Hongjun Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hongjun Zhang has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
Chunlin Liao
School of Graduate, Army Engineering University of PLA, Nanjing, China
1 shared publication- Kai Cheng
Army Engineering University of PLA
1 shared publication - Kai Cheng
Command and Control Engineering College, Army Engineering University of PLA,Nanjing,China,210000
1 shared publication - Kai Kang
Command and Control Engineering College, Army Engineering University of PLA,Nanjing,China,210000
1 shared publication - Ke Zhang
The 63rd Research Institute of National University of Defense Technology,Nanjing,China,210000
1 shared publication - Ke Zhang
National University of Defense Technology
1 shared publication
Latest Publications
- Open Access
- 21 October 2025
- Published by SAGE Publications in Science Progress
- Vol. 108 (4)
A decision-making framework using MCTS as a hierarchical task network and deep learning connector
Currently, purely deep learning-based agents struggle to make optimal decisions within a short timeframe in problems with a vast decision-making space. Human planning knowledge is required to assist agents in making better decisions. This manuscript proposes a novel knowledge-guided and data-driven decision-making framework, utilizing hierarchical task network as the carrier of knowledge, deep learning as the trainer for data, and the Monte Carlo Tree Search as the connector between hierarchical task network and deep learning. The experiments on the MiniRTS environment validated that the proposed framework in this manuscript can replace humans in collecting high-quality data, and it can train neural networks that perform equally well as the compared network even with only 20% of the available data, which provide a new direction for future research.
- 21 October 2025
- Published by SAGE Publications in Science Progress
- Vol. 108 (4)
Open AccessA decision-making framework using MCTS as a hierarchical task network and deep learning connector
Currently, purely deep learning-based agents struggle to make optimal decisions within a short timeframe in problems with a vast decision-making space. Human planning knowledge is required to assist agents in making better decisions. This manuscript proposes a novel knowledge-guided and data-driven decision-making framework, utilizing hierarchical task network as the carrier of knowledge, deep learning as the trainer for data, and the Monte Carlo Tree Search as the connector between hierarchical task network and deep learning. The experiments on the MiniRTS environment validated that the proposed framework in this manuscript can replace humans in collecting high-quality data, and it can train neural networks that perform equally well as the compared network even with only 20% of the available data, which provide a new direction for future research.
- 1 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Systems Engineering and Electronics
- Vol. 35 (5) , 1264-1275
Planning, Monitoring and Replanning Techniques for Handling Abnormity in HTN-Based Planning and Execution
A framework that integrates planning, monitoring and replanning techniques is proposed. It can devise the best solution based on the current state according to specific objectives and properly deal with the influence of abnormity on the plan execution. The framework consists of three parts: the hierarchical task network (HTN) planner based on Monte Carlo tree search (MCTS), hybrid plan monitoring based on forward and backward and norm-based replanning method selection. The HTN planner based on MCTS selects the optimal method for HTN compound task through pre-exploration. Based on specific objectives, it can identify the best solution to the current problem. The hybrid plan monitoring has the capability to detect the influence of abnormity on the effect of an executed action and the premise of an unexecuted action, thus trigger the replanning. The norm-based replanning selection method can measure the difference between the expected state and the actual state, and then select the best replanning algorithm. The experimental results reveal that our method can effectively deal with the influence of abnormity on the implementation of the plan and achieve the target task in an optimal way.
- 1 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Systems Engineering and Electronics
- Vol. 35 (5) , 1264-1275
Planning, Monitoring and Replanning Techniques for Handling Abnormity in HTN-Based Planning and Execution
A framework that integrates planning, monitoring and replanning techniques is proposed. It can devise the best solution based on the current state according to specific objectives and properly deal with the influence of abnormity on the plan execution. The framework consists of three parts: the hierarchical task network (HTN) planner based on Monte Carlo tree search (MCTS), hybrid plan monitoring based on forward and backward and norm-based replanning method selection. The HTN planner based on MCTS selects the optimal method for HTN compound task through pre-exploration. Based on specific objectives, it can identify the best solution to the current problem. The hybrid plan monitoring has the capability to detect the influence of abnormity on the effect of an executed action and the premise of an unexecuted action, thus trigger the replanning. The norm-based replanning selection method can measure the difference between the expected state and the actual state, and then select the best replanning algorithm. The experimental results reveal that our method can effectively deal with the influence of abnormity on the implementation of the plan and achieve the target task in an optimal way.
- Open Access
- 5 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 110670-110680
EwLoss: A Exponential Weighted Loss Function for Knowledge Graph Embedding Models
Knowledge graphs have been widely used in various domains such as intelligent question answering, information retrieval, transportation, medicine, e-commerce, and others, and have achieved significant success. To support these applications with a high-quality vector representation, numerous studies have proposed various knowledge graph embedding (KGE) models. However, most of these models primarily focus on the design of new scoring functions, disregarding the crucial role of the loss function during the model training stage. In light of this, we propose a new exponential weight coefficient loss function, named EwLoss, which can be incorporated into any typical KGE model. This loss function enables the model to better distinguish between triplets that are otherwise difficult to differentiate. Consequently, the model can learn the features of these challenging triplets more effectively, leading to improved embedding vectors. By conducting experiments on four commonly used data sets, where EwLoss is applied to six representative models, we demonstrate the superiority of EwLoss in distinguishing such hard-to-distinguish triplets. Moreover, we observe that the performance of the embedding models is significantly enhanced when utilizing EwLoss.
- 5 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 110670-110680
Open AccessEwLoss: A Exponential Weighted Loss Function for Knowledge Graph Embedding Models
Knowledge graphs have been widely used in various domains such as intelligent question answering, information retrieval, transportation, medicine, e-commerce, and others, and have achieved significant success. To support these applications with a high-quality vector representation, numerous studies have proposed various knowledge graph embedding (KGE) models. However, most of these models primarily focus on the design of new scoring functions, disregarding the crucial role of the loss function during the model training stage. In light of this, we propose a new exponential weight coefficient loss function, named EwLoss, which can be incorporated into any typical KGE model. This loss function enables the model to better distinguish between triplets that are otherwise difficult to differentiate. Consequently, the model can learn the features of these challenging triplets more effectively, leading to improved embedding vectors. By conducting experiments on four commonly used data sets, where EwLoss is applied to six representative models, we demonstrate the superiority of EwLoss in distinguishing such hard-to-distinguish triplets. Moreover, we observe that the performance of the embedding models is significantly enhanced when utilizing EwLoss.
- Open Access
- 1 January 2014
- Published by Wiley in Mathematical Problems in Engineering
- Vol. 2014 (1) , 1-16
Quantum‐Inspired Evolutionary Algorithm for Continuous Space Optimization Based on Multiple Chains Encoding Method of Quantum Bits
This study proposes a novel quantum evolutionary algorithm called four‐chain quantum‐inspired evolutionary algorithm (FCQIEA) based on the four gene chains encoding method. In FCQIEA, a chromosome comprises four gene chains to expand the search space effectively and promote the evolutionary rate. Different parameters, including rotational angle and mutation probability, have been analyzed for better optimization. Performance comparison with other quantum‐inspired evolutionary algorithms (QIEAs), evolutionary algorithms, and different chains of QIEA demonstrates the effectiveness and efficiency of FCQIEA.
- 1 January 2014
- Published by Wiley in Mathematical Problems in Engineering
- Vol. 2014 (1) , 1-16
Open AccessQuantum‐Inspired Evolutionary Algorithm for Continuous Space Optimization Based on Multiple Chains Encoding Method of Quantum Bits
This study proposes a novel quantum evolutionary algorithm called four‐chain quantum‐inspired evolutionary algorithm (FCQIEA) based on the four gene chains encoding method. In FCQIEA, a chromosome comprises four gene chains to expand the search space effectively and promote the evolutionary rate. Different parameters, including rotational angle and mutation probability, have been analyzed for better optimization. Performance comparison with other quantum‐inspired evolutionary algorithms (QIEAs), evolutionary algorithms, and different chains of QIEA demonstrates the effectiveness and efficiency of FCQIEA.