检索结果-内蒙古大学图书馆

Small Evolution Antenna Design Method Based on Dynamic Hill Climbing Algorithm and Orthogonal Experiment

学校读者我要写书评

暂无评论

Small Evolution Antenna Design Method Based on Dynamic Hill ...

2019 High Performance computing and Computational Intelligence Conference, HPCCI 2019

作者： Shi, Hui Li, Junjie Xie, Zhiming Department of Information Engineering Shanwei Polytechnic Shanwei Guangdong516600 China Research Institute of Cloud Computing and Data Center Project Design Shanwei Innovation Industrial Design and Research Institute Shanwei Guangdong516600 China Professional Committee Cloud Computing and Big Data in Higher Vocational College Higher Education Institute in Guangdong Province Shanwei Guangdong516600 China

Small-scale evolution antennas have the advantages of high gain, wide bandwidth and small size, and have broad application prospects in the aerospace field. Aiming at the complex high-dimensional and dynamic problems of aerial antenna modeling, a dynamic hill-climbing algorithm based on orthogonal design is designed and implemented to simulate and optimize the single-span antenna in high-frequency operation mode. In the optimization problem, this paper defines a new individual comparison mechanism, which can reflect the number of violations of constraints and the degree of violation of each constraint, making the comparison of individuals more reasonable. The antenna simulation results demonstrate the effectiveness and feasibility of using dynamic evolutionary algorithms for antenna design. © Published under licence by IOP Publishing Ltd.

关键词： Antennas

Bridging prediction and decision: Advances and challenges in data-driven optimization

学校读者我要写书评

暂无评论

Nexus 2025年第1期2卷

作者： Yanzhi Wang Jianxiao Wang Haoran Zhang Jie Song Department of Industrial Engineering and Management College of Engineering Peking University Beijing 100871 China National Engineering Laboratory for Big Data Analysis and Applications Peking University Beijing 100871 China PKU-Changsha Institute for Computing and Digital Economy Changsha 410000 China School of Urban Planning and Design Peking University Shenzhen Guangdong 518055 China

data-driven approaches have revolutionized traditional optimization methods by integrating prediction with decision-making. This review examines the theoretical foundations, strengths, recent advancements, and limitations of three key methods—sequential optimization, end-to-end learning, and direct learning—highlighting their practical applications in power grid scheduling, operations management, and intelligent autonomous control. A multidimensional comparison is presented, followed by a discussion of the challenges in data-centric methodology, optimization methodology, and decision-making application. This paper offers a methodological guide and outlines future directions for academia and industry to enhance decision-making in complex data environments.

关键词： convex optimization data-centric AI data-driven decision-making end-to-end learning intelligent autonomous control machine learning operations management power grid scheduling reinforcement learning uncertainty estimation

Extracting hierarchical relationships of aspects from reviews

学校读者我要写书评

暂无评论

Applied Soft computing 2025年 178卷

作者： Jiangtao Qiu Ling Lin Siyu Wang Chongqing Engineering Research Center for Intelligent Applications of Financial Big Data Chongqing Institute of Engineering Chongqing China School of Computing and Artificial Intelligence Southwestern University of Finance and Economics Chengdu China College of Artificial Intelligence (CUIT Shuangliu Industrial College) Chengdu University of Information Technology Chengdu China Gusu Laboratory of Materials Suzhou China

Aspect Based Sentiment Analysis (ABSA) attracts significant attention in recent years. Three elements of ABSA including aspect term, aspect, and domain (or entity) present a hierarchical relationships in e-commerce reviews. Extracting the hierarchical relationships can significantly enhance various applications, such as creating user profiles, identifying hierarchical topics, and visualizing review data. In this study, we proposed a framework to tackle this task, consisting of two primary components: a text adversarial autoencoder that efficiently encodes review content, and a deep network that extracts the clusters of aspect terms from review dataset and organizes them to a hierarchical structure using the Student-Teacher paradigm. Our framework also addresses the challenge of acquiring labeled training data by utilizing self-supervised learning. We evaluated the proposed framework on three public datasets and observed that it outperforms baseline models, indicating the feasibility and effectiveness of our approach.

关键词： Self-supervised learning

Invisible backdoor attack with sample-specific triggers

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Li, Yuezun Li, Yiming Wu, Baoyuan Li, Longkang He, Ran Lyu, Siwei Ocean University of China Qingdao China School of Data Science The Chinese University of Hong Kong Shenzhen China Secure Computing Lab of Big Data Shenzhen Research Institute of Big Data Shenzhen China Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen China NLPR/CRIPAC Institute of Automation Chinese Academy of Sciences Beijing China University at Buffalo SUNY NY United States

Recently, backdoor attacks pose a new security threat to the training process of deep neural networks (DNNs). Attackers intend to inject hidden backdoors into DNNs, such that the attacked model performs well on benign samples, whereas its prediction will be maliciously changed if hidden backdoors are activated by the attacker-defined trigger. Existing backdoor attacks usually adopt the setting that triggers are sample-agnostic, i.e., different poisoned samples contain the same trigger, resulting in that the attacks could be easily mitigated by current backdoor defenses. In this work, we explore a novel attack paradigm, where backdoor triggers are sample-specific. In our attack, we only need to modify certain training samples with invisible perturbation, while not need to manipulate other training components (e.g., training loss, and model structure) as required in many existing attacks. Specifically, inspired by the recent advance in DNN-based image steganography, we generate sample-specific invisible additive noises as backdoor triggers by encoding an attacker-specified string into benign images through an encoder-decoder network. The mapping from the string to the target label will be generated when DNNs are trained on the poisoned dataset. Extensive experiments on benchmark datasets verify the effectiveness of our method in attacking models with or without defenses. The code will be available at https://***/yuezunli/ISSBA. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

Towards Tailored Models on Private AIoT Devices: Federated Direct Neural Architecture Search

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhang, Chunhui Yuan, Xiaoming Zhang, Qianyun Zhu, Guangxu Cheng, Lei Zhang, Ning The Michtom School of Computer Science Brandeis University WalthamMA United States The Qinhuangdao Branch Campus Northeastern University Qinhuangdao066004 China The School of Cyber Science and Technology Beihang University Beijing China The Shenzhen Research Institute of Big Data Guangdong Shenzhen China The College of Information Science and Electronic Engineering Zhejiang University China Hangzhou310027 China The Department of Electrical and Computing Engineering University of Windsor WindsorONN9B 3P4 Canada

Neural networks often encounter various stringent resource constraints while deploying on edge devices. To tackle these problems with less human efforts, automated machine learning becomes popular in finding various neural architectures that fit diverse Artificial Intelligence of Things (AIoT) scenarios. Recently, to prevent the leakage of private information while enable automated machine intelligence, there is an emerging trend to integrate federated learning and neural architecture search (NAS). Although promising as it may seem, the coupling of difficulties from both tenets makes the algorithm development quite challenging. In particular, how to efficiently search the optimal neural architecture directly from massive non-independent and identically distributed (non-IID) data among AIoT devices in a federated manner is a hard nut to crack. In this paper, to tackle this challenge, by leveraging the advances in ProxylessNAS, we propose a Federated Direct Neural Architecture Search (FDNAS) framework that allows for hardware-friendly NAS from non-IID data across devices. To further adapt to both various data distributions and different type of devices with heterogeneous embedded hardware platforms, inspired by meta-learning, a Cluster Federated Direct Neural Architecture Search (CFDNAS) framework is proposed to achieve device-aware NAS, in the sense that each device can learn a tailored deep learning model for its particular data distribution and hardware constraint. Extensive experiments on non-IID datasets have shown the state-of-the-art accuracy-efficiency trade-offs achieved by the proposed solution in the presence of both data and device heterogeneity. © 2022, CC BY.

关键词： Network architecture

Cooperative Sensing and Heterogeneous Information Fusion in VCPS: A Multi-agent Deep Reinforcement Learning Approach

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xu, Xincao Liu, Kai Dai, Penglin Xie, Ruitao Cao, Jingjing Luo, Jiangtao The College of Computer Science Chongqing University Chongqing400040 China The School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China The National Engineering Laboratory of Integrated Transportation Big Data Application Technology Chengdu611756 China The College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China The School of Transportation and Logistics Engineering Wuhan University of Technology Hubei430063 China The Electronic Information and Networking Research Institute Chongqing University of Posts and Telecommunications Chongqing400065 China

Cooperative sensing and heterogeneous information fusion are critical to realize vehicular cyber-physical systems (VCPSs). This paper makes the first attempt to quantitatively measure the quality of VCPS by designing a new metric called Age of View (AoV). Specifically, we first present the system architecture where heterogeneous information can be cooperatively sensed and uploaded via vehicle-to-infrastructure (V2I) communications in vehicular edge computing (VEC). Logical views are constructed by fusing the heterogeneous information at edge nodes. Further, we formulate the problem by deriving a cooperative sensing model based on the multi-class M/G/1 priority queue, and defining the AoV by modeling the timeliness, completeness and consistency of the logical views. On this basis, a multi-agent deep reinforcement learning solution is proposed. In particular, the system state includes vehicle sensed information, edge cached information and view requirements. The vehicle action space consists of the sensing frequencies and uploading priorities of information. A difference-reward-based credit assignment is designed to divide the system reward, which is defined as the VCPS quality, into the difference reward for vehicles. Edge node allocates V2I bandwidth to vehicles based on predicted vehicle trajectories and view requirements. Finally, we build the simulation model and give a comprehensive performance evaluation, which conclusively demonstrates the superiority of the proposed solution. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

Dual-View Pyramid Pooling in Deep Neural Networks for Improved Medical Image Classification and Confidence Calibration

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Xiaoqing Nie, Qiushi Xiao, Zunjie Zhao, Jilu Wu, Xiao Guo, Pengxin Li, Runzhi Liu, Jin Wei, Yanjie Pan, Yi Center for High Performance Computing Shenzhen Key Laboratory of Intelligent Bioinformatics Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Shenzhen518055 China Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China Department of Statistics and Actuarial Science The University of Hong Kong 999077 Hong Kong Cooperative Innovation Center of Internet Healthcare Zhengzhou University Zhengzhou450001 China Hunan Provincial Key Lab on Bioinformatics School of Computer Science and Engineering Central South University Changsha410083 China Xinjiang Engineering Research Center of Big Data and Intelligent Software School of software Xinjiang University Wulumuqi830046 China Faculty of Computer Science and Control Engineering Shenzhen University of Advanced Technology Shenzhen China

Spatial pooling (SP) and cross-channel pooling (CCP) operators have been applied to aggregate spatial features and pixel-wise features from feature maps in deep neural networks (DNNs), respectively. Their main goal is to reduce computation and memory overhead without visibly weakening the performance of DNNs. However, SP often faces the problem of losing the subtle feature representations, while CCP has a high possibility of ignoring salient feature representations, which may lead to both miscalibration of confidence issues and suboptimal medical classification results. To address these problems, we propose a novel dual-view framework, the first to systematically investigate the relative roles of SP and CCP by analyzing the difference between spatial features and pixel-wise features. Based on this framework, we propose a new pooling method, termed dual-view pyramid pooling (DVPP), to aggregate multi-scale dual-view features. DVPP aims to boost both medical image classification and confidence calibration performance by fully leveraging the merits of SP and CCP operators from a dual-axis perspective. Additionally, we discuss how to fulfill DVPP with five parameter-free implementations. Extensive experiments on six 2D/3D medical image classification tasks show that our DVPP surpasses state-of-the-art pooling methods in terms of medical image classification results and confidence calibration across different DNNs. © 2024, CC BY.

关键词： Deep neural networks

Turbulent pressure support in galaxy clusters Impact of the hydrodynamical solver

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Groth, Frederick Valentini, Milena Steinwandel, Ulrich P. Vallés-Pérez, David Dolag, Klaus Universitäts-Sternwarte Fakultät für Physik Ludwig-Maximilians-Universität München Scheinerstr. 1 München81679 Germany Astronomy Unit Department of Physics University of Trieste via Tiepolo 11 TriesteI-34131 Italy INAF - Osservatorio Astronomico di Trieste via Tiepolo 11 TriesteI-34131 Italy ICSC - Italian Research Center on High Performance Computing Big Data and Quantum Computing via Magnanelli 2 Casalecchio di Reno40033 Italy Center for Computational Astrophysics Flatiron Institute 162 Fifth Avenue New YorkNY10010 United States Departament d’Astronomia i Astrofísica Universitat de València C/Doctor Moliner 50 Burjassot ValènciaE-46100 Spain Max-Planck-Institut für Astrophysik Karl-Schwarzschild-Straße 1 Garching85741 Germany

Context. The amount of turbulent pressure in galaxy clusters is still debated, especially as for the impact of the dynamical state and the hydro-method used for simulations. Aims. We study the turbulent pressure fraction in the intra cluster medium of massive galaxy clusters. We aim to understand the impact of the hydrodynamical scheme, analysis method, and dynamical state on the final properties of galaxy clusters from cosmological simulations. Methods. We perform non-radiative simulations of a set of zoom-in regions of seven galaxy clusters with Meshless Finite Mass (MFM) and Smoothed Particle Hydrodynamics (SPH). We use three different analysis methods based on: (i) the deviation from hydrostatic equilibrium, (ii) the solenoidal velocity component obtained by a Helmholtz-Hodge decomposition, and (iii) the small-scale velocity obtained through a multi-scale filtering approach. We split the sample of simulated clusters into active and relaxed clusters. Results. Our simulations predict an increased turbulent pressure fraction for active compared to relaxed clusters. This is especially visible for the velocity-based methods. For these, we also find increased turbulence for the MFM simulations compared to SPH, consistent with findings from more idealized simulations. The predicted non-thermal pressure fraction varies between a few percent for relaxed clusters and ≈ 13% for active ones within the cluster center and increases towards the outskirts. No clear trend with redshift is visible. Conclusions. Our analysis quantitatively assesses the importance played by the hydrodynamical scheme and the analysis method to determine the non-thermal/turbulent pressure fraction. While our setup is relatively simple (non-radiative runs), our simulations show agreement with previous, more idealized simulations, and make a step further toward the understanding of turbulence. Copyright © 2024, The Authors. All rights reserved.

关键词： Cosmology

Magnon trap by chiral spin pumping

学校读者我要写书评

暂无评论

Physical Review B 2020年第5期102卷 054429-054429页

作者： Tao Yu Hanchen Wang Michael A. Sentef Haiming Yu Gerrit E. W. Bauer Max Planck Institute for the Structure and Dynamics of Matter Luruper Chaussee 149 22761 Hamburg Germany Fert Beijing Institute School of Microelectronics Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing 100191 China WPI-AIMR Institute for Materials Research and CSRN Tohoku University Sendai 980-8577 Japan

Chiral spin pumping is the generation of a unidirectional spin current in half of ferromagnetic films or conductors by dynamic dipolar stray fields from close-by nanomagnets. We formulate a general theory of long-range chiral interactions between magnets mediated by unidirectional traveling waves, e.g., spin waves in a magnetic film or microwaves in a waveguide. The traveling waves emitted by an excited magnet can be perfectly trapped by a second, initially passive magnet by a dynamical interference effect. When both magnets are excited by a uniform microwave, the chiral interaction between them creates a large imbalance in their magnon numbers.

关键词： Chirality Magnetization dynamics Magnons Spin blockade Magnetic mirrors & traps Transistors