Ontology-guided Semantic Composition for Zero-Shot Learning (2020-06-30T00:00:00.000000Z)

TL;DR

This study proposes to model the compositional and expressive semantics of class labels by an OWL (Web Ontology Language) ontology, and further develop a new ZSL framework with ontology embedding.

Abstract

Zero-shot learning (ZSL) is a popular research problem that aims at predicting for those classes that have never appeared in the training stage by utilizing the inter-class relationship with some side information. In this study, we propose to model the compositional and expressive semantics of class labels by an OWL (Web Ontology Language) ontology, and further develop a new ZSL framework with ontology embedding. The effectiveness has been verified by some primary experiments on animal image classification and visual question answering.

Authors

Huajun Chen

2 papers

Jiaoyan Chen

17 papers

Jeff Z. Pan

7 papers

TL;DR

Abstract

Authors

References31 items

Explainable zero-shot learning via attentive graph convolutional network and knowledge graphs

Generative Adversarial Zero-shot Learning via Knowledge Graphs

Generative Adversarial Zero-Shot Relational Learning for Knowledge Graphs

Compositional Generalization in Image Captioning

ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks

OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge

EL Embeddings: Geometric construction of models for the Description Logic EL ++

Knowledge-based Transfer Learning Explanation

Rethinking Knowledge Graph Propagation for Zero-Shot Learning

Zero-Shot Recognition via Semantic Embeddings and Knowledge Graphs

Supporting shared hypothesis testing in the biomedical domain

Visual Question Answering as a Meta Learning Task

Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly

Semantic Autoencoder for Zero-Shot Learning

ConceptNet 5.5: An Open Multilingual Graph of General Knowledge

Latent Embeddings for Zero-Shot Classification

Predicting Deep Zero-Shot Convolutional Neural Networks Using Textual Descriptions

HermiT: An OWL 2 Reasoner

DeViSE: A Deep Visual-Semantic Embedding Model

Distributed Representations of Words and Phrases and their Compositionality

Semantic Compositionality through Recursive Matrix-Vector Spaces

TrOWL: Tractable OWL 2 Reasoning Infrastructure

Zero-shot Learning with Semantic Output Codes

Learning to detect unseen object classes by between-class attribute transfer

Information extraction from Wikipedia: moving down the long tail

DBpedia: A Nucleus for a Web of Open Data

Pushing the EL Envelope

WordNet: A Lexical Database for English

Embedding OWL Ontologies with OWL2Vec

jcel: A Modular Rule-based Reasoner

f-SWRL: A Fuzzy Extension of SWRL

Field of Study

Venue Information

Name

Type

URL

Alternate Names