Research map / Learning theory and algorithms / Combinatorial and search optimization
Sufficient Reasons for Classifier Decisions in the Presence of Domain Constraints
Niku Gorji, Sasha Rubin
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Treat constraint-violating instances as undefined rather than ignoring the constraint | Boolean classifier sufficient reasons · Constrained decision functions | reason succinctness | Improved |
| Represent constraints using implication (κ → φ), rather than ignoring them | Boolean classifier sufficient reasons · Constrained decision functions | reason parsimony | Improved |
| Represent constraints using conjunction (κ ∧ φ), rather than implication | Boolean classifier sufficient reasons · Constrained decision functions | reason parsimony | Hurt |
| Use one-hot constraints to simplify explanations | CRASH trial RPART decision tree · CRASH trial, selected positive instance | reason size | Improved |
| Use the no-overlap constraint for X and O pieces | Tic-Tac-Toe RPART decision tree · Tic-Tac-Toe, selected positive instance | number of reasons: 8 -> 11 | Improved |
| Use valid-play constraints, including move order and alternation | Tic-Tac-Toe RPART decision tree · Tic-Tac-Toe, selected positive instance | number of reasons: 46 | Mixed |
Closest papers by meaning
- On the Computation of Necessary and Sufficient Explanations (AAAI 2022 · similarity 0.86)
- Tractable Explanations for d-DNNF Classifiers (AAAI 2022 · similarity 0.84)
- Consistent Sufficient Explanations and Minimal Local Rules for explaining the decision of any classifier or regressor (NeurIPS 2022 · similarity 0.84)
- Estimating Accuracy from Unlabeled Data: A Probabilistic Logic Approach (NeurIPS 2017 · similarity 0.83)
- On Computing Probabilistic Explanations for Decision Trees (NeurIPS 2022 · similarity 0.83)
- Decomposing Constraint Networks for Calculating c-Representations (AAAI 2024 · similarity 0.82)
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.
