Graph Neural Network With Self-Attention And Multi-Task Learning For Credit Default Risk Prediction,
2022
University of Technology Sydney
Graph Neural Network With Self-Attention And Multi-Task Learning For Credit Default Risk Prediction, Zihao Li, Xianzhi Wang, Lina Yao, Yakun Chen, Guandong Xu, Ee-Peng Lim
Research Collection School Of Computing and Information Systems
We propose a graph neural network with self-attention and multi-task learning (SaM-GNN) to leverage the advantages of deep learning for credit default risk prediction. Our approach incorporates two parallel tasks based on shared intermediate vectors for input vector reconstruction and credit default risk prediction, respectively. To better leverage supervised data, we use self-attention layers for feature representation of categorical and numeric data; we further link raw data into a graph and use a graph convolution module to aggregate similar information and cope with missing values during constructing intermediate vectors. Our method does not heavily rely on feature engineering work and …
Daot: Domain-Agnostically Aligned Optimal Transport For Domain-Adaptive Crowd Counting,
2022
Singapore Management University
Daot: Domain-Agnostically Aligned Optimal Transport For Domain-Adaptive Crowd Counting, Huilin Zhu, Jingling Yuan, Xian Zhong, Zhengwei Yang, Zheng Wang, Shengfeng He
Research Collection School Of Computing and Information Systems
Domain adaptation is commonly employed in crowd counting to bridge the domain gaps between different datasets. However, existing domain adaptation methods tend to focus on inter-dataset differences while overlooking the intra-differences within the same dataset, leading to additional learning ambiguities. These domain-agnostic factors,e.g., density, surveillance perspective, and scale, can cause significant in-domain variations, and the misalignment of these factors across domains can lead to a drop in performance in cross-domain crowd counting. To address this issue, we propose a Domain-agnostically Aligned Optimal Transport (DAOT) strategy that aligns domain-agnostic factors between domains. The DAOT consists of three steps. First, individual-level differences …
Efficient Navigation For Constrained Shortest Path With Adaptive Expansion Control,
2022
Singapore Management University
Efficient Navigation For Constrained Shortest Path With Adaptive Expansion Control, Wenwen Xia, Yuchen Li, Wentian Guo, Shenghong Li
Research Collection School Of Computing and Information Systems
In many route planning applications, finding constrained shortest paths (CSP) is an important and fundamental problem. CSP aims to find the shortest path between two nodes on a graph while satisfying a path constraint. Solving CSPs requires a large search space and is prohibitively slow on large graphs, even with the state-of-the-art parallel solution on GPUs. The reason lies in the lack of effective navigational information and pruning strategies in the search procedure. In this paper, we propose SPEC, a Shortest Path Enhanced approach for solving the exact CSP problem. Our design rationales of SPEC rely on the observation that …
Development Of A Smartphone Application As An Asset To Pavement Management Engineers, Smartp3m,
2022
University of Texas at Tyler
Development Of A Smartphone Application As An Asset To Pavement Management Engineers, Smartp3m, Damien Stephens
Electrical Engineering Theses
An application specific multi-platform smartphone application can utilize on-board accelerometer, gyroscope, and GPS sensors, along with software derived signals from the same sensors, to sample vibrational and geolocation datasets to capture pavement distresses such as potholes when mounted in a standardized configuration in a vehicle. Several observations were made with regard to the signals obtained from the accelerometer, gyroscope, and GPS sensors, and it was determined that the raw sensor outputs are capable of sampling statistically significant datasets which can be used to distinguish pavement distress from normal driving conditions. Furthermore, an approximate sensor noise margin is established, and a …
Review On Ecological Construction Of Domestic High-Performance Parallel Application Software In Post Moore Era,
2022
1.Science and Technology on Parallel and Distribute Processing Laboratory, National University of Defense Technology, Changsha 410073, China;
Review On Ecological Construction Of Domestic High-Performance Parallel Application Software In Post Moore Era, Chunye Gong, Jie Liu, Weimin Bao, Dongmei Pan, Xinbiao Gan, Shengguo Li, Xuguang Chen, Tiaojie Xiao, Bo Yang, Ruibo Wang
Journal of System Simulation
Abstract: Domestic high performance computing (HPC) system is world-leading and the system chip architectures are in varied forms. The system operation relied on National Supercomputing Center has a good development trend. Several technical key points of domestic high-performance parallel application software are word-leading and the application supporting environment is developing fast. But industrial software and team building are facing huge challenges. In post Moore era, based on the progress of human civilization, it is necessary to promote the ecological development of parallel application software, and from the viewpoint of software products the industrial software must be aim foreign commercial software …
Research On Improved Feature Pyramid Algorithm Integrating Border Supervision Strategy,
2022
School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
Research On Improved Feature Pyramid Algorithm Integrating Border Supervision Strategy, Hong Sun, Yuelan Ling, Yuxiang Zhang
Journal of System Simulation
Abstract: Aiming at the inaccurate boundary division in semantic segmentation and the existence of multi-scale targets, an improved feature pyramid algorithm fused with boundary supervision strategies is proposed. By fusing the boundary supervision strategy and the improved feature pyramid algorithm, the problems of inaccurate boundary division and the existence of multi-scale targets are sloved respectively, and an attention mechanism is added in the upsampling process to further improve the segmentation effect. The experimental results show that the algorithm can reach 58.69% and 78.59% MIOU (mean intersection over union) indicators on the Camvid and PASCAL VOC2012 data sets respectively, and has …
Real-Time Scheduling Method For Railway Passenger Station Operations Based On Digital Twin,
2022
1.School of Transportation and Logistics, Southwest Jiaotong University, Chengdu 611756, China;2.National United Engineering Laboratory of Integrated and Intelligent Transportation, Southwest Jiaotong University, Chengdu 611756, China;3.Comprehensive Transportation Key Laboratory of Sichuan Province, Southwest Jiaotong University, Chengdu 611756, China;
Real-Time Scheduling Method For Railway Passenger Station Operations Based On Digital Twin, Bisheng He, Peng Chen, Hongxiang Zhang, Gongyuan Lu, Chunhui Zhang
Journal of System Simulation
Abstract: To improve the operation scheduling of railway passenger stations and reduce the train delays, digital twin technology is used to establish a railway passenger station operation model. Through real-time data acquisition, based on the operation time prediction of random forest method, operation simulation and decision-making, a real-time scheduling method of railway passenger station operations based on digital twin is proposed, and applied in a real railway passenger station. The experimental results show that the method can effectively forecast and simulate the actual operation. Three kinds of digital twin scheduling rules have been used, which can reduce the delay time …
A Quantum Krill Herd Fusion Algorithm And Its Application,
2022
1.School of Building Services Science and Engineering, Xi'an University of Architecture and Technology, Xi'an 710055, China;2.Anhui Key Laboratory of Intelligent Building and Building Energy Conservation, Anhui Jianzhu University, Hefei 230022, China;
A Quantum Krill Herd Fusion Algorithm And Its Application, Zengxi Feng, Jintong Zhao, Shiyan Li, Yalong Yang, Haiyue Chen, Cong Zhang
Journal of System Simulation
Abstract: Aiming at the defects of krill herd algorithm and quantum evolutionary algorithm, a quantum krill herd fusion algorithm (QKH) is proposed. The algorithm uses double-chain real numbers to encode the krill position, which can speed up the convergence speed, and avoids the randomness and complexity of quantum observations. The dynamically adjusted quantum krill herd rotation phase update strategy improves the convergence accuracy, and the efficiency of determining the quantum rotation phase. The introduction of an improved quantum full interference crossover strategy can prevent the fusion algorithm from falling into a local optimum, and can improve the optimization efficienal. The …
Control Of Quadruped Robot Based On Impedance And Virtual Model,
2022
1.School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
Control Of Quadruped Robot Based On Impedance And Virtual Model, Chikun Gong, Xunwei Wu, Lipeng Yuan
Journal of System Simulation
Abstract: In order to improve the motion stability of quadruped robot, a control method based on impedance and virtual model is proposed. The force-based impedance control method is used to control the leg swing phase to realize the more accurate trajectory tracking and leg compliance control. The virtual model control method is used to control the support phase to realize the attitude control of the robot body and the stable walking of the quadruped robot. Combined with the lateral stride strategy and the yaw angle control strategy based on virtual model, a robot anti-lateral impact control method is proposed, which …
Adaptive Crowd Evacuation Simulation Model Based On Bounded Rationality Constraints,
2022
1.College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China;
Adaptive Crowd Evacuation Simulation Model Based On Bounded Rationality Constraints, Liqiang Zhao, Mengqian Guo, Shuixiong Tang, Jinjin Tang
Journal of System Simulation
Abstract: To effectively improve the accuracy of evacuation simulation of a crowded environment, an adaptive crowd evacuation simulation model based on social force model and bounded rationality constraints is proposed. The desired direction and desired speed of the self-driving force that affects the pedestrian movement in the traditional social force model is improved. The adaptive calculation is used in the optimization of direction and speed of pedestrian in an obstacle avoidance situation. The rational route decision mechanism is proposed to describe the route selection behavior of pedestrians in a congested state more accurately. The results show that the proposed model …
Research On Modeling And Simulation Of Optimization Deployment For Cooperative Localization By Multiple Detection Sensors In Complex Environment,
2022
Northwest Institute of Nuclear Technology, Xi'an 710024, China;
Research On Modeling And Simulation Of Optimization Deployment For Cooperative Localization By Multiple Detection Sensors In Complex Environment, Gongguo Xu, Libing Cai, Peibing Du, Yu Liu
Journal of System Simulation
Abstract: Aiming at the difficulty of accurate cooperative localization by multi-sensor network in complex environment, an optimization deployment method is proposed. The GDOP evaluation index of target positioning accuracy is constructed based on PCRLB. The influence of undulating terrain, clutter jamming and illumination on sensor detection and positioning ability in complex environment is analyzed. An optimization deployment model of multi-sensor cooperative location is built and the intelligent optimization algorithm is used to quickly solve the model. Simulation results show that the proposed method can effectively improve the cooperative localization ability of multi-sensor network and can guide the multi-sensor cooperative …
Tactical Maneuver Strategy Learning From Land Wargame Replay Based On Convolutional Neural Network,
2022
1.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China;2.Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China;
Tactical Maneuver Strategy Learning From Land Wargame Replay Based On Convolutional Neural Network, Jiale Xu, Haidong Zhang, Donghai Zhao, Wancheng Ni
Journal of System Simulation
Abstract: Aiming at collecting the high valuable knowledge of action decisions in "man-in-the-loop" wargame's replay data, a method of using convolutional neural network to learn the tactical maneuver strategy model from the replay data of wargame is proposed. In this method, the tactical maneuver strategy is modeled as a classification problem of making a good choice from the target candidate locations under the influence of current situation. The key factors affecting commander's decision-making are summarized, and the basic situation features are defined, which are composed of seven attributes such as "maneuverability range and observation range". The feature dataset with positive …
Weighted Local Complexity Invariance For Time Series Classification,
2022
1.Guangzhou Institute of Technology, Xidian University, Guangzhou 510555, China;
Weighted Local Complexity Invariance For Time Series Classification, Yitong Li, Xiaotao Liu, Jing Liu, Kai Wu
Journal of System Simulation
Abstract: Aiming at the misclassification of existing algorithms for long or unevenly distributed time series, the local complexity information is extracted and weighted local complexity-invariant distance (WLCID) is proposed, which includes the local complexity representation model and the weighted global complexity integration model. Sliding window is used to split up time series, and combined with the complexity-invariant distance, the local complexity information can be extracted. As to the class representation model, the integration weights are quantified with the normalized cumulative between-class distance, with the perspective that the subsequence contributes more greatly with larger between-class distance. Compared with other …
A K-Modes Clustering Method Based On Maximal Information Coefficient Data Preprocessing,
2022
1.Hangzhou Dianzi University, Hangzhou 310018, China;
A K-Modes Clustering Method Based On Maximal Information Coefficient Data Preprocessing, Mingmei Li, Chenglin Wen, Shaolin Hu
Journal of System Simulation
Abstract: The existing k-modes clustering method ignores the weak correlation of variable attributes, which often results in poor clustering performance in practical applications. A new k-modes clustering method that includes the weak correlation of attributes is proposed. Maximum information coefficient (MIC) is introduced to measure the correlation of variable attributes in the data set. The obtained MIC value is merged with the original distance to establish a new measurement method containing weak attribute correlation information to enhance the completeness of related information of variable attributes, and a more refined k-modes clustering method is established. Three different data sets are used …
Research On Satellite Navigation Confrontation Deduction Model For Wargame System,
2022
1.College of Joint Operations, National Defense University, Beijing 100091, China;2.Graduate School, National Defense University, Beijing 100091, China;
Research On Satellite Navigation Confrontation Deduction Model For Wargame System, Runqing Yang, Xi Wu
Journal of System Simulation
Abstract: Following the increasing reliance on space-based time-space metric of information warfare, satellite navigation confrontation becomes the major combat operation of competing for superiority in space-based position, navigation, and timing. Aiming at building a satellite navigation operation deduction model, on the basis of analyzing the relevant research, the navigation information environment computing model and the navigation deduction operation simulation model are designed, and the way to realize the visualization of navigation information environment of the wargame system and a substitution of high-resolution simulation computing model low-resolution probability effect model are explored, which support the pre-war planning and wartime evaluation …
Wind Power Primary Frequency Regulation Simulation Based On Wind Speed Prediction Model,
2022
School of Economics and Management, North China Electric Power University, Beijing 102206, China;
Wind Power Primary Frequency Regulation Simulation Based On Wind Speed Prediction Model, Zhenyu Zhao, Xu Ma, Geriletu Bao
Journal of System Simulation
Abstract: To furtherly improve the accuracy of wind speed prediction, considering the coupled comprehensive features of data internal structure external influencing factors, and the characteristics of wind turbines power, a BP-ARIMA combined prediction model is constructed and the wind speed of wind farms is simulated. Compared with the actual data, the high prediction accuracy of the model is verified. Bases on the curve of typical power characteristic of wind turbine, the optimal configuration of energy storage device capacity and the stability of wind turbine are considered, and the MATLAB-SIMULINK platform is applied to simulate the primary frequency regulation of …
Multi-Sensory Fusion Method For Power Transformer Virtual Assembly,
2022
North China Electric Power University, Baoding 071000, China;
Multi-Sensory Fusion Method For Power Transformer Virtual Assembly, Xuqiang Shao, Haowei Zhang, Xiaohua Feng
Journal of System Simulation
Abstract: Virtual assembly technology is to truly restore the equipments and physical scenarios. In the real world, people can interact with the physical world through visual, auditory, tactile and other sense organ. Aiming at the existing virtual assembly system being limited the single sense human-computer interaction mode, so a multi-sense fusion information interaction method is proposed to improve the sense of immersion and operability. An improved AABB Octree bounding box collision detection algorithm of large size diffidence component to be assembled is proposed, which can greatly reduce the amount of calculation and improve the calculation accuracy. The experimental result verifies …
Reinforcement-Learning-Based Adaptive Tracking Control For A Space Continuum Robot Based On Reinforcement Learning,
2022
1.Dalian University of Technology, Dalian 116024, China;
Reinforcement-Learning-Based Adaptive Tracking Control For A Space Continuum Robot Based On Reinforcement Learning, Da Jiang, Zhiqin Cai, Zhongzhen Liu, Haijun Peng, Zhigang Wu
Journal of System Simulation
Abstract: Aiming at the tracking control for three-arm space continuum robot in space active debris removal manipulation, an adaptive sliding mode control algorithm based on deep reinforcement learning is proposed. Through BP network, a data-driven dynamic model is developed as the predictive model to guide the reinforcement learning to adjust the sliding mode controller's parameters online, and finally realize a real-time tracking control. Simulation results show that the proposed data-driven predictive model can accurately predict the robot's dynamic characteristics with the relative error within ±1% to random trajectories. Compared with the fixed-parameter sliding mode controller, the proposed adaptive controller …
Simulation Research On Aerodynamic Characteristics Of A Miniature Munition,
2022
Beijing Key Laboratory of High Dynamic Navigation Technology, Beijing Information Science & Technology University, Beijing 100192, China;
Simulation Research On Aerodynamic Characteristics Of A Miniature Munition, Haisen Wang, Junfang Fan
Journal of System Simulation
Abstract: Aiming at the existing constraints of layout design of miniature munition, a new layout optimization scheme is proposed. Based on the normal layout structure, a three-dimensional aerodynamic model of munition is established, and the aerodynamic hydrodynamics simulation of miniature munition is carried out, and the influence of wing size on static stability of munition is studied. To verify the feasibility of the design scheme, the aerodynamic simulation calculation is carried out, and the lift, drag and pitching moment parameters of miniature munition are obtained through simulated aerodynamic calculation under different flight conditions, and the lift-drag ratio …
Fatigue Detection Method Based On Facial Features And Head Posture,
2022
1.School of Electrical and Automation Engineering, East China Jiaotong University, Nanchang 330013, China;2.Key Laboratory of Advanced Control and Optimization of Jiangxi Province, Nanchang 330013, China;
Fatigue Detection Method Based On Facial Features And Head Posture, Rongxiu Lu, Bihao Zhang, Zhenlong Mo
Journal of System Simulation
Abstract: Aiming at the of the single fatigue characteristics, low robustness and inability to customize fatigue thresholds for different drivers of fatigue detection methods, a method based on facial features and head posture is proposed. In face detection and face key point positioning HOG feature operator and regression tree algorithm are used. In head posture estimation, head posture Euler angle is estimated by combining the face key points with the coordinate system transformation. In fatigue feature extraction, a deep residual neural network model is established to extract the eye fatigue features, which the eye, mouth aspect ratio and head posture …
