Research map / Learning theory and algorithms / Combinatorial and search optimization
Tractable Explanations for d-DNNF Classifiers
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 |
|---|---|---|---|
| enumerate AXps/CXps exactly with SAT oracle calls on d-DNNF instead of heuristic Anchor | Algorithm 3 on d-DNNF · 19 binary datasets | avg runtime: 3.07 -> 0.013 (chess) | Helped |
| enumerate on parity5+5 d-DNNF vs Anchor | Algorithm 3 · parity5+5 | avg runtime: 18.64 -> 0.056 | Helped |
| enumerate on promoters (334 features) d-DNNF vs Anchor | Algorithm 3 · promoters | avg runtime: 57.72 -> 0.003 | Helped |
| explain SDD instead of d-DNNF representation of the same classifier | Algorithm 3 · 19 datasets | max runtime: < 1.0 sec -> < 0.2 sec | Helped |
| explain SDD instead of d-DNNF (average total enumeration) | Algorithm 3 · 19 datasets | avg runtime: 0.4 sec -> 0.05 sec | Helped |
| SDD vs d-DNNF size for same RODT | compilation · 19 datasets | model size | Helped |
| SDD vs d-DNNF on tic tac toe | Algorithm 3 · tic tac toe | avg runtime: 0.096 -> 0.034 | Helped |
| SDD vs d-DNNF on compas | Algorithm 3 · compas | avg runtime: 0.309 -> 0.037 | Helped |
| SAT oracle calls in enumeration | Algorithm 3 · 19 datasets | runtime | No clear effect |
Closest papers by meaning
- New Compilation Languages Based on Restricted Weak Decomposability (AAAI 2025 · similarity 0.83)
- Data-Efficient Learning with Neural Programs (NeurIPS 2024 · similarity 0.83)
- A Compiler for Weak Decomposable Negation Normal Form (AAAI 2024 · similarity 0.83)
- A General Theoretical Framework for Learning Smallest Interpretable Models (AAAI 2024 · similarity 0.83)
- Neural Reasoning Networks: Efficient Interpretable Neural Networks with Automatic Textual Explanations (AAAI 2025 · similarity 0.83)
- Tensor Decomposition Meets Knowledge Compilation: A Study Comparing Tensor Trains with OBDDs (AAAI 2025 · 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.