检索结果-内蒙古大学图书馆

Symmetry Breaking in Neural network Optimization: Insights from Input Dimension Expansion

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Jun-Jie Cheng, Nan Li, Fu-Peng Wang, Xiu-Cheng Chen, Jian-Nan Pang, Long-Gang Meng, Deyu Northwest Institute of Nuclear Technology No. 28 Pingyu Road Shaanxi Xi’an710024 China School of Telecommunications Engineering Xidian University No. 2 South Taibai Road Shaanxi Xi’an710071 China State Key Laboratory of ISN No. 2 South Taibai Road Shaanxi Xi’an710071 China Institute of Particle Physics Central China Normal University No. 152 Luoyu Road Hubei Wuhan30079 China School of Mathematics and Statistics Ministry of Education Key Lab of Intelligent Networks and Network Security Xi’an Jiaotong University No. 28 Xianning West Road Shaanxi Xi’an710049 China

Understanding the mechanisms behind neural network optimization is crucial for improving network design and performance. While various optimization techniques have been developed, a comprehensive understanding of the underlying principles that govern these techniques remains elusive. Specifically, the role of symmetry breaking, a fundamental concept in physics, has not been fully explored in neural network optimization. This gap in knowledge limits our ability to design networks that are both efficient and effective. Here, we propose the symmetry breaking hypothesis to elucidate the significance of symmetry breaking in enhancing neural network optimization. We demonstrate that a simple input expansion can significantly improve network performance across various tasks, and we show that this improvement can be attributed to the underlying symmetry breaking mechanism. We further develop a metric to quantify the degree of symmetry breaking in neural networks, providing a practical approach to evaluate and guide network design. Our findings confirm that symmetry breaking is a fundamental principle that underpins various optimization techniques, including dropout, batch normalization, and equivariance. By quantifying the degree of symmetry breaking, our work offers a practical technique for performance enhancement and a metric to guide network design without the need for complete datasets and extensive training processes. © 2024, CC0.

关键词：

Novel Color Orthogonal Moments-Based Image Representation and Recognition

学校读者我要写书评

暂无评论

Research Square

Research Square 2020年

作者： He, Bing Liu, Jun Yang, Tengfei Xiao, Bin Peng, Yanguo School of Physics and Electrical Engineering Weinan Normal University Weinan714000 China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi’an University of Posts & Telecommunications Xi’an710121 China School of Computer Science and Technology Weinan Normal University Weinan714000 China School of Cyberspace Security Xi’an University of Posts & Telecommunications Xi’an710121 China Chongqing Key Laboratory of Computational Intelligence Chongqing University of Posts and Telecommunications Chongqing400065 China School of Computer Science and Technology Xidian University Xi’an710071 China

Inspired by quaternion algebra and the idea of fractional-order transformation, we propose a new set of quaternion fractional-order generalized Laguerre orthogonal moments (QFr-GLMs) based on fractional-order generalized Laguerre polynomials. Firstly, the proposed QFr-GLMs are directly constructed in Cartesian coordinate space, avoiding the need for conversion between Cartesian and polar coordinates;therefore, they are better image descriptors than circularly orthogonal moments constructed in polar coordinates. Moreover, unlike the latest Zernike moments based on quaternion and fractional-order transformations, which extract only the global features from color images, our proposed QFr-GLMs can extract both the global and local color features. This paper also derives a new set of invariant color-image descriptors by QFr-GLMs, enabling geometric-invariant pattern recognition in color images. Finally, the performances of our proposed QFr-GLMs and moment invariants were evaluated in simulation experiments of correlated color images. Both theoretical analysis and experimental results demonstrate the value of the proposed QFr-GLMs and their geometric invariants in the representation and recognition of color images. © 2020, CC BY.

关键词： Image reconstruction

A Multichannel CNN-GRU Hybrid Architecture for sEMG Gesture Recognition

学校读者我要写书评

暂无评论

A Multichannel CNN-GRU Hybrid Architecture for sEMG Gesture ...

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Shouliang Song Anming Dong Jiguo Yu Yubing Han You Zhou Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China School of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Big Data Institute Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Shandong Haikan New Media Research Institute Co. Ltd.

Surface electromyography (sEMG) signal is a physiological electrical signal produced by muscle contraction. Different gestures can be effectively recognized from the characteristics of the sEMG signal. Currently, convolutional neural networks (CNNs) have been widely used in sEMG gesture recognition systems due to their capabilities in acquiring spatial features of sEMG signals. However, these classical CNNs are inefficient in extracting temporal correlation that resides in the time serials of sEMG signals, which is definitely important for gesture recognition. To overcome such a drawback of traditional CNN-based gesture recognition methods, we propose a multichannel hybrid deep learning model for gesture recognition by combining the multichannel CNNs with a gated recurrent unit (GRU). Specifically, we use multiple CNNs to preprocess the original multichannel EMG signals in a one-by-one manner to obtain the spatial features in the current observing window. The outputs of the multiple CNNs are concatenated and fed to a temporal-feature extracting module, which is designed by cascading a GRU with an attention mechanism. Through the GRU, the temporal features of successive signal frames can be established, while the attention mechanism is introduced to further focus on the key information in recognizing the gestures, which is beneficial to improve the robustness and accuracy of the model. Experiments show that the recognition accuracy of the proposed method reaches 97.6% and 96.7% on the Ninapro DB2 and Ninapro DB5 datasets, respectively. Compared with the classical CNN method, the performance improvement is 2.9% and xx% higher than that of the traditona CNN model, respectively.

关键词：

Investigating Bi-level optimization for learning and vision from a unified perspective: A survey and beyond

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Risheng Gao, Jiaxin Zhang, Jin Meng, Deyu Lin, Zhouchen DUT-RU International School of Information Science & Engineering Dalian University of Technology Key Laboratory for Ubiquitous Network Service Software of Liaoning Province Dalian116024 China Department of Mathematics Southern University of Science and Technology National Center for Applied Mathematics Shenzhen China School of Mathematics and Statistics Ministry of Education Key Lab of Intelligent Networks and Network Security Xi’an Jiaotong University Xi’an Shaanxi China School of Electronics Engineering and Computer Science Peking University Beijing100871 China Cooperative Medianet Innovation Center Shanghai Jiao Tong University Shanghai200240 China

Bi-Level Optimization (BLO) is originated from the area of economic game theory and then introduced into the optimization community. BLO is able to handle problems with a hierarchical structure, involving two levels of optimization tasks, where one task is nested inside the other. In machine learning and computer vision fields, despite the different motivations and mechanisms, a lot of complex problems, such as hyper-parameter optimization, multi-task and meta learning, neural architecture search, adversarial learning and deep reinforcement learning, actually all contain a series of closely related subproblms. In this paper, we first uniformly express these complex learning and vision problems from the perspective of BLO. Then we construct a best-response-based single-level reformulation and establish a unified algorithmic framework to understand and formulate mainstream gradient-based BLO methodologies, covering aspects ranging from fundamental automatic differentiation schemes to various accelerations, simplifications, extensions and their convergence and complexity properties. Last but not least, we discuss the potentials of our unified BLO framework for designing new algorithms and point out some promising directions for future research. © 2021, CC BY.

关键词： Deep learning

STR: Secure computation on additive shares using the share-transform-reveal strategy

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Xia, Zhihua Gu, Qi Zhou, Wenhao Xiong, Lizhi Weng, Jian Xiong, Naixue N. College of Cyber Security Jinan University Guangzhou510632 China Engineering Research Center of Digital Forensics Ministry of Education School of Computer and Software Jiangsu Engineering Center of Network Monitoring Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology Nanjing University of Information Science & Technology Nanjing210044 China Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin541004 China The College of Intelligence and Computing Tianjin University Tianjin300350 China

The rapid development of cloud computing probably benefits many of us while the privacy risks brought by semi-honest cloud servers have aroused the attention of more and more people and legislatures. In the last two decades, plenty of works seek to outsource various specific tasks to servers while ensuring the security of private data. The tasks to be outsourced are countless;however, the computations involved are similar. In this paper, we construct a series of novel protocols that support the secure computation of various functions on numbers (e.g., the basic elementary functions) and matrices (e.g., the calculation of eigenvectors and eigenvalues) on arbitrary n ≥ 2 servers. All protocols only require constant rounds of interactions and achieve low computation complexity. Moreover, the proposed n-party protocols ensure the security of private data even though n − 1 servers collude. The convolutional neural network models are utilized as the case studies to verify the protocols. The theoretical analysis and experimental results demonstrate the correctness, efficiency, and security of the proposed protocols. Copyright © 2020, The Authors. All rights reserved.

关键词： Additives

Robust Imagined Speech Production from Electrocorticography with Adaptive Frequency Enhancement

学校读者我要写书评

暂无评论

Robust Imagined Speech Production from Electrocorticography ...

International Joint Conference on Neural networks (IJCNN)

作者： Xu Xu Chong Fu Junxin Chen Gwanggil Jeon David Camacho School of Computer Science and Engineering Northeastern University Shenyang China School of Software Dalian University of Technology Dalian China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Northeastern University Shenyang China Engineering Research Center of Security Technology of Complex Network System Ministry of Education Shenyang China College of Information Technology Incheon National University Incheon Korea Department of Computer Systems Engineering Universidad Politécnica de Madrid Madrid Spain

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Imagined speech production with electrocorticography (ECoG) plays a crucial role in brain-computer interface system. A challenging issue is the great variation underlying the frequency bands of the ECoG signals’ encode information, which makes current methods difficult to generate imagined speech with stable quality among different persons. To this end, we propose a robust model to generate high-quality imagined speech from ECoG. A frequency enhancement branch is first designed to adaptively modulate the frequency information, whose product is fed into the following multi-scale channel attention module for robust feature extraction and fusion. By incorporating both the mel-spectrum and audio as training constraints, a multi-constraint decoder branch is finally constructed for imagined speech production. The performance of our model is evaluated on a high-quality dateset, i.e, Single Word Production Dutch-iBIDS. It yields Pearson correlation scores that are all above 0.8, and the standard deviationsare are all below 0.2 in different volunteers. Experimental results demonstrate that our model is effective and robust for ECoG based imagined speech production, and has advantages over peer methods.

关键词： Training Adaptation models Frequency modulation Correlation Neuroscience Production Speech enhancement

G²Face: High-Fidelity Reversible Face Anonymization via Generative and Geometric Priors

学校读者我要写书评

暂无评论

IEEE Transactions on Information Forensics and security 2024年 19卷 8773-8785页

作者： Haoxin Yang Xuemiao Xu Cheng Xu Huaidong Zhang Jing Qin Yi Wang Pheng-Ann Heng Shengfeng He School of Computer Science and Engineering South China University of Technology Guangzhou China Guangdong Engineering Center for Large Model and GenAI Technology State Key Laboratory of Subtropical Building and Urban Science Guangzhou China Centre for Smart Health The Hong Kong Polytechnic University Hong Kong China School of Future Technology South China University of Technology Guangzhou China School of Computer Science and Network Security Dongguan University of Technology Dongguan China Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong SAR China School of Computing and Information Systems Singapore Management University Victoria St Singapore

Reversible face anonymization, unlike traditional face pixelization, seeks to replace sensitive identity information in facial images with synthesized alternatives, preserving privacy without sacrificing image clarity. Traditional methods, such as encoder-decoder networks, often result in significant loss of facial details due to their limited learning capacity. Additionally, relying on latent manipulation in pre-trained GANs can lead to changes in ID-irrelevant attributes, adversely affecting data utility due to GAN inversion inaccuracies. This paper introduces G2Face, which leverages both generative and geometric priors to enhance identity manipulation, achieving high-quality reversible face anonymization without compromising data utility. We utilize a 3D face model to extract geometric information from the input face, integrating it with a pre-trained GAN-based decoder. This synergy of generative and geometric priors allows the decoder to produce realistic anonymized faces with consistent geometry. Moreover, multi-scale facial features are extracted from the original face and combined with the decoder using our novel identity-aware feature fusion blocks (IFF). This integration enables precise blending of the generated facial patterns with the original ID-irrelevant features, resulting in accurate identity manipulation. Extensive experiments demonstrate that our method outperforms existing state-of-the-art techniques in face anonymization and recovery, while preserving high data utility. Code is available at https://***/Harxis/G2Face .

关键词： Faces Data privacy Information integrity Information filtering Face recognition Generative adversarial networks Feature extraction

Outage Analysis of Aerial Semi-Grant-Free NOMA Systems

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lei, Hongjiang Zhu, Chen Park, Ki-Hong Ansari, Imran Shafique Lei, Weijia Tang, Hong Kim, Kyeong Jin The School of Communication and Information Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China Chongqing Key Lab of Mobile Communications Technology Chongqing400065 China Shaanxi Key Laboratory of Information Communication Network and Security Xi’an University of Posts and Telecommunications Shaanxi Xi’an710121 China Thuwal23955-6900 Saudi Arabia James Watt School of Engineering University of Glasgow GlasgowG12 8QQ United Kingdom The Mitsubishi Electric Research Laboratories CambridgeMA02139 United States

In this paper, we analyze the outage performance of Unmanned Aerial Vehicles (UAVs)-enabled downlink Non-Orthogonal Multiple Access (NOMA) communication systems with the Semi-Grant-Free (SGF) transmission scheme. A UAV provides coverage services for a Grant-Based (GB) user and one Grant-Free (GF) user is allowed to utilize the same channel resource opportunistically. The analytical expressions for the exact and asymptotic Outage Probability (OP) of the GF user are derived. The results demonstrate that no-zero diversity order can be achieved only under stringent conditions on users’ quality of service requirements. Subsequently, an efficient Dynamic Power Allocation (DPA) scheme is proposed to relax such data rate constraints. The analytical expressions for the exact and asymptotic OP of the GF user with the DPA scheme are derived. Finally, Monte Carlo simulation results are presented to validate the correctness of the derived analytical expressions and demonstrate the effects of the UAV’s location and altitude on the OP of the GF user. © 2022, CC BY.

关键词： Antennas

Adaptively secure key-policy ABE scheme with fast decryption

学校读者我要写书评

暂无评论

Tongxin Xuebao/Journal on Communications 2014年 35卷 26-32页

作者： Li, Qi Ma, Jian-Feng Xiong, Jin-Bo Liu, Xi-Meng School of Computer Science and Technology Xidian University Xi'an710071 China Shaanxi Key Laboratory of Network and System Security Xidian University Xi'an710071 China Science and Technology on Communication Information Security Control Laboratory Jiaxing314033 China Faculty of Software Fujian Normal University Fuzhou350108 China School of Telecommunication Engineering Xidian University Xi'an710071 China

In the existing adaptively secure ABE (attribute-based encryption) schemes, the decryption cost goes linearly with the number of attributes that are used in decryption. An adaptively secure key-policy ABE (FKP-ABE) scheme with fast decryption is proposed, where constant size of computation overhead is implemented in decryption. This scheme is constructed on composite order bilinear groups and supports any monotonic access structure that can be expressed by LSSS (linear secret sharing schemes). The proposed scheme is proved to be adaptively secure in the standard model. ©, 2014, Editorial Board of Journal on Communications. All right reserved.

关键词： Cryptography