Research map / Optimization / Nonconvex and smooth optimization
Characterization of the Convex Łukasiewicz Fragment for Learning From Constraints
Francesco Giannini, Michelangelo Diligenti, Marco Gori, Marco Maggini
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 |
|---|---|---|---|
| add collective classification over the knowledge base on top of neural network outputs | Feedforward network (30 hidden units per predicate) · Winston animal image benchmark (3325 ImageNet images, 7 classes), 10-90% supervisions | test accuracy | Improved |
| initialize grounded predicates with neural network outputs instead of only training supervisions (others set to 0.5) (ablation) | Collective classification with convex Lukasiewicz constraints · Winston animal image benchmark, 10-90% supervisions | test accuracy | Improved |
| increase fraction of available training supervisions toward 90% (ablation) | Collective classification (supervisions vs neural network prior) · Winston animal image benchmark | gap in test accuracy between the two collective settings | No clear effect |
Closest papers by meaning
- A Constraint-Based Approach to Learning and Explanation (AAAI 2020 · similarity 0.85)
- An Integer Linear Programming Framework for Mining Constraints from Data (ICML 2021 · similarity 0.84)
- Near-Optimal Solutions of Constrained Learning Problems (ICLR 2024 · similarity 0.83)
- Generalized Convergence Analysis of Tsetlin Automaton Based Algorithms: A Probabilistic Approach to Concept Learning (AAAI 2025 · similarity 0.82)
- On the Modelling of Constraints with Tractable Logical Operators (AAAI 2025 · similarity 0.82)
- Learning to Learn in Interactive Constraint Acquisition (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.
