Research map / Learning theory and algorithms / Fairness
Certifying Fairness of Probabilistic Circuits
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 |
|---|---|---|---|
| Use branch-and-bound pruning instead of naive enumeration | Top-k pattern search on probabilistic circuits · COMPAS, Income, and Adult | search speedup: 2.73x; 1.22x; 24.20x | Helped |
| Use sampling rather than naive enumeration to find the top-1 pattern | Pattern mining on probabilistic circuits · COMPAS, Income, and Adult | average speedup: 26x; 143x; 49480x | Helped |
| Use sampling instead of exact search under a fixed timeout | Strudel PCs · COMPAS, Income, and Adult | discrimination patterns found: 347 -> 751; 209 -> 1090; 37167 -> 113763 | Helped |
| Use sampling instead of exact search under a fixed timeout | Strudel PCs · Income, δ=0.05, 5s timeout | highest discrimination score: 0.1076 -> 0.1658 | Helped |
| Use sampling instead of exact search under a fixed timeout | Strudel PCs · Adult, δ=0.05, 600s timeout | highest discrimination score: 0.6725 -> 0.6871 | Helped |
| Use sampling instead of exact search under a fixed timeout | Strudel PCs · COMPAS, δ=0.05, 3s timeout | highest discrimination score: 0.2236 -> 0.2230 | Hurt |
| Use sampled Pareto summaries in place of exact Pareto summaries | COMPAS pattern mining · COMPAS | Pareto optimal patterns recovered: 27 of 38 | Mixed |
| Apply fair data repair | COMPAS probabilistic circuits · COMPAS | discrimination patterns at δ=0.05: 3866; 1320; 488; 659; 894; 578 | Mixed |
Closest papers by meaning
- Learning Fair Naive Bayes Classifiers by Discovering and Eliminating Discrimination Patterns (AAAI 2020 · similarity 0.91)
- Group Fairness by Probabilistic Modeling with Latent Fair Decisions (AAAI 2021 · similarity 0.89)
- Bayes-Optimal Fair Classification with Multiple Sensitive Features (AAAI 2026 · similarity 0.88)
- Reconciling Predictive and Statistical Parity: A Causal Approach (AAAI 2024 · similarity 0.87)
- Interventional Fairness on Partially Known Causal Graphs: A Constrained Optimization Approach (ICLR 2024 · similarity 0.86)
- Individual Fairness In Strategic Classification (NeurIPS 2025 · similarity 0.86)
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.