Attention-Based Aggregation Graph Networks for Knowledge Graph Information Transfer

Zhao, Ming; Jia, Weijia; Huang, Yu‐Sheng

doi:10.1007/978-3-030-47436-2_41

Cited by 14 publications

(16 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To improve efficiency, some inductive methods adopt GNN to aggregate IKG neighbors to produce embeddings for OOKG entities (Hamaguchi et al, 2017;Wang et al, 2019;Bi et al, 2020;Zhao et al, 2020). These methods are effective but need relatively complex calculations.…”

Section: Related Workmentioning

confidence: 99%

“…In recent years, some inductive methods have been proposed for OOKG entities without retraining. Hamaguchi et al (2017); Wang et al (2019); Bi et al (2020); Zhao et al (2020) adopt Graph Neural Networks (GNN) to aggregate the IKG neighbors to represent the OOKG entities. These methods are effective but require relatively complex calculations, which could be simplified for higher efficiency.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions

Dai¹,

Zheng²,

Luo³

et al. 2021

Proceedings of the 6th Workshop on Representation Learning for NLP (RepL4NLP-2021)

View full text Add to dashboard Cite

Conventional Knowledge Graph Completion (KGC) assumes that all test entities appear during training. However, in real-world scenarios, Knowledge Graphs (KG) evolve fast with outof-knowledge-graph (OOKG) entities added frequently, and we need to efficiently represent these entities. Most existing Knowledge Graph Embedding (KGE) methods cannot represent OOKG entities without costly retraining on the whole KG. To enhance efficiency, we propose a simple and effective method that inductively represents OOKG entities by their optimal estimation under translational assumptions. Moreover, given pretrained embeddings of the in-knowledge-graph (IKG) entities, our method even needs no additional learning. Experimental results on two KGC tasks with OOKG entities show that our method outperforms the previous methods by a large margin with higher efficiency. 1 * Equal contribution.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions

Dai¹,

Zheng²,

Luo³

et al. 2021

Proceedings of the 6th Workshop on Representation Learning for NLP (RepL4NLP-2021)

View full text Add to dashboard Cite

show abstract

“…For instance, [4] chooses a semi-supervised strategy to predict the unseen entities based on their relations with the in-sample entities and absorb them into the KG. [273] places these operations as "transition" and "aggregation" modules.…”

Section: Graph/ontologymentioning

confidence: 99%

“…Graph modeling is a comprehensive task involving a variety of learning objects. When some objects are scarce or completely unseen, many LSL graph modeling methods attempt to exploit other supporting information to make up for the lack of partial training materials [4,226,237,264,273,274]. When the novel relations or entities have some correlation with the original elements in the KG, the metric-based method [48,139,236,260] can learn a reliable embedding space.…”

Section: Graph/ontologymentioning

confidence: 99%

What Can Knowledge Bring to Machine Learning? -- A Survey of Low-shot Learning for Structured Data

Hu¹,

Chapman²,

Wen³

et al. 2021

Preprint

View full text Add to dashboard Cite

Supervised machine learning has several drawbacks that make it difficult to use in many situations. Drawbacks include: heavy reliance on massive training data, limited generalizability and poor expressiveness of high-level semantics. Low-shot Learning attempts to address these drawbacks. Low-shot learning allows the model to obtain good predictive power with very little or no training data, where structured knowledge plays a key role as a high-level semantic representation of human. This article will review the fundamental factors of low-shot learning technologies, with a focus on the operation of structured knowledge under different low-shot conditions. We also introduce other techniques relevant to low-shot learning. Finally, we point out the limitations of low-shot learning, the prospects and gaps of industrial applications, and future research directions.

show abstract

“…There are also some recent works focusing on the OOKB problem in the KBC task recently.Some researchers focus on different type of tasks like link prediction [36], [37] in few-shot learning and entity detection [38] while we focus on triplet classification.Some researchers used the jointly embedding method [39] and multimodal data enhanced representation [40] to achieve OOKB entity embedding while in our work these external data are not considered.Reference [41] used attention-based aggregation to solve the new OOKB relation problem.Their idea and method are exciting and we want to extend our work to the OOKB relation problem in the future.Many studies assumed a specific scenario while our work considers only the standard scenario of the This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.…”

Section: Related Workmentioning

confidence: 99%

Knowledge Transfer for Out-of-Knowledge-Base Entities: Improving Graph-Neural-Network-Based Embedding Using Convolutional Layers

Zhang

Zhou

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Knowledge base completion (KBC) aims to predict missing information in a knowledge base. Most existing embedding-based KBC models assume that all test entities are available at training time. Thus, a question arises-that is, how to answer queries concerning test entities not observed at training time, which is called the out-of-knowledge-base (OOKB) entity problem. In this paper, we propose a parameter-efficient embedding model that combines the benefits of a graph neural network (GNN) and a convolutional neural network (CNN) to solve the KBC task with OOKB entities. First, we apply the GNN architecture to learn the information between nodes in the graph. Second, convolution layers are used as a transition matrix in GNN to learn more expressive embeddings with fewer parameters. Finally, we use a transition-based knowledge graph embedding model to solve the KBC task. The model has learnable weights that adapt based on information from neighbors and can exploit auxiliary knowledge for OOKB entities to compute their embedding while remaining parameter efficient. We demonstrate the effectiveness of the proposed model on OOKB datasets, and the code is available at https://github.com/Tianchen627/Knowledge-Transferfor-Out-of-Knowledge-Base-Entities.

show abstract

Attention-Based Aggregation Graph Networks for Knowledge Graph Information Transfer

Cited by 14 publications

References 14 publications

Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions

Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions

What Can Knowledge Bring to Machine Learning? -- A Survey of Low-shot Learning for Structured Data

Knowledge Transfer for Out-of-Knowledge-Base Entities: Improving Graph-Neural-Network-Based Embedding Using Convolutional Layers

Contact Info

Product

Resources

About