Abstract:External knowledge (a.k.a. side information) plays a critical role in zero-shot learning (ZSL) which aims to predict with unseen classes that have never appeared in training data. Several kinds of external knowledge, such as text and attribute, have been widely investigated, but they alone are limited with incomplete semantics. Some very recent studies thus propose to use Knowledge Graph (KG) due to its high expressivity and compatibility for representing kinds of knowledge. However, the ZSL community is still in short of standard benchmarks for studying and comparing different external knowledge settings and different KG-based ZSL methods. In this paper, we proposed six resources covering three tasks, i.e., zero-shot image classification (ZS-IMGC), zero-shot relation extraction (ZS-RE), and zero-shot KG completion (ZS-KGC). Each resource has a normal ZSL benchmark and a KG containing semantics ranging from text to attribute, from relational knowledge to logical expressions. We have clearly presented these resources including their construction, statistics, data formats and usage cases w.r.t. different ZSL methods. More importantly, we have conducted a comprehensive benchmarking study, with a few classic and state-of-the-art methods for each task, including a method with KG augmented explanation. We discussed and compared different ZSL paradigms w.r.t. different external knowledge settings, and found that our resources have great potential for developing more advanced ZSL methods and more solutions for applying KGs for augmenting machine learning. All the resources are available at https://github.com/China-UK-ZSL/Resources_for_KZSL.

A review on multimodal zero‐shot learning

Knowledge-aware Zero-Shot Learning: Survey and Perspective

A Review of Generalized Zero-Shot Learning Methods

Domain Adaptation Meets Zero-Shot Learning: an Annotation-Efficient Approach to Multi-Modality Medical Image Segmentation

OntoZSL: Ontology-enhanced Zero-shot Learning

Manifold Regularized Cross-Modal Embedding for Zero-Shot Learning

Zero-shot and Few-shot Learning with Knowledge Graphs: A Comprehensive Survey

Fine-Grained Zero-Shot Learning: Advances, Challenges, and Prospects

Multi-modal Generative Adversarial Network for Zero-Shot Learning

Meta-Transfer Networks for Zero-Shot Learning

Recognizing Unseen Objects via Multimodal Intensive Knowledge Graph Propagation

Zero-shot Learning with Regularized Cross-Modality Ranking.

K-ZSL: Resources for Knowledge-driven Zero-shot Learning

Learn More from Less: Generalized Zero-Shot Learning with Severely Limited Labeled Data

Transductive Unbiased Embedding for Zero-Shot Learning

Benchmarking Knowledge-driven Zero-shot Learning

Automatic Metrics Learning with Low‐noise Embedding for Zero‐shot Learning

Semantic Softmax Loss for Zero-Shot Learning

Towards Zero-Shot Learning: A Brief Review and an Attention-Based Embedding Network

Transductive Multi-class and Multi-label Zero-shot Learning

Recent Advances in Zero-shot Recognition