Yang Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yang Yang has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Wen Yao
Defense Innovation Institute, Chinese Academy of Military Science, Beijing, China
5 shared publications - Feifei Wang
Renmin University of China
2 shared publications - Handing Wang
Xidian University, Xi'an, China
2 shared publications - Wei Peng
Defense Innovation Institute, Chinese Academy of Military Science Intelligent Game and Decision Laboratory, Beijing, China
2 shared publications - Weien Zhou
Intelligent Game and Decision Laboratory, Beijing, China
2 shared publications Chao Ma
1 shared publication
Latest Publications
- 1 November 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (11)
Flow-sensing mechanism of harbor seal whisker to hydrodynamic perturbation
The harbor seal can accurately perceive minute hydrodynamic perturbations only by its beaded whiskers when vision or hearing is eliminated. The perception ability of whiskers is mainly affected by the whisker structure, the oscillating fluid (perturbation), and the fluid–structure interaction between them. However, the content has not been systematically studied. This paper first establishes a biomechanical theoretical model based on fluid–structure interaction between the oscillating fluid and the harbor seal whisker. Then, the effects of the whisker structure and the oscillating fluid on the hydrodynamic perturbation perception sensitivity are systematically investigated. For the whisker structure, reducing the whisker aspect ratio and density can improve the whisker sensitivity. Increasing the whisker height and the density can improve whisker tuning ability, that is, enhancing the sensing ability of whiskers to hydrodynamic perturbations. For the oscillating fluid, the fluid force, fluid density, and fluid oscillation direction can affect the force exerted on the whisker and then affect the sensing sensitivity of the whisker. We find that considering the additional drag and increasing fluid density can enhance the whisker tuning ability. When the oscillation direction of the fluid is perpendicular to the whisker's dorsal side, the bending stiffness of the dorsal side is less than the bending stiffness of the ventral side, resulting in a larger frequency response and sensing sensitivity and a stronger tuning ability. These findings provide new insights into the flow-sensing mechanisms of harbor seal whiskers and offer valuable inspiration for developing bioinspired flow-sensing technologies.
- 1 November 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (11)
Flow-sensing mechanism of harbor seal whisker to hydrodynamic perturbation
The harbor seal can accurately perceive minute hydrodynamic perturbations only by its beaded whiskers when vision or hearing is eliminated. The perception ability of whiskers is mainly affected by the whisker structure, the oscillating fluid (perturbation), and the fluid–structure interaction between them. However, the content has not been systematically studied. This paper first establishes a biomechanical theoretical model based on fluid–structure interaction between the oscillating fluid and the harbor seal whisker. Then, the effects of the whisker structure and the oscillating fluid on the hydrodynamic perturbation perception sensitivity are systematically investigated. For the whisker structure, reducing the whisker aspect ratio and density can improve the whisker sensitivity. Increasing the whisker height and the density can improve whisker tuning ability, that is, enhancing the sensing ability of whiskers to hydrodynamic perturbations. For the oscillating fluid, the fluid force, fluid density, and fluid oscillation direction can affect the force exerted on the whisker and then affect the sensing sensitivity of the whisker. We find that considering the additional drag and increasing fluid density can enhance the whisker tuning ability. When the oscillation direction of the fluid is perpendicular to the whisker's dorsal side, the bending stiffness of the dorsal side is less than the bending stiffness of the ventral side, resulting in a larger frequency response and sensing sensitivity and a stronger tuning ability. These findings provide new insights into the flow-sensing mechanisms of harbor seal whiskers and offer valuable inspiration for developing bioinspired flow-sensing technologies.
- 26 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 30 (3) , 898-910
Cross-Task Collaborative Optimization Based on Knowledge Transfer for Soft Robot Design
The automatic design of soft robots is an intertwined process of evolving morphology and learning control. As reinforcement learning is repeatedly used to learn the control policy for each candidate robot design, the design process becomes time-consuming. So far, the common design paradigm in robotics has been based on a single task. In fact, there is control similarity between different tasks. Learning a controller with combinatorial generalization capabilities across a variety of tasks can significantly reduce the computational cost of the design process. To this end, we propose a cross-task collaborative evolutionary algorithm that constructs a universal controller capable of solving a group of tasks simultaneously. Instead of “one robot, one controller, one task" paradigm, the proposed universal controller is to learn a control policy, which can generalize to unseen morphologies. After the controller learning on easy tasks, the universal controller can be further transferred to new hard tasks. Furthermore, the knowledge transfer is incorporated in the search strategy to enhance the performance of the universal controller. The experimental results on 13 test tasks demonstrate that the proposed algorithm outperforms the SOTA design algorithms on 8 of them. Compared to these algorithms, the proposed algorithm reduces the computational cost by 55% while achieving comparable performance, particularly for unseen hard tasks.
- 26 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 30 (3) , 898-910
Cross-Task Collaborative Optimization Based on Knowledge Transfer for Soft Robot Design
The automatic design of soft robots is an intertwined process of evolving morphology and learning control. As reinforcement learning is repeatedly used to learn the control policy for each candidate robot design, the design process becomes time-consuming. So far, the common design paradigm in robotics has been based on a single task. In fact, there is control similarity between different tasks. Learning a controller with combinatorial generalization capabilities across a variety of tasks can significantly reduce the computational cost of the design process. To this end, we propose a cross-task collaborative evolutionary algorithm that constructs a universal controller capable of solving a group of tasks simultaneously. Instead of “one robot, one controller, one task" paradigm, the proposed universal controller is to learn a control policy, which can generalize to unseen morphologies. After the controller learning on easy tasks, the universal controller can be further transferred to new hard tasks. Furthermore, the knowledge transfer is incorporated in the search strategy to enhance the performance of the universal controller. The experimental results on 13 test tasks demonstrate that the proposed algorithm outperforms the SOTA design algorithms on 8 of them. Compared to these algorithms, the proposed algorithm reduces the computational cost by 55% while achieving comparable performance, particularly for unseen hard tasks.
- 21 January 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 271
HeteroHTC: Enhancing Hierarchical Text Classification via Heterogeneity Encoding of Label Hierarchy
- 21 January 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 271
HeteroHTC: Enhancing Hierarchical Text Classification via Heterogeneity Encoding of Label Hierarchy
- Open Access
- 12 April 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Knowledge Discovery From Data
- Vol. 18 (6) , 1-32
Social Behavior Analysis in Exclusive Enterprise Social Networks by FastHAND
There is an emerging trend in the Chinese automobile industries that automakers are introducing exclusive enterprise social networks (EESNs) to expand sales and provide after-sale services. The traditional online social networks (OSNs) and enterprise social networks (ESNs), such as X (formerly known as Twitter) and Yammer, are ingeniously designed to facilitate unregulated communications among equal individuals. However, users in EESNs are naturally social stratified, consisting of both enterprise staffs and customers. In addition, the motivation to operate EESNs can be quite complicated, including providing customer services and facilitating communication among enterprise staffs. As a result, the social behaviors in EESNs can be quite different from those in OSNs and ESNs. In this work, we aim to analyze the social behaviors in EESNs. We consider the Chinese car manufacturer NIO as a typical example of EESNs and provide the following contributions. First, we formulate the social behavior analysis in EESNs as a link prediction problem in heterogeneous social networks. Second, to analyze this link prediction problem, we derive plentiful user features and build multiple meta-path graphs for EESNs. Third, we develop a novel Fast (H)eterogeneous graph (A)ttention (N)etwork algorithm for (D)irected graphs (FastHAND) to predict directed social links among users in EESNs. This algorithm introduces feature group attention at the node-level and uses an edge sampling algorithm over directed meta-path graphs to reduce the computation cost. By conducting various experiments on the NIO community data, we demonstrate the predictive power of our proposed FastHAND method. The experimental results also verify our intuitions about social affinity propagation in EESNs.
- 12 April 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Knowledge Discovery From Data
- Vol. 18 (6) , 1-32
Open AccessSocial Behavior Analysis in Exclusive Enterprise Social Networks by FastHAND
There is an emerging trend in the Chinese automobile industries that automakers are introducing exclusive enterprise social networks (EESNs) to expand sales and provide after-sale services. The traditional online social networks (OSNs) and enterprise social networks (ESNs), such as X (formerly known as Twitter) and Yammer, are ingeniously designed to facilitate unregulated communications among equal individuals. However, users in EESNs are naturally social stratified, consisting of both enterprise staffs and customers. In addition, the motivation to operate EESNs can be quite complicated, including providing customer services and facilitating communication among enterprise staffs. As a result, the social behaviors in EESNs can be quite different from those in OSNs and ESNs. In this work, we aim to analyze the social behaviors in EESNs. We consider the Chinese car manufacturer NIO as a typical example of EESNs and provide the following contributions. First, we formulate the social behavior analysis in EESNs as a link prediction problem in heterogeneous social networks. Second, to analyze this link prediction problem, we derive plentiful user features and build multiple meta-path graphs for EESNs. Third, we develop a novel Fast (H)eterogeneous graph (A)ttention (N)etwork algorithm for (D)irected graphs (FastHAND) to predict directed social links among users in EESNs. This algorithm introduces feature group attention at the node-level and uses an edge sampling algorithm over directed meta-path graphs to reduce the computation cost. By conducting various experiments on the NIO community data, we demonstrate the predictive power of our proposed FastHAND method. The experimental results also verify our intuitions about social affinity propagation in EESNs.
- 25 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 28 (6) , 1674-1688
Rapidly Evolving Soft Robots via Action Inheritance
The automatic design of soft robots characterizes as jointly optimizing structure and control. As reinforcement learning is gradually used to optimize control, the time-consuming controller training makes soft robots design an expensive optimization problem. Although surrogate-assisted evolutionary algorithms have made a remarkable achievement in dealing with expensive optimization problems, they typically suffer from challenges in constructing accurate surrogate models due to the complex mapping among structure, control, and task performance. Therefore, we propose an action inheritance-based evolutionary algorithm to accelerate the design process. Instead of training a controller, the proposed algorithm uses inherited actions to control a candidate design to complete a task and obtain its approximated performance. Inherited actions are near-optimal control policies that are partially or entirely inherited from optimized control actions of a real evaluated robot design. The action inheritance plays the role of surrogate models where its input is the structure and output is the near-optimal control actions. We also propose a random perturbation operation to estimate the error introduced by inherited control actions. The effectiveness of our proposed method is validated by evaluating it on a wide range of tasks, including locomotion and manipulation. Experimental results show that our algorithm is better than the other three state-of-the-art algorithms on most tasks when only a limited computational budget is available. Compared with the algorithm without surrogate models, our algorithm saves about half the computing cost.
- 25 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 28 (6) , 1674-1688
Rapidly Evolving Soft Robots via Action Inheritance
The automatic design of soft robots characterizes as jointly optimizing structure and control. As reinforcement learning is gradually used to optimize control, the time-consuming controller training makes soft robots design an expensive optimization problem. Although surrogate-assisted evolutionary algorithms have made a remarkable achievement in dealing with expensive optimization problems, they typically suffer from challenges in constructing accurate surrogate models due to the complex mapping among structure, control, and task performance. Therefore, we propose an action inheritance-based evolutionary algorithm to accelerate the design process. Instead of training a controller, the proposed algorithm uses inherited actions to control a candidate design to complete a task and obtain its approximated performance. Inherited actions are near-optimal control policies that are partially or entirely inherited from optimized control actions of a real evaluated robot design. The action inheritance plays the role of surrogate models where its input is the structure and output is the near-optimal control actions. We also propose a random perturbation operation to estimate the error introduced by inherited control actions. The effectiveness of our proposed method is validated by evaluating it on a wide range of tasks, including locomotion and manipulation. Experimental results show that our algorithm is better than the other three state-of-the-art algorithms on most tasks when only a limited computational budget is available. Compared with the algorithm without surrogate models, our algorithm saves about half the computing cost.
- Open Access
- 27 November 2022
- Published by Springer Nature in Scientific Reports
- Vol. 12 (1) , 1-10
Multiturn dialogue generation by modeling sentence-level and discourse-level contexts
Currently, multiturn dialogue models generate human-like responses based on pretrained language models given a dialogue history. However, most existing models simply concatenate dialogue histories, which makes it difficult to maintain a high degree of consistency throughout the generated text. We speculate that this is because the encoder ignores information about the hierarchical structure between sentences. In this paper, we propose a novel multiturn dialogue generation model that captures contextual information at the sentence level and at the discourse level during the encoding process. The context semantic information is dynamically modeled through a difference-aware module. A sentence order prediction training task is also designed to learn representation by reconstructing the order of disrupted sentences with a learning-to-rank algorithm. Experiments on the multiturn dialogue dataset, DailyDialog, demonstrate that our model substantially outperforms the baseline model in terms of both automatic and human evaluation metrics, generating more fluent and informative responses than the baseline model.
- 27 November 2022
- Published by Springer Nature in Scientific Reports
- Vol. 12 (1) , 1-10
Open AccessMultiturn dialogue generation by modeling sentence-level and discourse-level contexts
Currently, multiturn dialogue models generate human-like responses based on pretrained language models given a dialogue history. However, most existing models simply concatenate dialogue histories, which makes it difficult to maintain a high degree of consistency throughout the generated text. We speculate that this is because the encoder ignores information about the hierarchical structure between sentences. In this paper, we propose a novel multiturn dialogue generation model that captures contextual information at the sentence level and at the discourse level during the encoding process. The context semantic information is dynamically modeled through a difference-aware module. A sentence order prediction training task is also designed to learn representation by reconstructing the order of disrupted sentences with a learning-to-rank algorithm. Experiments on the multiturn dialogue dataset, DailyDialog, demonstrate that our model substantially outperforms the baseline model in terms of both automatic and human evaluation metrics, generating more fluent and informative responses than the baseline model.
- Open Access
VIRT: Improving Representation-based Models for Text Matching through Virtual Interaction
With the booming of pre-trained transformers, representation-based models based on Siamese transformer encoders have become mainstream techniques for efficient text matching. However, these models suffer from severe performance degradation due to the lack of interaction between the text pair, compared with interaction-based models. Prior arts attempt to address this through performing extra interaction for Siamese encoded representations, while the interaction during encoding is still ignored. To remedy this, we propose a \textit{Virtual} InteRacTion mechanism (VIRT) to transfer interactive knowledge from interaction-based models into Siamese encoders through attention map distillation. As a train-time-only component, VIRT could completely maintain the high efficiency of the Siamese structure and brings no extra computation cost during inference. To fully utilize the learned interactive knowledge, we further design a VIRT-adapted interaction strategy. Experimental results on multiple text matching datasets demonstrate that our method outperforms state-of-the-art representation-based models. What's more, VIRT can be easily integrated into existing representation-based methods to achieve further improvements.
Open AccessVIRT: Improving Representation-based Models for Text Matching through Virtual Interaction
With the booming of pre-trained transformers, representation-based models based on Siamese transformer encoders have become mainstream techniques for efficient text matching. However, these models suffer from severe performance degradation due to the lack of interaction between the text pair, compared with interaction-based models. Prior arts attempt to address this through performing extra interaction for Siamese encoded representations, while the interaction during encoding is still ignored. To remedy this, we propose a \textit{Virtual} InteRacTion mechanism (VIRT) to transfer interactive knowledge from interaction-based models into Siamese encoders through attention map distillation. As a train-time-only component, VIRT could completely maintain the high efficiency of the Siamese structure and brings no extra computation cost during inference. To fully utilize the learned interactive knowledge, we further design a VIRT-adapted interaction strategy. Experimental results on multiple text matching datasets demonstrate that our method outperforms state-of-the-art representation-based models. What's more, VIRT can be easily integrated into existing representation-based methods to achieve further improvements.
- Open Access
- 19 November 2021
- Published by Association for Computing Machinery (ACM)
- p. 53-62
Study and Application of Process Model of Product Function Extending Petri Net
In order to obtain the best original understanding of product system in the conceptual design stage, it is necessary to analyze the original understanding of product system and analyze the overall performance of the system in the conceptual design stage. Therefore, the design flow of the product functional expansion Petri net model is presented, various design schemes are analyzed and evaluated, the best design scheme is selected, and the effectiveness of the equipment for producing the Packer metal rubber expansion tube is verified.
- 19 November 2021
- Published by Association for Computing Machinery (ACM)
- p. 53-62
Open AccessStudy and Application of Process Model of Product Function Extending Petri Net
In order to obtain the best original understanding of product system in the conceptual design stage, it is necessary to analyze the original understanding of product system and analyze the overall performance of the system in the conceptual design stage. Therefore, the design flow of the product functional expansion Petri net model is presented, various design schemes are analyzed and evaluated, the best design scheme is selected, and the effectiveness of the equipment for producing the Packer metal rubber expansion tube is verified.
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 468-472
Study of a Community Partitioning Algorithm Based on Modularity Measurement
The current method of community division has the characteristics of single division method and inaccurate division result. Aiming at this problem, this algorithm proposes a community partitioning algorithm based on modularity measurement, which solves the problem of accuracy of community partitioning results. Firstly, the algorithm analyses the attributes of social networks, defines the structure of social network nodes, and describes two problems of node overlap structure and node hierarchy. Secondly, iteratively calculates the influence of two structures in social networks, split the nodes that are mixed together. Thirdly, measure the similarity and community module as two evaluation indicators, and divide the social network into different communities. Finally, using the three real data sets of Karate Club Network, Dolphin Social Network and The Pol Books Dataset, this algorithm is designed to compare with other algorithms. The experimental results show that the algorithm modularity of the algorithm has higher accuracy of community division.
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 468-472
Study of a Community Partitioning Algorithm Based on Modularity Measurement
The current method of community division has the characteristics of single division method and inaccurate division result. Aiming at this problem, this algorithm proposes a community partitioning algorithm based on modularity measurement, which solves the problem of accuracy of community partitioning results. Firstly, the algorithm analyses the attributes of social networks, defines the structure of social network nodes, and describes two problems of node overlap structure and node hierarchy. Secondly, iteratively calculates the influence of two structures in social networks, split the nodes that are mixed together. Thirdly, measure the similarity and community module as two evaluation indicators, and divide the social network into different communities. Finally, using the three real data sets of Karate Club Network, Dolphin Social Network and The Pol Books Dataset, this algorithm is designed to compare with other algorithms. The experimental results show that the algorithm modularity of the algorithm has higher accuracy of community division.