检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Liu, Bangzhen Zheng, Chenxi Xu, Xuemiao Xu, Cheng Zhang, Huaidong He, Shengfeng South China University of Technology Guangzhou China Guangdong Engineering Center for Large Model and GenAI Technology The State Key Laboratory of Subtropical Building and Urban Science The Ministry of Education Key Laboratory of Big Data and Intelligent Robot The Guangdong Provincial Key Lab of Computational Intelligence and Cyberspace Information China Singapore Management University Singapore

The vulnerability of 3D point cloud analysis to unpredictable rotations poses an open yet challenging problem: orientation-aware 3D domain generalization. Cross-domain robustness and adaptability of 3D representations are crucial but not easily achieved through rotation augmentation. Motivated by the inherent advantages of intricate orientations in enhancing generalizability, we propose an innovative rotation-adaptive domain generalization framework for 3D point cloud analysis. Our approach aims to alleviate orientational shifts by leveraging intricate samples in an iterative learning process. Specifically, we identify the most challenging rotation for each point cloud and construct an intricate orientation set by optimizing intricate orientations. Subsequently, we employ an orientation-aware contrastive learning framework that incorporates an orientation consistency loss and a margin separation loss, enabling effective learning of categorically discriminative and generalizable features with rotation consistency. Extensive experiments and ablations conducted on 3D cross-domain benchmarks firmly establish the state-of-the-art performance of our proposed approach in the context of orientation-aware 3D domain generalization. © 2025, CC BY-NC-SA.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

An Encoder-Based Framework for Privacy-Preserving Machine Learning 24th

An Encoder-Based Framework for Privacy-Preserving Machine L...

引用

24th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2024

作者： Wu, Jiayun Ren, Wei Zhang, Xianchao Zheng, Xianghan School of Computer Science China University of Geosciences Wuhan430074 China State Key Laboratory of Geo-Information Engineering and Key Laboratory of Surveying and Mapping Science and Geospatial Information Technology of MNR CASM Beijing China Ministry of Education Chengdu China Key Laboratory of Medical Electronics and Digital Health of Zhejiang Province Jiaxing University Jiaxing China Engineering Research Center of Intelligent Human Health Situation Awareness of Zhejiang Province Jiaxing University Jiaxing China College of Computer and Big Data Fuzhou University Fujian Fuzhou China School of Information and Intelligent Engineering Sanya College Hainan China

ISBN: (纸本)9789819615506

As a data-driven science, machine learning requires vast amounts of training data and computational resources. However, for highly privacy-sensitive data, it is crucial to protect the privacy of the data during both the training and utilization of machine learning models. In this paper, we propose a privacy-preserving machine learning approach using autoencoders and differential privacy mechanisms to safeguard data privacy while minimizing the impact on data availability. Specifically, we augment logistic regression and ResNet18 models with different architectures of autoencoders to perform data encryption? without compromising the machine learning tasks. Additionally, we employ differential privacy mechanisms to introduce gradient perturbations in the encoding part of the autoencoder, enhancing the algorithm’s security and further protecting data privacy. We also design the cosine similarity between the encoded and original data as a metric for evaluating data privacy, considering model performance, privacy budget, and data privacy collectively to balance data availability and privacy. Extensive experiments conducted on MNIST, CIFAR-10, PathMNIST, and BloodMNIST datasets demonstrate that for simple logistic regression models handling easily classifiable datasets, employing simple autoencoder structures can enhance classification accuracy, with significant performance impact after adding differential privacy. For ResNet18, utilizing convolutional autoencoders for data encryption generally has minimal impact on model classification performance and can even improve accuracy in most cases. Adding differential privacy has minor effects on model classification performance. Selecting appropriate model structures and privacy budgets for different usage scenarios can ensure both data availability and privacy. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A leaded sine–cosine artificial protozoa optimizer algorithm for solving multi-objective optimization problems

引用

Cluster Computing 2025年第5期28卷 1-65页

作者： Liao, Junjie Melethil Sethumadhavan, Syam Ke, Rui Gao, Zheng-Ming Zhao, Juan College of Computer and Information Technology China Three Gorges University Yichang443002 China School of Computer Engineering Jingchu University of Technology Jingmen448000 China Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang443002 China Big Data Research Center Jingchu University of Technology Jingmen448000 China School of Electronics and Information Engineering Jingchu University of Technology Jingmen448000 China

Many real-world engineering problems require simultaneous optimization of multiple objectives. The artificial protozoa optimizer (APO), a recently proposed and efficient optimization method, has demonstrated potential in solving multi-objective problems. In order to solve multi-objective problems, this paper proposes a novel algorithm based on the APO framework, the Leaded Sine Cosine Multi-objective APO (LSCMOAPO) algorithm. LSCMOAPO mainly integrates the sine–cosine algorithm (SCA) and a leader selection strategy to guide the population towards the true Pareto front. The algorithm was validated through extensive simulations involving 41 benchmark test functions (ZDT-series, UF-series, CF-series, and CEC2020-series) and five practical engineering problems. Performance was evaluated using the inverted generational distance, generational distance, Hypervolume, and spacing metrics. Comparative analysis with ten other multi-objective optimization algorithms (MOPSO, MOSSA, MOGWO, MODA, MOCryStAl, MOALO, MOGOA, MOSMA, MOGNDO, and MOAPO) also validated the better performance of LSCMOAPO in more than 86% occasions (The average of the evaluation indicators) in realizing high-quality solutions to all multi-objective problems, including linear, nonlinear, continuous, and discrete Pareto optimal front. It will set a new benchmark for new algorithm proposed. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Deep learning for parameter identification of nonlinear dynamical system driven by fractional Brownian motion

引用

Nonlinear Dynamics 2025年 1-20页

作者： Hou, Wentao Ma, Shaojuan School of Mathematics and Information Science North Minzu University Yinchuan China Department of Mathematics and Information Technology Yuncheng University Yuncheng China Ningxia Key Laboratory of Intelligent Information and Big Data Processing Yinchuan China

Accurate parameters identification in nonlinear dynamical systems is crucial for understanding complex phenomena. Fractional Brownian motion is widely used in modeling nonlinear systems due to its long memory, but it also complicates parameter identification. Traditional methods typically rely on linear models or assume Gaussian white noise, which makes them less effective for handling such complex systems. Although data-driven methods have been proposed, most of them utilize simple network architectures and focus primarily on identifying the Hurst exponent or parameters of linear systems. This paper presents a deep learning-based parameter identifier capable of jointly identifying all parameters of a nonlinear dynamical system driven by fractional Brownian motion. Firstly, a new parameter identifier (Transformer-Bidirectional Long Short-Term Memory Neural Network) is constructed by combining the advantages of Transformer in dealing with long-term dependence and Bidirectional Long Short-Term Memory in extracting local features. We then incorporate fractional Brownian motion as the random effect in the nonlinear dynamical system and provide three examples to simulate the system. Finally, the proposed identifier is used to identify parameters in the three systems individually, and its performance is compared to that of the Convolutional Neural Network-Bidirectional Long Short-Term Memory-Attention Neural Network and the Parameter Estimation Neural Network to demonstrate its effectiveness. Taking Example 3 and using the mean absolute error to evaluate identification accuracy. The mean absolute error values for the Transformer-Bidirectional Long Short-Term Memory Neural Network are 0.0171, 0.0699, 0.0328 and 0.0851, which are significantly better than the Convolutional Neural Network-Bidirectional Long Short-Term Memory-Attention Neural Network, with values of 0.0277, 0.1054, 0.0494 and 0.1037, and the Parameter Estimation Neural Network, with values of 0.0419, 0.1265,

关键词：

来源：评论

学校读者我要写书评

暂无评论

AMHF-TP:Multifunctional therapeutic peptides prediction based on multi-granularity hierarchical features

引用

Quantitative Biology 2025年第1期13卷 127-141页

作者： Shouheng Tuo YanLing Zhu Jiangkun Lin Jiewei Jiang School of Computer Science and Technology Xi’an University of Posts and TelecommunicationsXi’anChina Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi’anChina Xi’an Key Laboratory of Big Data and Intelligent Computing Xi’anChina School of Electronic Engineering Xi’an University of Posts and TelecommunicationsXi’anChina

Multifunctional therapeutic peptides(MFTP)hold immense potential in diverse therapeutic contexts,yet their prediction and identification remain challenging due to the limitations of traditional methodologies,such as extensive training durations,limited sample sizes,and inadequate generalization *** address these issues,we present AMHF-TP,an advanced method for MFTP recognition that utilizes attention mechanisms and multi-granularity hierarchical features to enhance *** AMHF-TP is composed of four key components:a migration learning module that leverages pretrained models to extract atomic compositional features of MFTP sequences;a convolutional neural network and selfattention module that refine feature extraction from amino acid sequences and their secondary structures;a hypergraph module that constructs a hypergraph for complex similarity representation between MFTP sequences;and a hierarchical feature extraction module that integrates multimodal peptide sequence *** with leading methods,the proposed AMHF-TP demonstrates superior precision,accuracy,and coverage,underscoring its effectiveness and robustness in MFTP *** comparative analysis of separate hierarchical models and the combined model,as well as with five contemporary models,reveals AMHFTP’s exceptional performance and stability in recognition tasks.

关键词： deep learning hypergraph multifunctional therapeutic peptides multi-granularity hierarchical features

来源：评论

学校读者我要写书评

暂无评论

3DFaceMAE: Pre-training of Masked Autoencoder Using Patch-Based Random Masking Reconstruction and Super-resolution for 3D Face Recognition 7th

3DFaceMAE: Pre-training of Masked Autoencoder Using Patch-Ba...

引用

7th Chinese Conference on Pattern Recognition and Computer Vision

作者： Gao, Ziqi Li, Qiufu Shen, Linlin Yang, Junpeng Shenzhen Univ Comp Vis Inst Shenzhen 518060 Guangdong Peoples R China Shenzhen Univ Natl Engn Lab Big Data Syst Comp Technol Shenzhen 518060 Guangdong Peoples R China Shenzhen Inst Artificial Intelligence Robot Soc A Shenzhen 518129 Guangdong Peoples R China Shenzhen Univ Guangdong Prov Key Lab Intelligent Informat Proc Shenzhen 518060 Guangdong Peoples R China

ISBN: (纸本)9789819787944;9789819787951

Compared to 2D face recognition, 3D face recognition exhibits stronger robustness against variations like pose and illumination. However, due to the limited training data, the accuracy of existing 3D face recognition methods is still unsatisfactory. In this paper, we introduce 3DFaceMAE, which is the first masked autoencoder (MAE) based 3D face recognition method using point clouds. Specifically, we first synthesize a large-scale 3D point cloud facial dataset and combine it with the small-scale real data. In the pre-training of 3DFaceMAE, we extract the key facial regions from the input 3D facial point cloud, using normal difference techniques, and reconstruct these key regions using patch-based random masking reconstruction and super-resolution. We finally fine-tune the encoder of 3DFaceMAE on the real 3D face point cloud data. In the experiments, we test 3DFaceMAE on three 3D face datasets, as high as 91.17% was achieved on the Lock3DFace dataset, which is the first reported result surpassing 90%. In addition, the experimental results indicate that 3DFaceMAE has strong cross-quality generalization performance. We also validate the effectiveness of different components of 3DFaceMAE through ablation study.

关键词： Face recognition 3D point cloud

来源：评论

学校读者我要写书评

暂无评论

GoPIM: GCN-Oriented Pipeline Optimization for PIM Accelerators

GoPIM: GCN-Oriented Pipeline Optimization for PIM Accelerato...

引用

IEEE Symposium on High-Performance Computer Architecture

作者： Siling Yang Shuibing He Wenjiong Wang Yanlong Yin Tong Wu Weijian Chen Xuechen Zhang Xian-He Sun Dan Feng The State Key Laboratory of Blockchain and Data Security Zhejiang University Zhejiang Lab Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security Zhejiang Key Laboratory of Big Data Intelligent Computing Washington State University Vancouver Illinois Institute of Technology Huazhong University of Science and Technology Wuhan National Laboratory for Optoelectronics

ISBN: (数字)9798331506476

ISBN: (纸本)9798331506483

Graph convolutional networks (GCNs) are popular for a variety of graph learning tasks. ReRAM-based processing-in-memory (PIM) accelerators are promising to expedite GCN training owing to their in-situ computing capability. However, existing accelerators can be severely underutilized even with pipelines, due to the oversight of the skewed execution times of various GCN stages and the ignorance of skewed degrees of graph vertices. In this work, we propose GOPIM, a GCN-oriented pipeline optimization for PIM accelerators to expedite GCN training. First, GOPIM proposes an ML-based scheme that allocates crossbar resources to the most needed stages to streamline the overall pipeline. Second, GOPIM utilizes a selective vertex updating technique that evenly distributes vertices on crossbars by interleaved mapping. These techniques collectively reduce the overall execution time without losing much accuracy. We also provide a practical architecture design for GOPIM. Our experimental results show that, GoPIM achieves up to 191 × speedup and 16.1 × energy saving, compared to the state-of-the-art work.

关键词： Training Accuracy Graph convolutional networks Pipelines Energy conservation Computer architecture Machine learning Parallel processing Optimization

来源：评论

学校读者我要写书评

暂无评论

TactCLNet: Tactile Continual Learning Network Based on Generative Replay for Object Hardness Recognition

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 10045-10055页

作者： Liu, Yiwen Yi, Zhengkun Fang, Senlin Zhang, Yupo Wan, Feng Yang, Zhi-Xin Lu, Xu Wu, Xinyu Chinese Acad Sci Shenzhen Inst Adv Technol Guangdong Prov Key Lab Robot & Intelligent Syst Shenzhen 518055 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China City Univ Macau Fac Data Sci Macau Peoples R China Univ Macau Fac Sci & Technol Ctr Cognit & Brain Sci Ctr Artificial Intelligence & RobotDept Elect & C Macau Peoples R China Univ Macau Inst Collaborat Innovat Taipa Macao Peoples R China Univ Macau Ctr Artificial Intelligence & Robot Dept Electromech Engn State Key Lab Internet Things Smart City SAR Macau Peoples R China Guangdong Polytech Normal Univ Sch Comp Sci Guangzhou 510665 Peoples R China Guangdong Polytech Normal Univ Guangdong Prov Key Lab Intellectual Property & Big Guangzhou 510665 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Shenzhen Inst Artificial Intelligence & Robot Soc SIAT Branch Shenzhen 518055 Peoples R China

Currently, deep neural networks can be extremely effective in robotic tactile perception. However, a major challenge is to solve the problem of continual learning of robotic tactile perception in an open and dynamic environment. In this paper, we propose a novel continual learning method for the domian incremental learning task in the field of tactile perception. To be specific, we introduce a morphology-specific variational autoencoders which can mitigate catastrophic forgetting by generating pseudo-samples for training in the continual learning process. We integrate the generative model and the discriminative model into one model, which reduces the size of model and improves the continual learning ability. In addition, considering the ordinal information between the hardness levels, we propose to add conditional information to the model and introduce a modified loss function to combine the latent value with the hardness information, which improves the continual learning performance by controlling the distribution and quality of pseudo-sample generation. Following this, we designed a tactile robot experiment, collected hardness data, and tested our model on this object hardness recognition task. We show experimentally that, after training, the model can still maintain the accuracy of more than 94% after learning three tasks in terms. Note to Practitioners-In the field of robotics tactile perception, the issue of continual learning in robots is a crucial problem that urgently requires resolution. We hope robots to effectively engage in continual learning across multiple tasks, ensuring the acquisition of new knowledge while mitigating the risk of forgetting previously acquired knowledge. In this paper, we propose a novel continual learning method for the domian incremental learning task. we introduce a morphology-specific variational autoencoders based on replaying pseudo-samples during continual learning process which reduces the size of model and improves the cont

关键词： Continuing education robots data models Accuracy Training Autoencoders intelligent systems Torque Incremental learning Computational modeling Continual learning robotic tactile perception object hardness recognition ordinal network

来源：评论

学校读者我要写书评

暂无评论

Using Depth-Enhanced Spatial Transformation for Student Gaze Target Estimation in Dual-View Classroom Images

Using Depth-Enhanced Spatial Transformation for Student Gaze...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Miao, Haonan Zhao, Peizheng Sun, Yuqi Nan, Fang Zhang, Xiaolong Wu, Yaqiang Tian, Feng School of Computer Science and Technology Xi'an Jiaotong University Xi'an710049 China Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi'an Jiaotong University Xi'an710049 China School of Advanced Technology Xi'an Jiaotong-Liverpool University Suzhou215123 China Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi'an Jiaotong University Xi'an710049 China

ISBN: (纸本)9798350368741

Dual-view gaze target estimation in classroom environments has not been thoroughly explored. Existing methods lack consideration of depth information, primarily focusing on 2D image information and neglecting the latent 3D spatial context, which could lead to suboptimal transformation and cause the gaze cone to intersect with an incorrect object. This paper introduces a novel dual-view gaze target estimation method tailored for classroom settings, leveraging depth-enhanced spatial transformations. By formulating a depth-enhanced 2D space, our method uses depth-enhanced spatial transformation to accurately project students' gaze cones to the teacher-oriented image. Additionally, we collected a dataset named DVSGE, specifically for student gaze target estimation in dual-view classroom images. Experimental results demonstrate significant performance improvements of 9.8% in AUC and 19.9% in L2-Distance for our method, surpassing existing methods. © 2025 IEEE.

关键词： classroom depth-enhanced 2D space dual-view gaze target estimation spatial transformation

来源：评论

学校读者我要写书评

暂无评论

Broadband High-Selectivity Stacked Filtering Antenna with Multiple Radiation Nulls

引用

Journal of Microwaves, Optoelectronics and Electromagnetic Applications 2025年第2期24卷

作者： Gao, Shuai Wang, Zhongbao Liu, Hongmei Fang, Shaojun School of Information Science and Technology Dalian Maritime University Liaoning Dalian116026 China Liaoning Key Laboratory of Radio Frequency and Big Data for Intelligent Applications Liaoning Technical University Liaoning Huludao125105 China

In this paper, a novel broadband high-selectivity stacked filtering antenna with multiple radiation nulls is proposed. The antenna consists of a short-circuited stepped impedance feeding line and two U-shaped strips on the bottom layer, four parasitic patches and a rectangular driven patch etched with two Г-shaped slots on the top layer. Short-circuited stepped impedance feeding line and Г-shaped slots respectively introduce two radiation nulls, which greatly improves low-band edge selectivity. Two U-shaped strips and four parasitic patches respectively introduce two radiation nulls, effectively increasing the upper stopband rejection level. For demonstration, one prototype of the filtering antenna with a center frequency of 2.4 GHz is fabricated and measured. Measurement results manifest that the impedance bandwidth is 15.8% (from 2.18 to 2.56 GHz) with the antenna gain of 7.69 dBi. Furthermore, four radiation nulls can be generated at 1.91, 2.01, 2.60, and 2.72 GHz, respectively. And the lower- and upper-edge selectivities are 157.3 and 310.9 dB/GHz, respectively. © 2025 SBMO/SBMag.

关键词： Microwave antennas

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：