Disentangled Ontology Embedding for Zero-shot Learning

Knowledge Graph (KG) and its variant of ontology have been widely used forrepresentation, and have shown to be quite effective in augmenting-shot Learning (ZSL). However, existing ZSL methods that utilize KGs allthe intrinsic complexity of inter-class relationships represented in. One typical feature is that a class is often related to other classes insemantic aspects. In this paper, we focus on ontologies forZSL, and propose to learn disentangled ontology embeddings guided byproperties to capture and utilize more fine-grained classin different aspects. We also contribute a new ZSL frameworkDOZSL, which contains two new ZSL solutions based on generative modelsgraph propagation models, respectively, for effectively utilizing theontology embeddings. Extensive evaluations have been conducted onbenchmarks across zero-shot image classification (ZS-IMGC) and zero-shotcompletion (ZS-KGC). DOZSL often achieves better performance than the-of-the-art, and its components have been verified by ablation studies andstudies. Our codes and datasets are available at://github.com/zjukg/DOZSL.