HyDI: A hybrid Deep Learning-Inductive Logic Programming ensemble for multi-label classification
cs.LG, cs.LO
Submitted: 2026-09-29
Updated: 2026-09-29
Comments: Accepted at IJCLR26 (6th International Joint Conference on Learning & Reasoning, 16-18 September 2026)
Code: https://github.com/bramucas/xclingo2
License: http://creativecommons.org/licenses/by/4.0/
The gist: While attaining remarkable results for many applications, Deep Learning models are notoriously difficult to explain.
Abstract
While attaining remarkable results for many applications, Deep Learning models are notoriously difficult to explain. This work introduces HyDI, a hybrid ensemble architecture for hierarchical multi-label classification. It combines a Deep Learning (DL) model with rule-based classifiers generated by Inductive Logic Programming (ILP). For leaf classes of the label hierarchy, the rule-based classifiers replace the DL model, leading to more transparent classification results. HyDI is applied to the Chemical Entities of Biological Interest (ChEBI) ontology, providing ILP-generated rules for 314 classes. For these classes, HyDI can generate global explanations as well as local explanations that combine visual and text-based descriptions.
Sources
- ChemLog: Making MSOL Viable for Ontological Classification and Learning
- The ILASP system for Inductive Learning of Answer Set Programs
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks