检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Zhang, Yechao Hu, Shengshan Zhang, Leo Yu Shi, Junyu Li, Minghui Liu, Xiaogeng Wan, Wei Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia School of Computer Science and Technology Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Adversarial examples for deep neural networks (DNNs) have been shown to be transferable: examples that successfully fool one white-box surrogate model can also deceive other black-box models with different architectures. Although a bunch of empirical studies have provided guidance on generating highly transferable adversarial examples, many of these findings fail to be well explained and even lead to confusing or inconsistent advice for practical use. In this paper, we take a further step towards understanding adversarial transferability, with a particular focus on surrogate aspects. Starting from the intriguing "little robustness" phenomenon, where models adversarially trained with mildly perturbed adversarial samples can serve as better surrogates for transfer attacks, we attribute it to a trade-off between two dominant factors: model smoothness and gradient similarity. Our research focuses on their joint effects on transferability, rather than demonstrating the separate relationships alone. Through a combination of theoretical and empirical analyses, we hypothesize that the data distribution shift induced by off-manifold samples in adversarial training is the reason that impairs gradient similarity. Building on these insights, we further explore the impacts of prevalent data augmentation and gradient regularization on transferability and analyze how the trade-off manifest in various training methods, thus building a comprehensive blueprint for the regulation mechanisms behind transferability. Finally, we provide a general route for constructing superior surrogates to boost transferability, which optimizes both model smoothness and gradient similarity simultaneously, e.g., the combination of input gradient regularization and sharpness-aware minimization (SAM), validated by extensive experiments. In summary, we call for attention to the united impacts of these two factors for launching effective transfer attacks, rather than optimizing one while ignoring the other,

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Unveiling the Cosmos: a Comprehensive Exploration of Key Factors Shaping Pulsar Candidates Using HTRU2 dataset Analysis and Predictions

Unveiling the Cosmos: a Comprehensive Exploration of Key Fac...

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Jianqiong Huang Wenlong Guo Waqar Khan Yajun Xie Yuxin Wan School of Big Data Fuzhou University of International Studies and Trade Fuzhou China College of Electronics and Information Science Fujian Jiangxia University Fuzhou China

ISBN: (数字)9798350356670

ISBN: (纸本)9798350356687

Following the completion of the 500-metre aperture spherical radio telescope (FAST) and the launch of the 19-beam receiver survey project, the number of candidates obtained from pulsar searches has exhibited a notable increase, resulting in a significant expansion in the number of observable celestial objects and astronomical phenomena. It has furnished a substantial corpus of data for astronomical research. However, the vast quantity of pulsar candidate data presents a considerable challenge in identifying genuine candidates, due to the presence of numerous interfering signals. (1) The search for pulsar candidates is a complex and time-consuming process, necessitating the acquisition and analysis of a substantial amount of observational data; (2) The discovery of each new pulsar contributes to our understanding of extreme physical processes in the Universe. It is therefore necessary to search for new pulsar candidates on an ongoing basis. This paper examines the key factors influencing the identification of pulsar candidates, with a view to developing an efficient screening process. The High Time Resolution Universe Survey 2 (HTRU2) dataset from the University of California, Irvine (UCI or UC Irvine) platform was employed for data analysis and the construction of predictive analysis models. These were evaluated using classification metrics, including precision, recall, F1-score, and ROC-AUC curves. The latter were constructed using decision trees (DT), random forest (RF) and alpha-investing algorithms. The experimental results demonstrate that the three most critical factors affecting the pulsar candidates are the mean value of the integrated profile (Mean_IP), the excess kurtosis of the integrated profile (Excess_kurtosis_IP), and the skewness of the integrated profile (Skewness_IP).

关键词： Surveys Measurement Software algorithms Collaboration Prediction algorithms Kurtosis Classification algorithms Decision trees Random forests Physics

来源：评论

学校读者我要写书评

暂无评论

ADDRESSING CLASS VARIABLE IMBALANCE IN FEDERATED SEMI-SUPERVISED LEARNING

arXiv

引用

arXiv 2023年

作者： Dong, Zehui Liu, Wenjing Liu, Siyuan Chen, Xingzhi School of Data Science and Application Inner Mongolia University of Technology Hohhot China

Federated Semi-supervised Learning (FSSL) combines techniques from both fields of federated and semi-supervised learning to improve the accuracy and performance of models in a distributed environment by using a small fraction of labeled data and a large amount of unlabeled data. Without the need to centralize all data in one place for training, it collect updates of model training after devices train models at local, and thus can protect the privacy of user data. However, during the federal training process, some of the devices fail to collect enough data for local training, while new devices will be included to the group training. This leads to an unbalanced global data distribution and thus affect the performance of the global model training. Most of the current research is focusing on class imbalance with a fixed number of classes, while little attention is paid to data imbalance with a variable number of classes. Therefore, in this paper, we propose Federated Semi-supervised Learning for Class Variable Imbalance (FCVI) to solve class variable imbalance. The class-variable learning algorithm is used to mitigate the data imbalance due to changes of the number of classes. Our scheme is proved to be significantly better than baseline methods, while maintaining client privacy. © 2023, CC BY.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

DC-YOLOv3: A novel efficient object detection algorithm

DC-YOLOv3: A novel efficient object detection algorithm

引用

2021 Guangdong-Hong Kong-Macao Greater Bay Area Artificial Intelligence and big data Forum, AIBDF 2021

作者： Zhang, Xu Han, Fang Wang, Ping Jiang, Wei Wang, Chen Science and Education Center of Big Data and Computer Application Huanghe University of Science and Technology Henan Province Zhengzhou450000 China

Feature pyramids have become an essential component in most modern object detectors, such as Mask RCNN, YOLOv3, RetinaNet. In these detectors, the pyramidal feature representations are commonly used which represent an image with multi-scale feature layers. However, the detectors can't be used in many real world applications which require real time performance under a computationally limited circumstance. In the paper, we study network architecture in YOLOv3 and modify the classical backbone-darknet53 of YOLOv3 by using a group of convolutions and dilated convolutions (DC). Then, a novel one-stage object detection network framework called DC-YOLOv3 is proposed. A lot of experiments on the Pascal 2017 benchmark prove the effectiveness of our framework. The results illustrate that DC-YOLOv3 achieves comparable results with YOLOv3 while being about 1.32× faster in training time and 1.38× faster in inference time. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

MEPNet: Medical Entity-Balanced Prompting Network for Brain CT Report Generation

arXiv

引用

arXiv 2025年

作者： Zhang, Xiaodan Shi, Yanzhao Ji, Junzhong Zheng, Chengxin Qu, Liangqiong College of Computer Science Beijing University of Technology Beijing China School of Computing and Data Science The University of Hong Kong Hong Kong

The automatic generation of brain CT reports has gained widespread attention, given its potential to assist radiologists in diagnosing cranial diseases. However, brain CT scans involve extensive medical entities, such as diverse anatomy regions and lesions, exhibiting highly inconsistent spatial patterns in 3D volumetric space. This leads to biased learning of medical entities in existing methods, resulting in repetitiveness and inaccuracy in generated reports. To this end, we propose a Medical Entity-balanced Prompting Network (MEPNet), which harnesses the large language model (LLM) to fairly interpret various entities for accurate brain CT report generation. By introducing the visual embedding and the learning status of medical entities as enriched clues, our method prompts the LLM to balance the learning of diverse entities, thereby enhancing reports with comprehensive findings. First, to extract visual embedding of entities, we propose Knowledge-driven Joint Attention to explore and distill entity patterns using both explicit and implicit medical knowledge. Then, a Learning Status Scorer is designed to evaluate the learning of entity visual embeddings, resulting in unique learning status for individual entities. Finally, these entity visual embeddings and status are elaborately integrated into multi-modal prompts, to guide the text generation of LLM. This process allows LLM to self-adapt the learning process for biased-fitted entities, thereby covering detailed findings in generated reports. We conduct experiments on two brain CT report generation benchmarks, showing the effectiveness in clinical accuracy and text coherence. Copyright © 2025, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Pairwise Preference over Multi-type Implicit Feedback Confidence Based Bayesian Personalized Ranking for Collaborative Filtering

Pairwise Preference over Multi-type Implicit Feedback Confid...

引用

IEEE Joint International Information technology and Artificial Intelligence Conference (ITAIC)

作者： Shan Gao Junwei Jin Bicao Li Cuijuan Lou Yuying Jiang School of Artificial Intelligence and Big Data Henan University of Technology Zhengzhou China School of Electronic and Information Engineering Zhongyuan University of Technology Zhengzhou China

Implicit feedback (e.g., purchases, views, clicks) have been receiving more attention due to their close relationship with real applications, including shopping, making friends, healthcare, etc. In some recent works, pairwise ranking algorithms have been shown to leverage different types of users' behaviors to solve the recommendation problem of heterogeneous implicit feedback. The ever-growing healthcare websites impose a new challenge for medical services recommendation. In this paper, we develop a novel preference learning algorithm to learn a confidence over multiple users' actions, which is called Multi-type Implicit Feedback Confidence Learned Bayesian Personalized Ranking (MTC-BPR). MTC-BPR has the merits of generating positive data from multiple auxiliary feedback and accommodates both the original and generated data for pairwise ranking. Experimental results on the real-world dataset from a healthcare website demonstrate that MTC-BPR achieves more accurate recommendations than several state-of-the-art methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A data-dependent algorithm for querying earth mover’s distance with low doubling dimensions

A data-dependent algorithm for querying earth mover’s dista...

引用

2021 SIAM International Conference on data Mining, SDM 2021

作者： Ding, Hu Chen, Tan Yang, Fan Wang, Mingyue School of Computer Science and Technology University of Science and Technology of China China School of Data Science University of Science and Technology of China China

ISBN: (纸本)9781611976700

In this paper, we consider the following query problem: given two weighted point sets A and B in the Euclidean space Rd, we want to quickly determine that whether their earth mover’s distance (EMD) is larger or smaller than a pre-specified threshold T ≥ 0. The problem finds a number of important applications in the fields of machine learning and data mining. In particular, we assume that the dimensionality d is not fixed and the sizes |A| and |B| are large. Therefore, most of existing EMD algorithms are not quite efficient to solve this problem due to their high complexities. Here, we consider the problem under the assumption that A and B have low intrinsic dimensions, which is common for high-dimensional data in real world. Inspired by the geometric method net tree, we propose a novel "data-dependent" algorithm to avoid directly computing the EMD between A and B, so as to solve this query problem more efficiently. We also study the performance of our method in practice. The experimental results suggest that our method can save a large amount of running time comparing with existing EMD algorithms. © 2021 by SIAM.

关键词： Earth (planet)

来源：评论

学校读者我要写书评

暂无评论

AdvMap: Crafting Adversarial Maps to Counter AI Aimbot in First-Person Shooter Games

引用

IEEE Transactions on Games 2025年

作者： Yan, Kai Chen, Xianyi Cui, Qi Yuan, Haoqin Tan, Zhenshan Nanjing University of Information Science and Technology Engineering Research Center of Digital Forensics Ministry of Education the School of Computer Science Nanjing210044 China Beijing Jiaotong University Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport Beijing100044 China

AI-based automatic aiming cheats (a.k.a., AI aimbots) have proliferated in first-person shooter (FPS) games, which grant malicious users an unfair gameplay advantage. Since AI aimbots operate independently of game data and are developed using object detection algorithms, they are difficult to detect with traditional anti-cheating methods. To actively counter AI aimbots, we propose AdvMap, which introduces invisible adversarial perturbations into game scene elements. In optimizing these adversarial perturbations, we design a mixture-of-misleading loss function that increases the total target confidence score within each misleading bounding box. It mitigates the risk of segment missing even when AdvMap is obscured, thereby enhancing the robustness. Besides, an L1-norm constraint with a small scale is employed during each update of the adversarial perturbations, which preserves the fidelity of the game scene. Additionally, to enable effective adaptation to interact with various elements within game environments, we introduce an Image-Subspace-Based Multi-Directional (ISMD) optimization strategy. It enables the adversarial perturbations to adaptively fit into each element by leveraging the mapping relationship between the game's 3D scenes and its corresponding 2D images. Furthermore, we construct a comprehensive benchmark, which includes various FPS games with different graphics styles and perspectives. Extensive experimental results demonstrate the efficacy of our method in countering various AI aimbot tools on different state-of-the-art object detection methods. © 2021 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Financial Impact of ESG News Sentiment on ESG Finance Trends 10

Analyzing the Financial Impact of ESG News Sentiment on ESG ...

引用

10th International Conference on Platform technology and Service, PlatCon 2024

作者： Kim, Minjoong Kim, Sujin Kim, Yongsung Moon, Jihoon Dept. of ICT Convergence Soonchunhyang University Asan Korea Republic of Graduate School of Education Chungnam National University Daejeon Korea Republic of Dept. of Technology Education Chungnam National University Daejeon Korea Republic of Dept. of AI and Big Data Soonchunhyang University Asan Korea Republic of

ISBN: (纸本)9798350367874

This paper examines the interconnections between environmental, social, and governance (ESG) financial trends and the sentiment analysis of ESG-related news from 2019 to 2022. A substantial corpus of news articles and ESG management data is subjected to analysis employing a range of techniques, including term frequency-inverse document frequency (TF-IDF), latent Dirichlet allocation (LDA) topic modeling, and sentiment analysis tools such as valence-aware dictionary and sentiment reasoner (VADER) and bidirectional encoder representations from transformers (BERT). This analysis identifies significant trends and relationships between sentiment direction, ESG ratings, and financial performance. The results demonstrate that while ESG ratings do not have a robust direct correlation with financial metrics, sentiment analysis of ESG news offers insights into the potential impact of ESG management. This study underscores the necessity for standardized ESG metrics to enhance the clarity and efficacy of measuring the financial impact of ESG management. These insights are intended to inform the formulation of policies and strategies for future ESG management. © 2024 IEEE.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

X-ray Flaw Detection of Welds Based on Slice Inference

X-ray Flaw Detection of Welds Based on Slice Inference

引用

6th International Conference on Intelligent Computing and Signal Processing (ICSP)

作者： Rui Wang Shaoze Gao Chen Wang Peng Zhou School of Artificial Intelligence and Data Science Hebei University of Technology Tianjin China

Aiming at the problems of high detection difficulty and low recognition rate due to the large length-to-width ratio of the weld image and complex defect imaging, this paper proposes a YOLO-SD model with a slight increase in the number of parameters. The model adds a convolutional attention mechanism module CBAM(Convolutional Block Attention Module) to the YOLOv5s (You Only Look Once) backbone network, which improves the accuracy of the model by enhancing effective features and suppressing invalid features; for large-size images with a fixed step size Slice the image, perform detection on the smaller image after the image is taken, and merge the prediction results on the image after the full image detection is completed. The experimental results show that the mAP value of the improved detection model in this paper is increased from 89.73% to 93.46%,and the YOLO-SD model has a stronger ability to identify weld defects. This method has higher accuracy and recognition rate under the premise of controlling the cost of weld defect detection.

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：