检索结果-内蒙古大学图书馆

1st International Conference on Intelligent Manufacturing and Cloud Computing, ICIMCC 2024

作者： Guan, Yunxin Yun, Jing Sun, Pengfei Su, Xiaoming College of Data Science and Application Inner Mongolia University of Technology Hohhot China

ISBN: (纸本)9781643685779

In the design of modern compilers, the generation and optimization of intermediate code play a crucial role. Serving as a bridge between source code and target machine code, intermediate code provides ample opportunities for optimization. This paper focuses on porthole optimization techniques, proposing a new optimization strategy aimed at reducing unnecessary computations and memory accesses by analyzing control and data flows within intermediate code, thereby enhancing program execution efficiency. We first outline the various stages of the compilation process, followed by a detailed exploration of the basic principles and implementation methods of porthole optimization. Finally, we validate the effectiveness and advantages of this optimization strategy through experiments. The experimental results indicate significant improvements in the execution performance of intermediate code after the adoption of porthole optimization, providing new insights and directions for compiler optimization. © 2025 The Authors.

关键词： Program compilers

来源：评论

学校读者我要写书评

暂无评论

引用

23rd China National Conference on Computational Linguistics, CCL 2024

作者： Sun, Yaqi Yun, Jing Ma, Zhuoqun College of Data Science and Application Inner Mongolia University of Technology Huhhot China

ISBN: (纸本)9789819783663

There is a continuous and explosive growth of multimodal data. Efficient cross-modal hashing retrieval is of significant importance in conserving computational *** further enhance the attention to informative data within modalities and capture the semantic correlations in cross-modal data, we propose an enhanced deep Joint-Semantics Reconstructing Hashing algorithm, which is the Joint Similarity Guidance Hash Coding Based on Adaptive Weight Mixing Strategy(JSGHCA). The algorithm focuses on delving deeper into the correlations of the data in cross-modal. We introduce the adaptive weight mixing strategy to construct the semantic affinity matrix, so that the matrix can identify each modal data with specific weight in each batch. At the same time, in the process of the hash code generation, we introduce collaborative attention mechanism. It helps the model to pay more attention to the local information of each modality, thereby capturing the semantic features within each modality more accurately. Additionally, it enables the model to jointly process the attention across different modalities and extract shared semantic features more precisely. Experimental results show that the proposed model is significantly better than the deep joint semantic reconstruction hash algorithm on multiple benchmark datasets. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Research on Neural Machine Translation Based on Adversarial Knowledge Distillation 24

Research on Neural Machine Translation Based on Adversarial ...

引用

8th International Conference on Computer science and Artificial Intelligence, CSAI 2024

作者： Ma, Chang Tian, Hong Yong Zheng, Li Xiao Sun, Kang Kang College of Data Science and Application Inner Mongolia University of Technology Inner Hohhot China

ISBN: (纸本)9798400718182

Since the traditional Neural Machine Translation (NMT) model has a large number of model parameters and redundant structure, how to compress the existing model without affecting the performance has become a current research hot-spot. Knowledge distillation, as a commonly used model compression technique in the field of Natural Language Processing (NLP), aims to reduce the number of model parameters and computational resources by transferring the knowledge from a large, complex teacher model to a smaller, simple student model. In this regard, we propose a lightweight neural machine translation model (A Lite NMT, ALNMT) model construction in an adversarial teacher-student framework. By compressing the model using adversarial knowledge distillation techniques and incorporating neural knowledge base techniques in the teacher model, the teacher model is able to obtain a valid target language when dealing with scarce vocabulary, which in turn improves the quality of the translation. We conducted experiments on two large-scale machine translation tasks, WMT19 Chinese-English and WMT14 English-German, and one low- and medium-resource language, Mongolian-Chinese translation task, respectively, and our method improved the BLEU values by 2.83, 2.79, and 1.69, respectively, compared to the baseline model. © 2024 Copyright held by the owner/author(s).

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Stock trend prediction based on multi-factor screening machine 24

Stock trend prediction based on multi-factor screening machi...

引用

5th International Conference on Computer science and Management Technology, ICCSMT 2024

作者： Hu, Ruijie Faculty of Science and Technology Guangdong Provincial Key Laboratory of Interdisciplinary Research and Application for Data Science BNU-HKBU United International College Zhuhai China

ISBN: (纸本)9798400709999

With the rapid development of financial markets, stock price prediction, as a highly complex and challenging topic, has attracted more and more attention from academia and industry. This study aims to use multiple machine learning algorithms to predict stock market prices, especially exploring the advantages of deep learning models in capturing market nonlinear and time series characteristics. In this article, a machine learning model is used for contribution analysis, analyzing multiple features of stocks, performing a weighted combination analysis of features, and then using a convolutional neural network to perform feature extraction and analysis of stocks, and combining the attention mechanism for deep feature extraction, and finally use the long short-term memory network (LSTM) model to output the model to analyze and predict future stock trends. The experimental results show that combining deep learning analysis technology and factor screening machine can significantly improve the prediction accuracy of the model. At the same time, this article also compared the performance of different models in different market environments, and concluded that in markets with higher volatility, models based on deep learning are more stable. Ultimately, this research provides a new technical means for risk management and investment decisions in financial markets. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

PD-YOLO:Colon Polyp Detection Model Based on Enhanced Small-Target Feature Extraction

引用

Computers, Materials & Continua 2025年第1期82卷 913-928页

作者： Yicong Yu Kaixin Lin Jiajun Hong Rong-Guei Tsai Yuanzhi Huang New Engineering Industry College Putian UniversityPutian351100China Engineering Research Center of Big Data Application in Private Health Medicine(Putian University) Fujian Province UniversityPutian351100China Computer College Guangdong University of Science&TechnologyDongguan523668China

In recent years,the number of patientswith colon disease has increased *** polyps are the precursor lesions of colon *** not diagnosed in time,they can easily develop into colon cancer,posing a serious threat to patients’lives and health.A colonoscopy is an important means of detecting colon ***,in polyp imaging,due to the large differences and diverse types of polyps in size,shape,color,etc.,traditional detection methods face the problem of high false positive rates,which creates problems for doctors during the diagnosis *** order to improve the accuracy and efficiency of colon polyp detection,this question proposes a network model suitable for colon polyp detection(PD-YOLO).This method introduces the self-attention mechanism CBAM(Convolutional Block Attention Module)in the backbone layer based on YOLOv7,allowing themodel to adaptively focus on key information and ignore the unimportant *** help themodel do a better job of polyp localization and bounding box regression,add the SPD-Conv(Symmetric Positive Definite Convolution)module to the neck layer and use deconvolution instead of *** results indicate that the PD-YOLO algorithm demonstrates strong robustness in colon polyp *** to the original YOLOv7,on the Kvasir-SEG dataset,PD-YOLO has shown an increase of 5.44 percentage points in AP@0.5,showcasing significant advantages over other mainstream methods.

关键词： Polyp detection YOLOv7 SPD-Conv CBAM deconvolution

来源：评论

学校读者我要写书评

暂无评论

YOLO-FSE: An Improved Target Detection Algorithm for Vehicles in Autonomous Driving

引用

IEEE Internet of Things Journal 2025年第10期12卷 13922-13933页

作者： Xue, Tao Liu, Zhiwei Lan, Siwu Zhang, Qingda Yang, Aimin Li, Jie Beijing Information Science and Technology University Beijing100000 China North China University of Science and Technology Key Laboratory of Data Science and Application of Hebei Province Tangshan063210 China North China University of Science and Technology College of Science Tangshan063210 China North China University of Science and Technology College of Metallurgy and Energy Tangshan063210 China North China University of Science and Technology College of Metallurgy and Energy Tangshan063009 China

Real-time object detection plays a critical role in advancing autonomous driving technologies. To meet the demands for real-time performance, a lightweight model is essential to reduce parameter size, simplify complexity, and enhance detection speed. This paper presents YOLO-FSE, a compact vehicle detection model built upon the YOLOv5 framework. In YOLO-FSE, the C3 module is substituted with the C3faster module from the FasterNet lightweight architecture. The backbone network is augmented with the Shuffle Attention module, which improves feature fusion by separating spatial and channel attention mechanisms. Additionally, the original loss function is replaced by EIoU loss, which directly penalizes width and height predictions, thereby improving the model's generalization ability. During experimentation, it was assumed that the number of test images was sufficiently large and clear, with PyTorch serving as the primary framework for model development and training. Experimental results show that, on the UA-DETRAC and BIT-Vehicle datasets, the mean average precision (mAP) achieved 98.9% and 97.6%, respectively. Computational complexity (FLOPs) was reduced by 12.65%, while the average frame rate (Fps) increased by 46.13%. Overall, the enhanced model exhibits substantial improvements in both speed and performance, demonstrating its potential for deployment in autonomous driving applications. © 2014 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Dual Test-Time Training for Out-of-Distribution Recommender System

引用

IEEE Transactions on Knowledge and data Engineering 2025年第6期37卷 3312-3326页

作者： Yang, Xihong Wang, Yiqi Chen, Jin Fan, Wenqi Zhao, Xiangyu Zhu, En Liu, Xinwang Lian, Defu National University of Defense Technology College of Computer Science and Technology Changsha410073 China The Hong Kong University of Science and Technology School of Business and Management Hong Kong Hong Kong City University of Hong Kong Department of Data Science Hong Kong University of Science and Technology of China Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology Hefei Anhui230000 China

Deep learning has been widely applied in recommender systems, which has recently achieved revolutionary progress. However, most existing learning-based methods assume that the user and item distributions remain unchanged between the training phase and the test phase. However, the distribution of user and item features can naturally shift in real-world scenarios, potentially resulting in a substantial decrease in recommendation performance. This phenomenon can be formulated as an Out-Of-Distribution (OOD) recommendation problem. To address this challenge, we propose a novel Dual Test-Time-Training framework for OOD Recommendation, termed DT3OR. In DT3OR, we incorporate a model adaptation mechanism during the test-time phase to carefully update the recommendation model, allowing the model to adapt specially to the shifting user and item features. To be specific, we propose a self-distillation task and a contrastive task to assist the model learning both the user's invariant interest preferences and the variant user/item characteristics during the test-time phase, thus facilitating a smooth adaptation to the shifting features. Furthermore, we provide theoretical analysis to support the rationale behind our dual test-time training framework. To the best of our knowledge, this paper is the first work to address OOD recommendation via a test-time-training strategy. We conduct experiments on five datasets with various backbones. Comprehensive experimental results have demonstrated the effectiveness of DT3OR compared to other state-of-the-art baselines. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Time Series Decision Analysis Based on Linear Programming and SARIMA

Time Series Decision Analysis Based on Linear Programming an...

引用

6th International Conference on Cyber Security Intelligence and Analytics, CSIA 2024

作者： Li, Shuai Zhang, Zeyuan Jiang, Dongming College of Data Science and Application Inner Mongolia University of Technology Hohhot China College of Electric Power Inner Mongolia University of Technology Hohhot China

ISBN: (纸本)9783031882869

In this paper, a decision analysis method based on time series analysis and planning model is proposed. By establishing multiple regression equations, this paper analyzes the correlation between sales volume and cost-plus pricing, and uses SARIMA model to model the time distribution characteristics of cost data and predict the future trend of data. Furthermore, this paper constructs a linear programming model to maximize the expected revenue, taking into account the constraints such as loss rate and inventory capacity. Finally, the genetic algorithm is used to solve the planning model to achieve the optimal decision of daily replenishment and pricing. The empirical analysis shows that we apply this method to the purchase strategy of supermarkets, and the results can be effectively applied to practical problems. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Linear programming

来源：评论

学校读者我要写书评

暂无评论

GBC: Gaussian-Based Colorization and Super-Resolution for 3D Reconstruction 19

GBC: Gaussian-Based Colorization and Super-Resolution for 3D...

引用

19th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its applications in Industry, VRCAI 2024

作者： Liao, Jiecheng He, Shi Yuan, Yichen Zhang, Hui Department of Computer Science and Technology BNU-HKBU United International College ZhuHai China Guangdong Provincial Key Laboratory of Interdisciplinary Research and Application for Data Science BNU-HKBU United International College Zhuhai China

ISBN: (纸本)9798400713484

In this paper, we introduce GBC, an advanced framework for transforming low-resolution, monochrome video sequences into high-resolution, colorized, and geometrically accurate 3D models. GBC combines Bidirectional Optical Flow Super-Resolution (BOF-SR) for temporal super-resolution with a novel colorization approach, Temporal Optical Flow-based Colorization (TOF-CO), designed to enhance frame-to-frame temporal consistency. The integration of these modules with COLMAP-based 3D Gaussian splatting further extends GBC's capability to reconstruct high-fidelity 3D scenes. Additionally, we created a custom dataset tailored to the challenges of low-light, low-quality historical footage, enabling robust evaluation alongside public datasets. This solution offers a new approach to video restoration, advancing temporal coherence, color accuracy, and 3D scene fidelity. The source code is publicly available at https://***/ffftuanxxx/GBC. © 2024 ACM.

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

Improving Transferable Targeted Adversarial Attack via Normalized Logit Calibration and Truncated Feature Mixing

引用

IEEE Transactions on Information Forensics and Security 2025年 20卷 4595-4609页

作者： Weng, Juanjuan Luo, Zhiming Li, Shaozi Jinan University College of Information Science and Technology Guangzhou510632 China Xiamen University Department of Artificial Intelligence Xiamen361005 China Wuyi University Fujian Key Laboratory of Big Data Application and Intellectualization for Tea Industry Wuyishan354300 China

This paper aims to enhance the transferability of adversarial samples in targeted attacks, where attack success rates remain comparatively low. To achieve this objective, we propose two distinct techniques for improving the targeted transferability from the loss and feature aspects. First, in previous approaches, logit calibrations used in targeted attacks primarily focus on the logit margin between the targeted class and the untargeted classes among samples, neglecting the standard deviation of the logit. In contrast, we introduce a new normalized logit calibration method that jointly considers the logit margin and the standard deviation of logits. This approach effectively calibrates the logits, enhancing the targeted transferability. Second, previous studies have demonstrated that mixing the features of clean samples during optimization can significantly increase transferability. Building upon this, we further investigate a truncated feature mixing method to reduce the impact of the source training model, resulting in additional improvements. The truncated feature is determined by removing the Rank-1 feature associated with the largest singular value decomposed from the high-level convolutional layers of the clean sample. Extensive experiments conducted on the ImageNet-Compatible, CIFAR-10 and ImageNet-1k datasets demonstrate the individual and mutual benefits of our proposed two components, which outperform the state-of-the-art methods by a large margin in black-box targeted attacks. © 2005-2012 IEEE.

关键词： Calibration Closed box Training Standards Translation Perturbation methods Optimization Predictive models Feature extraction Industries

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：