactuals (AAAI 2024 · similarity 0.86)
- Generating Likely Counterfactuals Using Sum-Product Networks (ICLR 2025 · similarity 0.86)
- GNNX-BENCH: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarking (ICLR 2024 · similarity 0.85)
- Interpreting Robustness Proofs of Deep Neural Networks (ICLR 2024 · similarity 0.85)
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.