检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Dong, Wei Yan, Dawei Lin, Zhijun Wang, Peng College of Information and Control Engineering Xi’an University of Architecture and Technology China School of Computer Science Northwestern Polytechnical University China School of Computer Science and Engineering University of Electronic Science and Technology of China China Xi’an Hypersonic Measurement Technology Co. Ltd. China

The advent of high-capacity pre-trained models has revolutionized problem-solving in computer vision, shifting the focus from training task-specific models to adapting pre-trained models. Consequently, effectively adapting large pre-trained models to downstream tasks in an efficient manner has become a prominent research area. Existing solutions primarily concentrate on designing lightweight adapters and their interaction with pre-trained models, with the goal of minimizing the number of parameters requiring updates. In this study, we propose a novel Adapter ReComposing (ARC) strategy that addresses efficient pre-trained model adaptation from a fresh perspective. Our approach considers the reusability of adaptation parameters and introduces a parameter-sharing scheme. Specifically, we leverage symmetric down-/up-projections to construct bottleneck operations, which are shared across layers. By learning low-dimensional re-scaling coefficients, we can effectively re-compose layer-adaptive adapters. This parameter-sharing strategy in adapter design allows us to further reduce the number of new parameters while maintaining satisfactory performance, thereby offering a promising approach to compress the adaptation cost. We conduct experiments on 24 downstream image classification tasks using various Vision Transformer variants to evaluate our method. The results demonstrate that our approach achieves compelling transfer learning performance with a reduced parameter count. Our code is available at https://***/DavidYanAnDe/ARC. Copyright © 2023, The Authors. All rights reserved.

关键词： Reusability

来源：评论

学校读者我要写书评

暂无评论

Vulnerability Assessment of Urban Road Network in Face of Ride-Hailing Malicious Dispatch Attack: A Case Study of Nanjing

SSRN

引用

SSRN 2023年

作者： Zhao, Na Tian, Yongshang Lu, Yuwen Zhai, Guofang School of Architecture and Urban Planning Nanjing University 210093 China Department of Computer Science and Technology Nanjing University 210023 China

As a typical ICT-based transportation infrastructure in smart cities, ride-sharing platforms significantly enhance the efficiency of urban public transportation. However, they face severe cybersecurity risks. Cities like Moscow have experienced incidents where hackers infiltrated ride-sharing platforms to maliciously dispatch vehicles, leading to extensive traffic congestion. Existing literature has paid much attention to road network vulnerability against natural disasters. Yet, less focus has been put on disturbances due to such cyber attacks. In this study, we assess the vulnerability of the downtown road network in Nanjing during the morning peak hours when subjected to ride-hailing malicious dispatch attacks. We begin by adopting the perspective of attackers, leveraging relevant theories from graph theory to devise an optimized malicious vehicle dispatch scheme under the real-time road conditions. Subsequently, we conduct simulated attacks against the target network in five major commuting directions. Results show that this attack can significantly decrease the road traffic capacity in some critical directions with just a dispatch of tens of vehicles. Additionally, there is a noticeable disparity in vulnerability across different directions. In response, we further identify the primary factors contributing to these variations, which aims to enhance the resilience of urban road network in the information age. © 2023, The Authors. All rights reserved.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

DACB-NET: DUAL ATTENTION GUIDED COMPACT BILINEAR CONVOLUTION NEURAL NETWORK FOR SKIN DISEASE CLASSIFICATION

arXiv

引用

arXiv 2024年

作者： Ahmad, Belal Usama, Mohd Ahmad, Tanvir Saeed, Adnan Khatoon, Shabnam Chen, Min School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Departments of Diagnostics and Intervention and Biomedical Engineering Umea University Sweden School of Information and Control Engineering Xi'an University of Architecture and Technology Xian China School of Civil and Hydraulic Engineering Huazhong University of Science and Technology Wuhan China School of Management Science and Engineering China University of Geosciences Wuhan China

This paper introduces the three-branch Dual Attention-Guided Compact Bilinear CNN (DACBNet) by focusing on learning from disease-specific regions to enhance accuracy and alignment. A global branch compensates for lost discriminative features, generating Attention Heat Maps (AHM) for relevant cropped regions. Finally, the last pooling layers of global and local branches are concatenated for fine-tuning, which offers a comprehensive solution to the challenges posed by skin disease diagnosis. Although current CNNs employ Stochastic Gradient Descent (SGD) for discriminative feature learning, using distinct pairs of local image patches to compute gradients and incorporating a modulation factor in the loss for focusing on complex data during training. However, this approach can lead to dataset imbalance, weight adjustments, and vulnerability to overfitting. The proposed solution combines two supervision branches and a novel loss function to address these issues, enhancing performance and interpretability. The framework integrates data augmentation, transfer learning, and fine-tuning to tackle data imbalance to improve classification performance, and reduce computational costs. Simulations on the HAM10000 and ISIC2019 datasets demonstrate the effectiveness of this approach, showcasing a 2.59% increase in accuracy compared to the state-of-the-art. © 2024, CC BY.

关键词： Gradient methods

来源：评论

学校读者我要写书评

暂无评论

Content and Relation Fuzzy Mitigation Framework for Intent Perception

引用

IEEE Transactions on Fuzzy Systems 2024年

作者： Zhang, Wenxin Qi, Hao Wang, Shutong Lin, Ziqi Wang, Binglu Northwestern Polytechnical University School of Computer Science Xi'an710072 China Nankai University College of Economic and Social Development Tianjin300071 China Xi'an University of Architecture and Technology Xi'an710311 China Northwestern Polytechnical University School of Astronautics China

In this paper, we tackle the content fuzzy and relation fuzzy in image-based intent perception. Current research primarily focuses on additional modeling mechanisms and multi-modal information, but it is often difficult to deal with the relation fuzzy under inaccurate annotation in Intent Perception. We exploit the powerful representation capabilities of large language models and innovatively use them to solve the image-based intent perception task. Comprising two innovative modules: Multi-Dimension Adapter and Fuzzy Harmony Refiner. Multi-Dimension Adapter projects Intent image features of different difficulty levels to feature dimensions of different sizes to alleviate content fuzzy of different difficulties caused by image diversity. Besides, intent category labeling is somewhat subjective, and some categories are more likely to appear collaboratively. We design Fuzzy Harmony Refiner to mine the relationship between various types to enhance perception results. Different from traditional expert-specified rules, we learn the co-occurrence frequencies between different intent categories from the training data, generate a Fuzzy Harmony Matrix for correcting the score, and alleviate the relationship fuzzy in intent perception. We achieved new state-of-the-art performance on the Intentonomy dataset, with Macro F1 score of 42.52%, Micro F1 score of 54.80%, Samples F1 score of 57.57%, and Average F1 score of 51.63%. Compared with the current existing methods, our method improves by 9.8%. © 1993-2012 IEEE.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Research on automatic cascade dispatching system of whole ship cold and heat energy for LNG-IHES

Research on automatic cascade dispatching system of whole sh...

引用

2022 IEEE International Conference on Advances in Electrical Engineering and computer Applications, AEECA 2022

作者： Lei, Shaomin Zhang, Jufu Shao, Xingdong Xu, Zhaoyang Li, Chuantao Zhang, Zunhua Wuhan University of Technology School of Naval Architecture Ocean and Energy Power Engineering Hubei Wuhan430063 China Wuhan University of Technology School of Computer Science and Artificial Intelligence Hubei Wuhan430063 China Wuhan University of Technology School of Information Engineering Hubei Wuhan430063 China

ISBN: (数字)9781665480901

ISBN: (纸本)9781665480901

In order to solve the problem of large waste of high-quality cold and heat energy and low energy utilization rate that may occur in the future LNG-Inland Hybrid Electric Ship (LNG-IHES), this paper designs a set of LNG-IHE-Soriented The whole ship's cooling and heating cascade dispatching system. In this paper, through a comprehensive analysis of the cold and heat energy modules in different temperature ranges on LNG-IHES, based on the principle of 'corresponding temperature, cascade utilization', water, steam and Phase change material are used to collect the cold and heat energy of different temperature gradients generated on the ship to construct a multi-load The advanced cold and heat energy cascade dispatching system realizes the high-quality cold and heat energy cascade management and reuse. The system divides the cold and heat energy into five steps according to the difference of the carrier temperature after the cold and heat energy is collected. The system itself can be divided into three major sections, namely, the hot and cold energy distribution center, the self-induction control integrated system, and the human-computer interaction system. The system uses the cold and heat energy distribution center to collect, store, manage and deploy the cold and heat energy in different temperature ranges on the ship in a centralized manner;uses the self-induction control integrated system to monitor, control and dominate the cold and heat energy on board;uses human-computer interaction system realizes the visualization of important control and equipment temperature, the situation of the distribution center, and the proportion of pipeline flow. After economic benefit analysis and calculation, if all the cold and heat energy recovered by this system is used for power generation, the annual net income will be 1,155,600 yuan. Considering the initial investment and operating costs of each equipment, the investment payback period is half a year. This system has very import

关键词： Investments

来源：评论

学校读者我要写书评

暂无评论

Distilling object detectors with feature richness 21

Distilling object detectors with feature richness

引用

Proceedings of the 35th International Conference on Neural Information Processing Systems

作者： Zhixing Du Rui Zhang Ming Chang Xishan Zhang Shaoli Liu Tianshi Chen Yunji Chen University of Science and Technology of China and SKL of Computer Architecture Institute of Computing Technology CAS Beijing China and Cambricon Technologies China SKL of Computer Architecture Institute of Computing Technology CAS Beijing China and Cambricon Technologies China Cambricon Technologies China SKL of Computer Architecture Institute of Computing Technology CAS Beijing China and University of Chinese Academy of Sciences China

ISBN: (纸本)9781713845393

In recent years, large-scale deep models have achieved great success, but the huge computational complexity and massive storage requirements make it a great challenge to deploy them in resource-limited devices. As a model compression and acceleration method, knowledge distillation effectively improves the performance of small models by transferring the dark knowledge from the teacher detector. However, most of the existing distillation-based detection methods mainly imitating features near bounding boxes, which suffer from two limitations. First, they ignore the beneficial features outside the bounding boxes. Second, these methods imitate some features which are mistakenly regarded as the background by the teacher detector. To address the above issues, we propose a novel Feature-Richness Score (FRS) method to choose important features that improve generalized detectability during distilling. The proposed method effectively retrieves the important features outside the bounding boxes and removes the detrimental features within the bounding boxes. Extensive experiments show that our methods achieve excellent performance on both anchor-based and anchor-free detectors. For example, RetinaNet with ResNet-50 achieves 39.7% in mAP on the COCO2017 dataset, which even surpasses the ResNet-101 based teacher detector 38.9% by 0.8%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

TAPEMAN: Towards an Optimal Data Gathering Mechanism in Wireless Sensor Networks

引用

Chinese Journal of Electronics 2023年第4期19卷 594-598页

作者： Yu Gu Yusheng Ji Hongyang Chen Baohua Zhao Department of Computer Science University of Science and Technology of China Hefei China State Key Laboratory of Networking and Switching Technology Beijing China Information Systems Architecture Science Research Division NII Tokyo Japan Institute of Industrial Science University of Tokyo Tokyo Japan

In this paper, we study the lifetime optimization problem using a mobile sink node in a storage-constrained wireless sensor network, and propose an optimal data gathering mechanism named (TAPEMAN) which runs in three steps: first a tight upper bound of the network lifetime is derived through analysis of energy consumption, then we generate a Traveling salesman problem (TSP) solution based on a 2-approximation $O(n^{2})$ algorithm. A judgment will be made about whether this solution is optimal. If the answer is yes, TAPEMAN terminates. If not, a novel data diffusion scheme is used to distribute data to neighboring nodes, in order to avoid data leaking. We prove that under some reasonable assumptions, our algorithm can achieve this upperbound. Simulation results demonstrate the efficiency of our proposed solution and substantiate the importance of using sink mobility for energy-constrained sensor networks.

关键词： Wireless communication Wireless sensor networks Energy consumption Upper bound Simulation Traveling salesman problems Delays

来源：评论

学校读者我要写书评

暂无评论

Optimal Design of Heliostat Field Using Random Search Algorithm

Optimal Design of Heliostat Field Using Random Search Algori...

引用

IEEE International Conference on Power, Intelligent Computing and Systems (ICPICS)

作者： Chenglin Mou Kaiqi Wu Pengcheng Du Kaibo Yang Department of Engineering Mechanics School of Civil Engineering and Architecture Henan University of Science and Technology Luoyang China Department of Computer Science and Technology School of Information Engineering Henan University of Science and Technology Luoyang China Department of Automation School of Information Engineering Henan University of Science and Technology Luoyang China Department of Civil Engineering School of International Education Henan University of Science and Technology Luoyang China

ISBN: (数字)9798350374315

ISBN: (纸本)9798350374322

Tower solar thermal power generation is a new type of low-carbon and environmentally friendly clean energy technology. In this paper, a single-objective optimization model is established with the maximum annual average thermal output per unit mirror area of the heliostat field as the objective function. By optimizing the heliostat width, heliostat height and heliostat coordinate, the annual average output thermal power can be maximized while meeting the rated power. The heliostat field is distributed in a circular shape. As the seasons change, the light intensity received by the heliostat field is uneven, so the optical efficiency of the heliostats is also different. We simplify the heliostat field into S area (area with higher optical efficiency) and W area (area with lower optical efficiency) based on the optical efficiency distribution of heliostats. At different locations, According to the heliostat width, heliostat height and heliostat coordinate generated by random functions, substituting these parameters into the efficiency model in this paper, and the random search algorithm is used to traverse the maximum annual output thermal power and the corresponding heliostat parameters. In the shadow occlusion loss model, we have grid to set up the heliostat and use the coordinate conversion to solve the shadow occlusion loss efficiency.

关键词： Integrated optics Green energy Shape Optimization models Computational modeling Poles and towers Linear programming Mirrors Power generation

来源：评论

学校读者我要写书评

暂无评论

Using Mobile Sinks in WSN: Computational Complexity and a Theoretical Bound

引用

Chinese Journal of Electronics 2023年第1期20卷 147-150页

In this paper, we study the lifetime optimization problem in wireless sensor networks using mobile sink nodes. This problem is inherently difficult since we need to consider both sink scheduling and data routing. Through a simple case study we develop a novel notation named the Placement pattern (PP) to bound traffic patterns with candidate locations. This significantly decreases the number of elements needed to be scheduled. Based on the PP, we mathematically formulate this optimization problem as a Mixed-integer non-linear programming (MINLP), which is very tough and time consuming to solve. By proving that the problem is NP-complete, we point out that instead of seeking an optimal algorithm, heuristic algorithms, especially those with performance guarantee, would be much more desirable to develop. Furthermore, in order to help identify performance gains of heuristic algorithms proposed in the future, we develop a Linear programming (LP) formulation which serves as an upper bound by adopting a reformulation and relaxation technique.

关键词： Wireless sensor networks Upper bound Heuristic algorithms Traffic control Programming Performance gain Routing

来源：评论

学校读者我要写书评

暂无评论

SEALing Neural Network Models in Encrypted Deep Learning Accelerators 21

SEALing Neural Network Models in Encrypted Deep Learning Acc...

引用

Proceedings of the 58th Annual ACM/IEEE Design Automation Conference

作者： Pengfei Zuo Yu Hua Ling Liang Xinfeng Xie Xing Hu Yuan Xie Huazhong University of Science and Technology and University of California Santa Barbara Huazhong University of Science and Technology University of California Santa Barbara SKL of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences

ISBN: (纸本)9781665432740

Deep learning (DL) accelerators suffer from a new security problem, i.e., being vulnerable to physical access based attacks. An adversary can easily obtain the entire neural network (NN) model by physically snooping the memory bus that connects the accelerator chip with DRAM memory. Therefore, memory encryption becomes important for DL accelerators to improve their security. Nevertheless, we observe that traditional memory encryption techniques that have been efficiently used in CPU systems cause significant performance degradation when directly used in DL accelerators, due to the big bandwidth gap between the memory bus and the encryption engine. To address this problem, our paper proposes SEAL, a Secure and Efficient Accelerator scheme for deep Learning to enhance the performance of encrypted DL accelerators by improving the data access bandwidth. Specifically, SEAL leverages a criticality-aware smart encryption scheme that identifies partial data having no impact on the security of NN models and allows them to bypass the encryption engine, thus reducing the amount of data to be encrypted without affecting security. Extensive experimental results demonstrate that, compared with existing memory encryption techniques, SEAL achieves 1.34 -- 1.4× overall performance improvement.

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：