检索结果-内蒙古大学图书馆

10th AAAI Conference on Human Computation and Crowdsourcing, HCOMP 2022

作者： Yang, Yi Zheng, Yueyuan Deng, Didan Zhang, Jindi Huang, Yongxiang Yang, Yumeng Hsiao, Janet H. Cao, Caleb Chen Huawei Research Hong Kong Hong Kong Department of Psychology University of Hong Kong Hong Kong Department of Electronic and Computer Engineering Hong Kong University of Science and Technology Hong Kong State Key Laboratory of Brain and Cognitive Sciences University of Hong Kong Hong Kong Institute of Data Science University of Hong Kong Hong Kong

ISBN: (纸本)9781577358787

Model explanations are generated by XAI (explainable AI)methods to help people understand and interpret machine learning models. To study XAI methods from the human perspective, we propose a human-based benchmark dataset, i.e., human saliency benchmark (HSB), for evaluating saliency based XAI methods. Different from existing human saliency annotations where class-related features are manually and subjectively labeled, this benchmark collects more objective human attention on vision information with a precise eyetracking device and a novel crowdsourcing experiment. Taking the labor cost of human experiment into consideration, we further explore the potential of utilizing a prediction model trained on HSB to mimic saliency annotating by humans. Hence, a dense prediction problem is formulated, and we propose an encoder-decoder architecture which combines multimodal and multi-scale features to produce the human saliency maps. Accordingly, a pretraining-finetuning method is designed to address the model training problem. Finally, we arrive at a model trained on HSB named human saliency imitator(HSI).We show, through an extensive evaluation, that HIS can successfully predict human saliency on our HSB dataset, and the HSI-generated human saliency dataset on Image Netshow cases the ability of benchmarking XAI methods both qualitatively and quantitatively. © 2022, Association for the Advancement of Artificial Intelligence.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Tunnel Disease Detection Based on Spark and Deep Learning 8th

Tunnel Disease Detection Based on Spark and Deep Learning

引用

8th International Conference on Artificial Intelligence and Security , ICAIS 2022

作者： Zhao, Xia Zheng, Ying Yu, Chongchong You, Xiangjun Zhao, Lu China Light Industry Key Laboratory of Industrial Internet and Big Data Beijing Technology and Business University Beijing100048 China School of Artificial Intelligence Beijing Technology and Business University Beijing100048 China Zhejiang HuaZhan Institute of Engineering Research and Design Ningbo315012 China Secregen Technology San MateoCA94402 United States

ISBN: (纸本)9783031067662

Aiming at metro shield disease detection with LiDAR scanning data, this paper proposes an approach of distributed parallel point cloud data processing and shield disease detection. The approach is based on HDFS and Spark, and converts the LiDAR point cloud data into the orthographic projection space, and generates projected image data efficiently. Based on the optimized deep neural network YOLOv5 target detection model, our approach detects four types of metro shield diseases and labels their locations and the classifications on the orthographic projection image. The experiments show that our distributed parallel process framework archives the speedup ratio of 4.96 to the approach of a single server on the LiDAR data processing. Using the optimized YOLO v5 model, with the appropriate IOU threshold, the accuracy of the water seepage and spalling detection box is higher than 90%, and that of the cracks and the cracks on joints is higher than 80%. Our approach can improve the efficiency and accuracy of metro shield disease detection significantly, and promote the intelligence of metro operation and maintenance in the rail transit industry. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Optical radar

来源：评论

学校读者我要写书评

暂无评论

Multi-parameter optical fibre sensors based on W-type leaky fibre grating 13

Multi-parameter optical fibre sensors based on W-type leaky ...

引用

Advanced Sensor Systems and Applications XIII 2023

作者： Luo, Yanhua Yan, Binbin Xie, Junjie Qi, Haifeng Yang, Liwei Wang, Kuiru Yu, Chongxiu Sang, Xinzhu Yuan, Jinhui Ni, Jiasheng Peng, Gang-Ding Key Laboratory of Specialty Fiber Optics and Optical Access Networks Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication Shanghai Institute for Advanced Communication and Data Science Shanghai University Shanghai200444 China State Key Laboratory of Information Photonics and Optical Communications Beijing University of Posts and Telecommunications Beijing100876 China Institute of Technology and Standards China Academy of Information and Communication Technology Beijing100191 China Laser Institute QiLu University of Technology Shandong Academy of Sciences Jinan250014 China College of Information and Electrical Engineering China Agricultural University Beijing100083 China Photonics and Optical Communications Group University of New South Wales SydneyNSW2052 Australia

ISBN: (纸本)9781510667914

A novel multi-parameter sensor has been proposed based on the W-type leaky fibre grating, after the systematic investigation of its characteristics in response to the variation of temperature, surrounding refractive index and bend. The resonant wavelength for different mode in the reflection spectra has the same temperature sensitivity of ~9.13 pm/oC, but the reflection intensity has no dependence upon the temperature change. The resonant wavelength of different mode linearly varies a little with the refractive index change, however, its reflection intensity follows the negative exponential decay with refractive index increasing, where for the higher modes the sensitivity is higher. The resonant wavelength for each mode almost linearly responds to the displacement, giving out a displacement sensitivity of 99.94 pm/mm. However, the reflection response to the displacement varies a lot. Based on these results, simple and compact multi-parameter fibre sensors are proposed based on W fibre grating. © 2023 SPIE.

关键词： Refractive index

来源：评论

学校读者我要写书评

暂无评论

Towards chemical accuracy using a multi-mesh adaptive finite element method in all-electron density functional theory

arXiv

引用

arXiv 2023年

作者： Kuang, Yang Shen, Yedan Hu, Guanghui Guangdong University of Technology China School of Mathematics and Information Science Guangzhou University China Hunan Key Laboratory for Computation and Simulation in Science and Engineering Xiangtan University China Department of Mathematics Guangdong-Hong Kong-Macao Joint Laboratory for Data-Driven Fluid Mechanics and Engineering Applications University of Macau China Zhuhai UM Science and Technology Research Institute Zhuhai China

Chemical accuracy serves as an important metric for assessing the effectiveness of the numerical method in Kohn-Sham density functional theory. It is found that to achieve chemical accuracy, not only the Kohn-Sham wavefunctions but also the Hartree potential, should be approximated accurately. Under the adaptive finite element framework, this can be implemented by constructing the a posteriori error indicator based on approximations of the aforementioned two quantities. However, this way results in a large amount of computational cost. To reduce the computational cost, we propose a novel multi-mesh adaptive method, in which the Kohn-Sham equation and the Poisson equation are solved in two different meshes on the same computational domain, respectively. With the proposed method, chemical accuracy can be achieved with less computational consumption compared with the adaptive method on a single mesh, as demonstrated in a number of numerical experiments. © 2023, CC BY.

关键词： Poisson equation

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Based Weed Detection and Target Spraying System At Seedling Stage of Cotton Field

SSRN

引用

SSRN 2023年

作者： Fan, Xiangpeng Chai, Xiujuan Zhou, Jianping Sun, Tan Agricultural Information Institute Chinese Academy of Agricultural Sciences Beijing100081 China Key Laboratory of Agricultural Big Data Ministry of Agriculture and Rural Affairs Beijing100081 China School of Mechanical Engineering Xinjiang University Urumqi830017 China

The precision spraying robot dispensing herbicides only on unwanted plants based on machine vision detection is the most appropriate approach to ensure the sustainable agro-ecosystem and the minimum impact of nuisance weeds. However, the coexistence of crops and a variety of weeds, similar targets and uneven weed distribution makes reliable weed detection difficult, leading to serious limitations in the application of deep learning method to target spraying in the field environment. In this paper, 4694 representative images are acquired from cotton field scenario as the data basis for deep learning model. A novel weed detection model is constructed by employing CAM module, BiFPN structure and Bilinear interpolation algorithm. The proposed network can effectively learn the deep information and distinguish cotton seedlings from weeds in various complicated growth states. Evaluation experiments on our constructed dataset indicate that the proposed method reaches an mAP of 98.43% with faster inference speed. Our proposed weed detection model is also deployed in the spraying robot, and field trials are conducted for detection and spraying, which could maintain the excellent performance with mAP of 97.42% and effective spraying rate of 98.93%. The ability to successfully execute the weed detection and herbicide spraying management in the field lays foundation for targeted spraying in precision weed control, which has an excellent impact on cotton cultivation and growth. © 2023, The Authors. All rights reserved.

关键词： Cotton

来源：评论

学校读者我要写书评

暂无评论

Ponzi Scheme Detection in Smart Contracts Using Heterogeneous Semantic Graph 6th

Ponzi Scheme Detection in Smart Contracts Using Heterogeneo...

引用

6th International Conference on Blockchain and Trustworthy Systems, BlockSys 2024

作者： Chen, Wei Jiang, Xinjun Lan, Tian Liu, Leyuan Li, Chengyu Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of China Chengdu610000 China School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu610000 China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu610000 China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen518000 China

ISBN: (纸本)9789819614103

Behind the booming decentralized finance (DeFi) ecosystem driven by blockchain technology, various financial risks lurking, including money laundering, gambling, Ponzi schemes, and phishing. Due to the decentralization and anonymity of Ethereum, Ponzi schemes can be easily deployed, causing huge economic losses to investors. Existing detection methods based on transaction data are difficult to provide early risk warnings, while detection methods based on smart contract source code and opcodes have insufficient feature fusion at multiple levels. We propose an approach for constructing a multi-level Heterogeneous Semantic Graph (HSG) of smart contracts, and improve the HAN model to detect Ponzi scheme smart contracts based on the Heterogeneous Semantic Graph. The experimental results demonstrate the effectiveness of our approach, achieving an accuracy of 97.21%, a precision of 94.29%, and an F1 score of 90.41%. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

Multistep Traffic Speed Prediction: A Sequence-to-Sequence Spatio-Temporal Attention Model

SSRN

引用

SSRN 2023年

作者： Yang, Di Wang, Peng Yuan, Lihong Li, Hong Changchun University of Science and Technology. Changchun130022 China Jilin Provincial Joint Key Laboratory of Big Data Science and Engineering Trivandrum Changchun130022 China Chongqing Institute of Changchun University of Science and Technology Chongqing401120 China

Multistep traffic speed prediction plays a crucial role in alleviating road congestion and improving transport efficiency. In actual traffic networks, the spatio-temporal dependence among roads dynamically changes over time concerning factors such as road conditions and unforeseen incidents bringing great challenges to multistep traffic speed prediction. Additionally, multistep traffic speed prediction commonly faces the problem of error accumulation, resulting in a loss of prediction accuracy. To address these issues, we propose a Sequence-to-Sequence Spatio-Temporal Attention model (STSSTA) for multistep traffic speed prediction. Specifically, we build an adaptive tuning module to select road preferences and automatically obtain global information about the roads. Then we construct a Sequence-to-Sequence architecture for spatio-temporal feature learning and multistep traffic speed prediction. In particular, in the encoder, we design a Diffusion Graph Convolutional Network (DGCN) and combine it with a Gated Recurrent Unit (GRU) to effectively capture the complex spatio-temporal feature within the traffic networks. In the decoder, we introduce a Recalling attention mechanism to alleviate the problem of local information loss caused by encoder compression, thereby reducing the error accumulation in multistep prediction. Experiments on METR-LA and PeMS-BAY datasets demonstrate that STSSTA outperforms the baseline models in long-term prediction. © 2023, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Communication Encryption Scheme of 5G Power Trading Private Network Based on Dynamic key

Communication Encryption Scheme of 5G Power Trading Private ...

引用

Electrical, Power and Computer engineering (EPCE), Asia Conference on

作者： Rui Zhou Yiyang Song Ying Xu Electrical Engineering and Automation School Harbin Institute of Technology Harbin China Information Data Department Guangdong Power Exchange Center Guangzhou China Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Beijing China

Security has been the primary concern for the 5G power trading private network, which involves a large amount of power trading information. The communication must be encrypted to prevent the leakage of important information. The communication encryption scheme for 5G power trading private network should meet the characteristics of small communication overhead, low storage space requirement and strong resistance to attack. Based on these characteristics, the dynamic key-based encryption scheme is designed. To ensure the security and reduce the computation overhead, the optimized symmetric encryption algorithm is used to update the shared key. And the dynamic key generation method is improved by adding random numbers to solve the problem that the correct transmission of the data link does not have true randomness. The encryption mode used can generate dynamic keys directly and synchronously at both ends of the communication, reducing the overhead of key exchange and eliminating the need for trusted third parties. The algorithm is also analyzed for security and efficiency and compared with other algorithms. The experimental results show that the improved algorithm enhances security with guaranteed efficiency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Rethinking the Evaluation for Conversational Recommendation in the Era of Large Language Models

arXiv

引用

arXiv 2023年

作者： Wang, Xiaolei Tang, Xinyu Zhao, Wayne Xin Wang, Jingyuan Wen, Ji-Rong Gaoling School of Artificial Intelligence Renmin University of China China School of Information Renmin University of China China Beijing Key Laboratory of Big Data Management and Analysis Methods China School of Computer Science and Engineering Beihang University China

The recent success of large language models (LLMs) has shown great potential to develop more powerful conversational recommender systems (CRSs), which rely on natural language conversations to satisfy user needs. In this paper, we embark on an investigation into the utilization of ChatGPT for CRSs, revealing the inadequacy of the existing evaluation protocol. It might overemphasize the matching with ground-truth items annotated by humans while neglecting the interactive nature of CRSs. To overcome the limitation, we further propose an interactive Evaluation approach based on LLMs, named iEvaLM, which harnesses LLM-based user simulators. Our evaluation approach can simulate various system-user interaction scenarios. Through the experiments on two public CRS datasets, we demonstrate notable improvements compared to the prevailing evaluation protocol. Furthermore, we emphasize the evaluation of explainability, and ChatGPT showcases persuasive explanation generation for its recommendations. Our study contributes to a deeper comprehension of the untapped potential of LLMs for CRSs and provides a more flexible and realistic evaluation approach for future research about LLM-based CRSs. The code is available at https://***/RUCAIBox/iEvaLM-CRS. Copyright © 2023, The Authors. All rights reserved.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Memory-Efficient Multi-Scale Residual Dense Network for Single Image Rain Removal

SSRN

引用

SSRN 2023年

作者： Zheng, Ziyang Chen, Zhixiang Wang, Shuqi Wang, Wenpeng Wang, Hui College of Physics and Information Engineering Minnan Normal University Zhangzhou363000 China College of Computer Minnan Normal University Zhangzhou363000 China Key Laboratory of Data Science and Intelligence Application Minnan Normal University Zhangzhou363000 China

Outdoor photography is greatly affected by bad rainfall, which degrades the quality of captured images and hinders practical application. In response to the above problems, we propose a memory-efficient multi-scale residual dense network (MMRD-Net) for single image rain removal. Firstly, we use a combination or fusion of extended convolution blocks of different sizes to accurately extract multi-scale rain streak features while retaining rich local detail. Secondly, we add attention modules to the residual block, which performs adaptive operations on rainfall and non-rainfall areas to extract more and more accurate image features and rain streak details. During this period, we opened up a space for shared use by all the concatenation layers in the network, which reduces the memory consumption of the network. It allows for better aggregation of rain streak features across multiple layers and also allows the network to converge faster. Finally, we tested on both synthetic and real datasets to obtain subjective results and objective evaluations. The experimental results show that the robustness of our algorithm is better for the task of rain image deraining for different densities of rain streaks. Our algorithm outperforms twelve other different classical algorithms in terms of its ability to remove rain streaks. It can produce visually sharper deraining images and the enhanced visibility of the images is effective for computer vision applications (Google Vision API). © 2023, The Authors. All rights reserved.

关键词： Rain

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：