Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
To resolve the interpretability bottleneck in hierarchical multi-label classification, the authors develop HyDI, a neuro-symbolic ensemble that couples deep neural networks with Inductive Logic Programming (ILP). By substituting black-box neural predictions at the taxonomy's leaf nodes with explicit logic-based classifiers, the framework achieves transparent decision-making at granular classification levels. Evaluated on the Chemical Entities of Biological Interest (ChEBI) ontology across 314 classes, the system successfully outputs both global rule-based logic and multimodal (visual and text) local explanations.
Swapping deep networks for symbolic logic rules specifically at taxonomic leaf nodes delivers verifiable, human-readable explanations across hundreds of granular classes without sacrificing hierarchical context.
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.