检索结果-内蒙古大学图书馆

National Science Review 2023年第6期10卷 13-15页

作者： Yinpeng Dong Chang Liu Wenzhao Xiang Hang Su Jun Zhu Department of Computer Science and Technology Institute for AI Tsinghua-Bosch Joint ML Center THBI Lab BNRist Center Tsinghua University Institute of Image Communication and Networks Engineering in the Department of Electronic Engineering (EE) Shanghai Jiao Tong University Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Peng Cheng Laboratory Pazhou Laboratory (Huangpu)

PROBLEM In recent years,the rapid development of artificial intelligence (AI) technology,especially machine learning and deep learning, is profoundly changing human production and *** various fields,such as robotics,face recognition,autonomous driving and healthcare,AI is playing an important ***,although AI is promoting the technological revolution and industrial progress,its security risks are often *** studies have found that the wellperforming deep learning models are extremely vulnerable to adversarial examples [1-3].The adversarial examples are crafted by applying small,humanimperceptible perturbations to natural examples,but can mislead deep learning models to make wrong *** vulnerability of deep learning models to adversarial examples can raise security and safety threats to various realworld applications.

关键词： robustness adversarial example adversarial training deep learning

来源：评论

学校读者我要写书评

暂无评论

A Compact Penta-Band Printed Monopole Antenna for Multiple Wireless communication Systems

A Compact Penta-Band Printed Monopole Antenna for Multiple W...

引用

International Conference on Futuristic communication and Network Technologies, CFCNT 2020

作者： Das, Sudipta Chowdhury, Apurba Lala, Bikram Dwivedi, Ravi Prakash Vasu Babu, K. EL Ghzaoui, Mohammed Department of Electronics and Communication Engineering IMPS College of Engineering and Technology Malda West Bengal India Mallabhum Institute of Polytechnic Bishnupur Bankura India SENSE VIT University Chennai Campus Chennai Tamilnadu India Department of Electronics and Communication Engineering Vasireddy Venkatadri Institute of Technology Guntur India Intelligent Systems Networks and Telecommunications Team IMAGE Laboratory ENS-Moulay Ismail University of Meknes Meknes Morocco

ISBN: (纸本)9789811646249

A compact offset microstrip line-fed printed multiband monopole antenna is proposed. In this work, the suggested antenna operates at multiple resonant frequencies due to the incorporation of narrow parallel slits and a hammer-shaped slot on the radiating patch with a modified ground plane. The proposed antenna has been configured on a 26 × 25 mm2 FR-4 substrate (Εr= 4.4 and thickness = 1.6 mm). The simulated return loss results confirm the penta-band resonances from 2.9 to 3.2 GHz, 3.22 to 3.65 GHz, 5.6 to 6.5 GHz, 7.8–8.4 GHz, and 9.4–11.1 GHz to support multi-standard wireless systems such as maritime radio navigation, Wi-MAX, WLAN, X-band satellite communication uplink, ITU band, and aeronautical radio navigation bands. The simulated antenna parameters such as VSWR, radiation pattern, gain and directivity along with surface current distribution have been analyzed in detail and presented in this article. The designed structure maintains VSWR © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Microstrip antennas

来源：评论

学校读者我要写书评

暂无评论

AEMIM: Adversarial Examples Meet Masked image Modeling

arXiv

引用

arXiv 2024年

作者： Xiang, Wenzhao Liu, Chang Su, Hang Yu, Hongyang Beijing100190 China Peng Cheng Laboratory Shenzhen10587 China University of the Chinese Academy of Sciences Beijing100049 China Institute of Image Communication and Networks Engineering Department of Electronic Engineering Shanghai Jiao Tong University Shanghai200240 China Department of Computer Science and Technology Tsinghua University Beijing100084 China

Masked image modeling (MIM) has gained significant traction for its remarkable prowess in representation learning. As an alternative to the traditional approach, the reconstruction from corrupted images has recently emerged as a promising pretext task. However, the regular corrupted images are generated using generic generators, often lacking relevance to the specific reconstruction task involved in pre-training. Hence, reconstruction from regular corrupted images cannot ensure the difficulty of the pretext task, potentially leading to a performance decline. Moreover, generating corrupted images might introduce an extra generator, resulting in a notable computational burden. To address these issues, we propose to incorporate adversarial examples into masked image modeling, as the new reconstruction targets. Adversarial examples, generated online using only the trained models, can directly aim to disrupt tasks associated with pre-training. Therefore, the incorporation not only elevates the level of challenge in reconstruction but also enhances efficiency, contributing to the acquisition of superior representations by the model. In particular, we introduce a novel auxiliary pretext task that reconstructs the adversarial examples corresponding to the original images. We also devise an innovative adversarial attack to craft more suitable adversarial examples for MIM pre-training. It is noted that our method is not restricted to specific model architectures and MIM strategies, rendering it an adaptable plug-in capable of enhancing all MIM methods. Experimental findings substantiate the remarkable capability of our approach in amplifying the generalization and robustness of existing MIM methods. Notably, our method surpasses the performance of baselines on various tasks, including imageNet, its variants, and other downstream tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

MULTITRUST: a comprehensive benchmark towards trustworthy multimodal large language models 24

MULTITRUST: a comprehensive benchmark towards trustworthy mu...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Yichi Zhang Yao Huang Yitong Sun Chang Liu Zhe Zhao Zhengwei Fang Yifan Wang Huanran Chen Xiao Yang Xingxing Wei Hang Su Yinpeng Dong Jun Zhu Dept. of Comp. Sei. and Tech. Institute for AI Tsinghua-Boseh Joint ML Center THBI Lab BNRist Center Tsinghua University Beijing China and RealAI Institute of Artificial Intelligence Beihang University Beijing China Institute of Image Communication and Networks Engineering in the Department of Electronic Engineering (EE) Shanghai Jiao Tong University Shanghai 200240 China RealAI Dept. of Comp. Sei. and Tech. Institute for AI Tsinghua-Boseh Joint ML Center THBI Lab BNRist Center Tsinghua University Beijing China Dept. of Comp. Sei. and Tech. Institute for AI Tsinghua-Boseh Joint ML Center THBI Lab BNRist Center Tsinghua University Beijing China and Pazhou Lab (Huangpu) Guangzhou China

ISBN: (纸本)9798331314385

Despite the superior capabilities of Multimodal Large Language Models (MLLMs) across diverse tasks, they still face significant trustworthiness challenges. Yet, current literature on the assessment of trustworthy MLLMs remains limited, lacking a holistic evaluation to offer thorough insights into future improvements. In this work, we establish MultiTrust, the first comprehensive and unified benchmark on the trustworthiness of MLLMs across five primary aspects: truthfulness, safety, robustness, fairness, and privacy. Our benchmark employs a rigorous evaluation strategy that addresses both multimodal risks and cross-modal impacts, encompassing 32 diverse tasks with self-curated datasets. Extensive experiments with 21 modern MLLMs reveal some previously unexplored trustworthiness issues and risks, highlighting the complexities introduced by the multimodality and underscoring the necessity for advanced methodologies to enhance their reliability. For instance, typical proprietary models still struggle with the perception of visually confusing images and are vulnerable to multimodal jailbreaking and adversarial attacks; MLLMs are more inclined to disclose privacy in text and reveal ideological and cultural biases even when paired with irrelevant images in inference, indicating that the multimodality amplifies the internal risks from base LLMs. Additionally, we release a scalable toolbox for standardized trustworthiness research, aiming to facilitate future advancements in this important field. Code and resources are publicly available at: https://***/.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Frequency-Aware Physics-Inspired Degradation Model for Real-World image Super-Resolution

arXiv

引用

arXiv 2021年

作者： Dong, Zhenxing Cao, Hong Shen, Wang Gan, Yu Ling, Yuye Zhai, Guangtao Su, Yikai John Hopcroft Center for Computer Science Shanghai Jiao Tong University China Institute of Image Communication and Network Engineering Shanghai Jiao Tong University China Department of Electrical and Computer Engineering University of Alabama TuscaloosaAL35401 United States State Key Lab of Advanced Optical Communication Systems and Networks Shanghai Jiao Tong University China

Current learning-based single image super-resolution (SISR) algorithms underperform on real data due to the deviation in the assumed degradation process from that in the real-world scenario. Conventional degradation processes consider applying blur, noise, and downsampling (typically bicubic downsampling) on high-resolution (HR) images to synthesize low-resolution (LR) counterparts. However, few works on degradation modelling have taken the physical aspects of the optical imaging system into consideration. In this paper, we analyze the imaging system optically and exploit the characteristics of the real-world LR-HR pairs in the spatial frequency domain. We formulate a real-world physics-inspired degradation model by considering both optics and sensor degradation;The physical degradation of an imaging system is modelled as a low-pass filter, whose cut-off frequency is dictated by the object distance, the focal length of the lens, and the pixel size of the image sensor. In particular, we propose to use a convolutional neural network (CNN) to learn the cutoff frequency of real-world degradation process. The learned network is then applied to synthesize LR images from unpaired HR images. The synthetic HR-LR image pairs are later used to train an SISR network. We evaluate the effectiveness and generalization capability of the proposed degradation model on real-world images captured by different imaging systems. Experimental results showcase that the SISR network trained by using our synthetic data performs favorably against the network using the traditional degradation model. Moreover, our results are comparable to that obtained by the same network trained by using real-world LR-HR pairs, which are challenging to obtain in real scenes. © 2021, CC BY.

关键词： Cutoff frequency

来源：评论

学校读者我要写书评

暂无评论

Estimating Available Bandwidth Using Overloading Stream with Variable Packet Size 3

Estimating Available Bandwidth Using Overloading Stream with...

引用

3rd International Conference for Convergence in Technology, I2CT 2018

作者： Yu, Cao Liang, Qian Lianghui, Ding Feng, Yang Institute of Wireless Communication Technology Shanghai Jiao Tong University Science and Technology on Communication Networks Laboratory Shanghai China Shanghai Jiao Tong University Institute of Wireless Communication Technology Shanghai China Shanghai Jiao Tong University Institute of Image Communication and Network Engineering Shanghai China

ISBN: (纸本)9781538642733

End to end available bandwidth estimation is required for areas such as bandwidth dependent applications, quality of service management and traffic engineering. Available bandwidth estimation methods have been proposed quite a lot past years, e.g. Pathload and Pathchirp. These methods focus on observation of delay and finding a transition point to estimate available bandwidth, resulting in heavy overhead loads on the network, sometimes loss of accuracy. In this paper, we propose a new method based on a statistical model for network path to estimate available bandwidth. This method estimates available bandwidth using an overload stream with variable packet size(linear increased packet size), requiring no prior knowledge of network path. Evaluation results show that our method performs better in terms of intrusiveness, estimation duration and accuracy © 2018 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

A loose-strategy-based complexity scalable motion estimation method for video coding

A loose-strategy-based complexity scalable motion estimation...

引用

2010 3rd International Congress on image and Signal Processing, CISP 2010

作者： Lin, Weiyao Panusopone, Krit Baylon, David M. Sun, Ming-Ting Li, Hongxiang Institute of Image Communication and Information Processing Shanghai Jiao Tong University Shanghai 200240 China Department of Electrical Engineering University of Washington Seattle WA 98105 United States Advanced Technology Department CTO Office Home and Networks Mobility Motorola Inc. San Diego CA 92121 United States Department of Electrical and Computer Engineering North Dakota State University Fargo ND 58108 United States

ISBN: (纸本)9781424465149

In this paper, a new Complexity Scalable Motion Estimation (CSME) method is proposed which can perform Motion Estimation adaptively under different computation or power budgets while keeping high coding performance. We first propose to use a new class-based method to measure the Macroblock (MB) importance. Based on the new MB importance measure, a complete CSME framework is then proposed. The proposed method allocates computation in a one-pass flow and is based on a loose strategy where frames can borrow computation from other frames to achieve better performance under the same computation budget. Experimental results demonstrate that the proposed method can allocate computation more accurately than previous methods. ©2010 IEEE.

关键词： Motion estimation

来源：评论

学校读者我要写书评

暂无评论

A loose-strategy-based Complexity Scalable Motion Estimation method for video coding

A loose-strategy-based Complexity Scalable Motion Estimation...

引用

Congress on image and Signal Processing, CISP

作者： Weiyao Lin Krit Panusopone David M. Baylon Ming-Ting Sun Hongxiang Li Institute of Image Communication and Information Processing Shanghai Jiaotong University Shanghai China Advanced Technology Department CTO Office Home & Networks Mobility Motorola Inc. San Diego CA USA Department of Electrical Engineering University of Washington Seattle WA USA Department of Electrical and Computer Engineering North Dakota State University Fargo ND USA

关键词： Encoding Resource management Computer integrated manufacturing Motion estimation Video coding Algorithm design and analysis Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

Windows computer laboratory as a tutorial for undergraduate electromagnetics

引用

Computer Applications in engineering Education 1995年第3期3卷 183-183页

作者： McCulloch, J. Keddy, W.A. Stuchly, M.A. University of Victoria Department of Electrical and Computer Engineering University of Victoria Victoria British Columbia Canada Janet McCulloch:is a native of Toronto Canada where she briefly attended both York University and the University of Toronto before she set off on a year-long adventure through the Middle East and North Africa. When she returned to Canada she relocated to California. She spent a number of years living both in Los Angeles and coastal Northern California interspersed with several extended trips through the United States Mexico and Central America. Ms. McCulloch moved to Victoria British Columbia in 1989 and a year later registered in the cooperative education program in Electrical Engineering at the University of Victoria. She is presently in the fifth and final year of the program and has her sights set on graduate school. Besides the work described in this article she has been very much involved in the use of technology to assist in the delivery of curriculum at the senior secondary level. She lives on a country property a half-hour from the University with her husband her horse and three cats. W. Alfred Keddy:is an applications programmer/analyst for the Department of Electrical and Computer Engineering at the University of Victoria. He received a BEng (1988) in Electrical Engineering and an MASc (1994) from the University of Victoria. His research interests include distributed systems image processing local area networks VLSI and DSP system design for computation and communication and human–computer interface design. Maria A. Stuchly:received an MSc in Electrical Engineering in 1962 from Warsaw Technical University and a PhD in Electrical Engineering from the Polish Academy of Sciences in 1970. Between the years of 1962 and 1970 she was with the Warsaw Technical University an Institute of the Polish Academy of Sciences. After immigrating to Canada during 1970 her first position was with the University of Manitoba. In 1976 she became employed by the Bureau of Radi

Contents and implementation of a computer laboratory for undergraduate electromagnetics are described. The laboratory consists of four 3-hour sessions covering vector calculus, Maxwell's equations (integral and differential forms), wave propagation in materials, and wave behavior at planar interfaces. Each session contains theory (in the Help file), animations (where relevant) and a quiz. The program runs on IBM compatible 486-based PCs in a Windows environment and uses the Borland C ++ 4.0 compiler.

关键词：

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：