Stability-Based Generalization Analysis For Mixtures Of Pointwise And Pairwise Learning,
2023
Huazhong Agricultural University
Stability-Based Generalization Analysis For Mixtures Of Pointwise And Pairwise Learning, Jiahuan Wang, Jun Chen, Hong Chen, Bin Gu, Weifu Li, Xin Tang
Machine Learning Faculty Publications
Recently, some mixture algorithms of pointwise and pairwise learning (PPL) have been formulated by employing the hybrid error metric of “pointwise loss + pairwise loss” and have shown empirical effectiveness on feature selection, ranking and recommendation tasks. However, to the best of our knowledge, the learning theory foundation of PPL has not been touched in the existing works. In this paper, we try to fill this theoretical gap by investigating the generalization properties of PPL. After extending the definitions of algorithmic stability to the PPL setting, we establish the high-probability generalization bounds for uniformly stable PPL algorithms. Moreover, explicit convergence …
Class-Independent Regularization For Learning With Noisy Labels,
2023
Beijing Jiaotong University
Class-Independent Regularization For Learning With Noisy Labels, Rumeng Yi, Dayan Guan, Yaping Huang, Shijian Lu
Computer Vision Faculty Publications
Training deep neural networks (DNNs) with noisy labels often leads to poorly generalized models as DNNs tend to memorize the noisy labels in training. Various strategies have been developed for improving sample selection precision and mitigating the noisy label memorization issue. However, most existing works adopt a class-dependent softmax classifier that is vulnerable to noisy labels by entangling the classification of multi-class features. This paper presents a class-independent regularization (CIR) method that can effectively alleviate the negative impact of noisy labels in DNN training. CIR regularizes the class-dependent softmax classifier by introducing multi-binary classifiers each of which takes care of …
Deep Learning Enhancement And Privacy-Preserving Deep Learning: A Data-Centric Approach,
2023
University of South Florida
Deep Learning Enhancement And Privacy-Preserving Deep Learning: A Data-Centric Approach, Hung S. Nguyen
USF Tampa Graduate Theses and Dissertations
Deep Learning and its applications have become attractive to a lot of research recentlybecause of its capability to capture important information from large amounts of data. While most of the work focuses on finding the best model parameters, improving machine learning performance from data perspective still needs more attention. In this work, we propose techniques to enhance the robustness of deep learning classification by tackling data issue. Specifically, our data processing proposals aim to alleviate the impacts of class-imbalanced data and non- IID data in deep learning classification and federated learning scenarios. In addition, data pre-processing strategies such that dimensionality …
Probing As A Technique To Understand Abstract Spaces,
2023
University of Minnesota Morris
Probing As A Technique To Understand Abstract Spaces, Ashlen A. Plasek
Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal
Machine learning models, while very powerful, have their operation obfuscated behind millions of parameters. This obfuscation can make deriving a human meaningful process from a machine learning model very difficult. However, while the intermediate states of a machine learning model are similarly obfuscated, using probing, we can start to explore looking at possible structure in those intermediate states. Large language models are a prime example of this obfuscation, and probing can begin to allow novel experimentation to be performed.
Lidar Segmentation-Based Adversarial Attacks On Autonomous Vehicles,
2023
University of Minnesota Morris
Lidar Segmentation-Based Adversarial Attacks On Autonomous Vehicles, Blake Johnson
Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal
Autonomous vehicles utilizing LiDAR-based 3D perception systems are susceptible to adversarial attacks. This paper focuses on a specific attack scenario that relies on the creation of adversarial point clusters with the intention of fooling the segmentation model utilized by LiDAR into misclassifying point cloud data. This can be translated into the real world with the placement of objects (such as road signs or cardboard) at these adversarial point cluster locations. These locations are generated through an optimization algorithm performed on said adversarial point clusters that are introduced by the attacker.
The Use Of Artificial Intelligence In Higher Education: A Study On Faculty Perspectives In Universities In Egypt,
2023
American University in Cairo
The Use Of Artificial Intelligence In Higher Education: A Study On Faculty Perspectives In Universities In Egypt, Farah S. Sharawy
Theses and Dissertations
Artificial Intelligence (AI) is an emerging technology that is transforming various aspects of society, including higher education. This paper examines faculty perspectives from five different institutions; The American University in Cairo (AUC), The German University in Cairo (GUC), The Arab Academy for Science and Technology (AAST), Ain Shams University, and Cairo University, on the use of AI in higher education in teaching and learning in Egypt, with all its challenges and resources available to support it, and how it can be used to achieve equity and accessibility. This research was conducted through a qualitative study using semi-structured one- on-one interviews …
Point Cloud Surface Matching Method Based On Precise Matching Of Critical Point,
2023
School of Computer Science and Engineering, Xi'an University of Technology, Xi'an 710048, China; Shaanxi Provincial Key Laboratory of Network Computing and Security Technology, Xi'an 710048, China
Point Cloud Surface Matching Method Based On Precise Matching Of Critical Point, Xiaojuan Ning, Chunxu Li, Jiahao Wang, Jing Tang, Yinghui Wang, Haiyan Jin
Journal of System Simulation
To solve the low matching efficiency and insufficient accuracy of feature-based point cloud surface matching method during critical point matching, a point cloud surface matching method based on the pairing exaction of critical points is proposed.An improved 3D scale-invariant feature transform(3D-SIFT) algorithm based on curvature information is presented to extract the critical points. Fast point feature histograms(FPFH) feature, the angle between the vector from the center to critical points and the principal direction of the model are taken as the constraints to obtain the exact critical point matching point pair set. The initial matching of the model surface is …
Semantic Segmentation Model Based On Adaptive Fusion And Attention Refinement,
2023
School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
Semantic Segmentation Model Based On Adaptive Fusion And Attention Refinement, Yun Wei, Qi Luo, Yingzhi Zhao
Journal of System Simulation
Aiming at the insufficient use of context information and loss of detail information of the existing semantic segmentation, a model based on adaptive fusion and attention refinement is proposed.The model introduces an adaptive fusion module in the process of coding, and solves the insufficient use of context information by fusing each feature map according to the corresponding weight. An attention thinning module is designed in the process of decoding, so that the low-order features and high-order features can guide and optimize each other to solve the loss of detail information.The experimental results show that the average intersection union …
Fault Indicator Configuration Optimization Based On Cooperative Game Particle Swarm Algorithm,
2023
College of Computer Science and Information Engineering, Harbin Normal University, Harbin 150025, China
Fault Indicator Configuration Optimization Based On Cooperative Game Particle Swarm Algorithm, Xu Wang, Weidong Ji, Guohui Zhou
Journal of System Simulation
In order to balance the reliability and economy of distribution network fault indicator, a multispace cooperative game particle swarm optimization algorithm is proposed. Based on the idea of population space grouping, the population activity space is adaptively divided, the particle game evolution in subspace is achieved, and the game calculation of particle fusion cosine similar reverse strategy is carried out, which well balances the convergence and diversity of the algorithm. The simulation results show that the adaptive multi-space division of population is conducive to jumping out of the local extreme value, and the game calculation integrating cosine similar reverse is …
Ar-Assisted Sign Language Letter Recognition Method Based On Improved Mobilenet Network,
2023
School of Computer Science & Technology, Southwest University of Science and Technology, Mianyang 621010, China
Ar-Assisted Sign Language Letter Recognition Method Based On Improved Mobilenet Network, Chunhong Liu, Song Wang, Fupan Wang, Wensheng Tang, Yunqiang Pei, Dongsheng Tian, Yadong Wu
Journal of System Simulation
An AR-assisted sign language letter recognition algorithm MS-MobileNet is proposed for the problems of sign language gestures needing to be standardized and low recognition rate. A multi-scale convolution module is designed to extract the low-level features and enhance the feature extraction ability. ELU activation function is used to retain the negative feature information, which combined with a lightweight MobileNet model for the web to improve the recognition accuracy and real-time performance for mobile AR applications. Test results show that compared with the original model, the recognition accuracy of MS-MobileNet on the datasets ASL-M, NUS-II and Creative Senz3D is improved by …
Summary Of Simulation Technology And Its Application In Training Field,
2023
Naval Research Academy PLA Beijing 100161 China
Summary Of Simulation Technology And Its Application In Training Field, Zhiming Qiu, Heng Li, Yufang Zhou, Duzheng Qing
Journal of System Simulation
High-tech weapons operations and battlefield environment become more and more increasingly sophisticated. It is an effective way to resolve the conflicts between huge equipment training cost and limited resources and improve training quality by building networked systematic and intelligent simulation training equipment and system and conducting systematic countermeasure simulation training based on present simulation technologies and training resources. The current situation and trend of the development of system architecture of simulation simulation modeling enemy simulation information system simulation etc are evaluated and summarized. It initially analyses the application and efficacy on the US Navy's normalized training environment typical systems such …
Simulation Research On Cooperative Control For Aircraft Ground Deicing Operation,
2023
College of Aeronautical Engineering, Civil Aviation University of China, Tianjin 300300, China
Simulation Research On Cooperative Control For Aircraft Ground Deicing Operation, Liwen Wang, Biao Li, Zhiwei Xing, Guan Lian
Journal of System Simulation
Aiming at the weak cooperation and low efficiency in aircraft ground deicing operation,a cooperative control method for deicing operations is designed based on the consistency of information states.A two-stage deicing Agent model is proposed and a multi-objective collaborative optimization model with the best deicing effect and the shortest deicing time is established. Lagrangian relaxation algorithm is applied to construct a feasible solution for the optimal operating parameters. Combined with the deicing process and mode, the cooperative graph of deicing operation is studied based on the operating topology, and the cooperative control method of ground deicing is constructed …
Research On Energy Coordination Control Strategy In Dc Microgrid,
2023
College of Physics and Electronic Information, Anhui Normal University, Wuhu 241000, China;Anhui Provincial Engineering Laboratory on Information Fusion and Control of Intelligent Robot, Wuhu 241000, China
Research On Energy Coordination Control Strategy In Dc Microgrid, Zibao Lu, Hao Ding, Fangyun Sun, Ziqiong Ding, Li Gong, Rui Zheng
Journal of System Simulation
Aiming of the bus voltage stability and energy flow balance in DC microgrids,a switching control strategy based on the energy balance relationship of DC microgrid is proposed. The bus voltage balance of DC microgrid is transformed into energy balance, and DC microgrid is modeled as a linear switching system with five modes. A controller is designed for each mode to stabilize the bus voltage. To achieve the seamless and smooth switching between modes, on the basis of maintaining the stability of any switching, the corresponding mode switching rules are given based on the energy flow characteristics of microgrid.Theoretical …
An Approach To Solving The Incoming Target Based On Uncertain Time Series,
2023
Navy Submarine College, Qingdao 266041, China; College of System and Engineering, National University of Defence and Technology, Changsha 410073, China
An Approach To Solving The Incoming Target Based On Uncertain Time Series, Jing Yang, Minghua Lu, Xingchen Hu, Jinping Wu
Journal of System Simulation
The traditional solution method for the incoming attacking target in water lacks the time series characteristics mining on multi-dimensional and uncertain observation data. Aiming at the time series prediction with high complexity and missing data, a method based on adaptive window interpolation and deep variable weight long-term and short-term memory network model for missing time series observation data with multiple sampling frequencies is proposed, which is compared and verified on the simulation data and public test data set.Aiming at the random missing problem caused by the inconsistency of sampling frequency of multi-source observation information, an adaptive window imputation method …
Particle Swarm Optimization For New Energy Truck Scheduling In Network Environment,
2023
School of Physics and Electronic Information, Anhui Normal University, Wuhu 241002, China; Anhui Provincial Engineering Laboratory on Information Fusion and Control of Intelligent Robot (Anhui Normal University), Wuhu 241002, China
Particle Swarm Optimization For New Energy Truck Scheduling In Network Environment, Chuanchao Zhao, Rui Zheng, Li Gong, Xiaolu Ma
Journal of System Simulation
In V2X intelligent network environment, the dispatching system of new energy trucks needs real-time dynamic information.The system under traditional particle swarm scheduling method is prone to fall into local optimum and low solution efficienty.An improved particle swarm scheduling method for new energy trucks is proposed on the basis of multi-objective optiminaztion research. The inertia weight update method is improvedso that the inertia weight decreases non-linearly, andthe risk of the system falling into local optimum is reduced.A priori path encoding method is designed and optimized,the solution efficiency of the algorithm is improved, …
Research On Application Of Monarch Butterfly Optimization Particle Filter In Slam,
2023
College of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, China
Research On Application Of Monarch Butterfly Optimization Particle Filter In Slam, Zhiqiang Chen, Menglong Cao, Wenbin Zhao
Journal of System Simulation
In traditional particle filter resampling, weight degradation and loss of particle diversity are prone to occur, which leads to the decrease in filtering accuracy and result in inaccurate robot positioning and inaccurate mapping. An optimized particle filter algorithm based on the improved monarch butterfly algorithm is proposed.The algorithm replaces the particle individual with the monarch butterfly individual, and integrates the migration operator and adjustment operator in the monarch butterfly algorithm into the particle filter algorithm. The adaptive genetic parameters are introduced to the iterative update process of the monarch butterfly, and the linear combination optimization resampling method is used …
A Uav Target Tracking And Control Algorithm Based On Siamrpn,
2023
Department of Automation, North China Electric Power University, Baoding 071003, China
A Uav Target Tracking And Control Algorithm Based On Siamrpn, Songming Jiao, Hui Ding, Yufei Zhong, Xin Yao, Jiahao Jiahao Zheng
Journal of System Simulation
Aiming at the requirement of autonomously tracking land moving targets of rotary-wing UAVs, an autonomous and stable UAV tracking and control system that can adapt to the common interference environments such as scale changes, occlusions, and attitude changes is constructed.The system extracts the imaging position of the target in airborne camera through the twin network based on deep learning, and obtains the relative pose of the target. The image processing algorithm is designed to process the icons in the tracking frame, and the yaw angle of UAV relative to the tracking target is obtained, Kalman filter is introduced to …
Application Of Digital Twin In Digital Transformation Of Thermal Power Units,
2023
Hebei Technology Innovation Center of Simulation & Optimized Control for Power Generation North China Electric Power University Baoding 071000 China;
School of Control and Computer Engineering North China Electric Power University Beijing 102206 China
Application Of Digital Twin In Digital Transformation Of Thermal Power Units, Ze Dong, Wei Jiang, Xiaoyan Wang, Lei Liu
Journal of System Simulation
A new state estimation algorithm is proposed to improve the accuracy to obtain the optimal state estimation of distribution network against FDI attack. In the case of phasor measurement units being attacked and the measurement results being alteredthe optimal Kalman estimate can be decomposed into a weighted sum of local state estimates. Focusing on the insecurity of the weighted sum methoda convex optimization based on local estimation is proposed to replace the method and combine the local estimation into a secure state estimation. The simulation results show that the proposed estimator is consistent with the Kalman …
Dynamic Simulation Of Urban Agglomeration Passenger Transport Network Vulnerability Based On Multi-Agent,
2023
School of Transportation, Inner Mongolia University, Hohhot 010070, China
Dynamic Simulation Of Urban Agglomeration Passenger Transport Network Vulnerability Based On Multi-Agent, Chengbing Li, Yunfei Li, Peng Wu
Journal of System Simulation
Research on vulnerability of comprehensive passenger transportation network in urban agglomerations helps to ensure the transportation efficiency of intercity travel.A comprehensive passenger transport network model of urban agglomeration is built based on multi-layer complex network theory. Urban transportation transfer factors are considered, actual passenger flow is used to calibrate the station load and capacity and a dynamic model of network cascading failure is constructed. Multi-agents are used to simulate the actual passenger flow, Dijkstra algorithm is used to find the shortest path, and two time-dimensional vulnerability evaluation indicators are proposed.MATLAB is used to carry out the dynamic simulation …
Secure State Estimation Of Distribution Network Based On Kalman Filter Decomposition,
2023
School of Electrical Engineering, Xi'an University of Technology, Xi'an 710048, China
Secure State Estimation Of Distribution Network Based On Kalman Filter Decomposition, Xinghua Liu, Siwen Dong, Jiaqiang Tian
Journal of System Simulation
A new state estimation algorithm is proposed to improve the accuracy to obtain the optimal state estimation of distribution network against FDI attack. In the case of phasor measurement units being attacked and the measurement results being altered,the optimal Kalman estimate can be decomposed into a weighted sum of local state estimates. Focusing on the insecurity of the weighted sum method,a convex optimization based on local estimation is proposed to replace the method and combine the local estimation into a secure state estimation. The simulation results show that the proposed estimator is consistent with the Kalman …
