Research map / Architectures and efficiency / RNNs and normalization
Multi-value Rule Sets for Interpretable Classification with Feature-Efficient Representations
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 multi-value rules with Bayesian prior instead of single-value rule models | rule-set classifier · Credit card | number of conditions: 15 -> 10 | Helped |
| use MRS instead of single-value rule baselines | rule-based classifiers · six real-world datasets | number of unique features | Helped |
| use MRS instead of single-value rule baselines | rule-based classifiers · six real-world datasets | accuracy | No clear effect |
| use MRS instead of CBA and SBRL | rule-based classifiers · Readmission | accuracy: 6 conditions vs 39 (CBA), 21 (SBRL) | Hurt |
| use MRS instead of XGBoost black box | classifier · Census | accuracy: .80 vs .87 | Hurt |
| increase prior penalty on model size (larger beta_M and beta_L) (ablation) | MRS · synthetic data, 100k observations | hold-out error | Hurt |
| use MRS instead of CBA and Ripper | rule models · human evaluation, credit card models | response time | Helped |
| ordered rule list (SBRL) instead of unordered rule sets | rule models · human evaluation | human accuracy | Hurt |
Closest papers by meaning
- Scalable Rule-Based Representation Learning for Interpretable Classification (NeurIPS 2021 · similarity 0.85)
- Interpretability with full complexity by constraining feature information (ICLR 2023 · similarity 0.84)
- Learning Interpretable Decision Rule Sets: A Submodular Optimization Approach (NeurIPS 2021 · similarity 0.84)
- Explainable Neural Networks with Guarantee: A Sparse Estimation Approach (AAAI 2025 · similarity 0.81)
- Large Language Models are Interpretable Learners (ICLR 2025 · similarity 0.81)
- Differentiable Rule Induction from Raw Sequence Inputs (ICLR 2025 · similarity 0.81)
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.