HackRL: Reinforcement learning with hierarchical attention for cross-graph knowledge fusion and collaborative reasoning

doi:10.1016/j.knosys.2021.107498

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 先进控制与自动化团队

	HackRL: Reinforcement learning with hierarchical attention for cross-graph knowledge fusion and collaborative reasoning
	Yang, Linyao 3,6; Wang, Xiao 4,6; Dai, Yuxin 5; Xin, Kejun 1,4; Zheng, Xiaolong 6; Ding, Weiping 2; Zhang, Jun 4,5,6; Wang, Fei-Yue 4,6
刊名	KNOWLEDGE-BASED SYSTEMS
	2021-12-05
卷号	233 页码:14
关键词	Knowledge fusion Knowledge reasoning Decision-making Hierarchical graph attention Reinforcement learning
ISSN号	0950-7051
DOI	10.1016/j.knosys.2021.107498
通讯作者	Wang, Xiao(x.wang@ia.ac.cn)
英文摘要	Reasoning aiming at inferring implicit facts over knowledge graphs (KGs) is a critical and fundamental task for various intelligent knowledge-based services. With multiple distributed and complementary KGs, the effective and efficient capture and fusion of knowledge from different KGs is becoming an increasingly important topic, which has not been well studied. To fill this gap, we propose to explore cross-KG relation paths with the anchor links identified by entity alignment for the knowledge fusion and collaborative reasoning of multiple KGs. To address the heterogeneity of different KGs, this paper proposes a novel reasoning model named HackRL based on the reinforcement learning framework, which incorporates the long short-term memory and hierarchical graph attention in the policy network to infer indicative cross-KG relation paths from the history trajectory and the heterogeneous environment for predicting corresponding relations. Meanwhile, an entity alignment oriented representation learning method is utilized to embed different KGs into a unified vector space based on the anchor links to reduce the impact of distinct vector spaces, and two training mechanisms, action mask and retrain with sampled paths, are proposed to optimize the training process to learn more successful indicative paths. The proposed HackRL is validated on three cross-lingual datasets built from DBpedia on the link prediction and fact prediction tasks. Experimental results demonstrate that HackRL achieves better performance on most tasks than existing methods. This work provides an industrially-applicable framework for fusing distributed KGs to make better decisions. (c) 2021 Elsevier B.V. All rights reserved.
WOS关键词	FRAMEWORK
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000709919000011
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/46264]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_先进控制与自动化团队
通讯作者	Wang, Xiao
作者单位	1.Nanjing Joinmap Data Res Inst, Nanjing 211100, Peoples R China 2.Nantong Univ, Sch Informat Sci & Technol, Nantong 226019, Peoples R China 3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 4.Qingdao Acad Intelligent Ind, Qingdao 266000, Peoples R China 5.Wuhan Univ, Sch Elect Engn & Automat, Wuhan 430072, Peoples R China 6.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Yang, Linyao,Wang, Xiao,Dai, Yuxin,et al. HackRL: Reinforcement learning with hierarchical attention for cross-graph knowledge fusion and collaborative reasoning[J]. KNOWLEDGE-BASED SYSTEMS,2021,233:14.
APA	Yang, Linyao.,Wang, Xiao.,Dai, Yuxin.,Xin, Kejun.,Zheng, Xiaolong.,...&Wang, Fei-Yue.(2021).HackRL: Reinforcement learning with hierarchical attention for cross-graph knowledge fusion and collaborative reasoning.KNOWLEDGE-BASED SYSTEMS,233,14.
MLA	Yang, Linyao,et al."HackRL: Reinforcement learning with hierarchical attention for cross-graph knowledge fusion and collaborative reasoning".KNOWLEDGE-BASED SYSTEMS 233(2021):14.