检索结果-内蒙古大学图书馆

Vibration Control of an Experimental Flexible Manipulator Against Input Saturation

IEEE/CAA Journal of Automatica Sinica 2023年第5期10卷 1340-1342页

作者： Zhijia Zhao Sentao Cai Ge Ma F.Richard Yu the School of Mechanical and Electrical Engineering Guangzhou UniversityGuangzhou 510006 the Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ) Shenzhen 518060China the Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ) College of Computer Science&Software EngineeringShenzhen UniversityShenzhen 518060China

Dear Editor, This letter considers the control problem of an experimental flexible manipulator in position tracking, vibration suppression, and saturation compensation. Based on the backstepping technology and a Nussbaum function, we develop an anti-windup control to restrain the manipulator’s vibration, realize the desire trajectory tracking, and eliminate the saturation.

关键词： restrain eliminate Vibration

来源：评论

学校读者我要写书评

暂无评论

Binary Electric Eel Foraging Optimization Algorithm for Solving 0-1 Knapsack Problems 6

Binary Electric Eel Foraging Optimization Algorithm for Solv...

引用

6th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2024

作者： Limane, Abdelhadi Zitouni, Farouq Ferhat, Aridj Lakbichi, Rihab Kasdi Merbah University Laboratory of Artificial Intelligence And Information Technologies Department of Computer Science And Information Technology Ouargla Algeria

ISBN: (纸本)9798350350265

Over the past few decades, the field of metaheuristics optimization has undergone rapid development. More than 650 different metaheuristic algorithms have been created so far. Among these, the Electric Eel Foraging Optimization (EEFO) stands out as a recently developed bio-inspired algorithm. To extend its application to binary optimization, particularly the 0-1 knapsack problem (KP01), we introduce a binary version of EEFO, termed EB-EEFO. The proposed algorithm employs two kinds of transfer functions, S-shaped and V-shaped, to shift from the continuous to binary search space. Additionally, we propose an enhanced repair and improvement (RI) method to address infeasible solutions, aiming to enhance solution diversity by reducing the inherent greedy nature of the traditional RI method. To Study the competitiveness of the new RI method, we develop the binary versions of five different algorithms using the RI method found in the literature, including, EEFO (B-EEFO), B-RIME, One-to-One-Based Optimizer (B-OOBO), Sinh Cosh Optimizer (B-SCHO), and Newton-Raphson-Based Optimizer (B-NRBO). The performance of these algorithms is evaluated using a dataset of 25 instances of KP01. The concluding remarks commend the performance of the EB-EEFO. © 2024 IEEE.

关键词： Transfer functions

来源：评论

学校读者我要写书评

暂无评论

Layer-Wise Learning Rate Optimization for Task-Dependent Fine-Tuning of Pre-Trained Models: An Evolutionary Approach

引用

ACM Transactions on Evolutionary Learning and Optimization 2024年第4期4卷 1-23页

作者： Bu, Chenyang Liu, Yuxin Huang, Manzong Shao, Jianxuan Ji, Shengwei Luo, Wenjian Wu, Xindong Key Laboratory of Knowledge Engineering with Big Data Ministry of Education and School of Computer Science and Information Engineering Hefei University of Technology Hefei China School of Artificial Intelligence and Big Data Hefei University Hefei China Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies School of Computer Science and Technology Harbin Institute of Technology Shenzhen China

The superior performance of large-scale pre-Trained models, such as Bidirectional Encoder Representations from Transformers (BERT) and Generative Pre-Trained Transformer (GPT), has received increasing attention in both academic and industrial research and has become one of the current research hotspots. A pre-Trained model refers to a model trained on large-scale unlabeled data, whose purpose is to learn general language representation or features for fine-Tuning or transfer learning in subsequent tasks. After pre-Training is complete, a small amount of labeled data can be used to fine-Tune the model for a specific task or domain. This two-stage method of "pre-Training+fine-Tuning"has achieved advanced results in natural language processing (NLP) tasks. Despite widespread adoption, existing fixed fine-Tuning schemes that adapt well to one NLP task may perform inconsistently on other NLP tasks given that different tasks have different latent semantic structures. In this article, we explore the effectiveness of automatic fine-Tuning pattern search for layer-wise learning rates from an evolutionary optimization perspective. Our goal is to use evolutionary algorithms to search for better task-dependent fine-Tuning patterns for specific NLP tasks than typical fixed fine-Tuning patterns. Experimental results on two real-world language benchmarks and three advanced pre-Training language models show the effectiveness and generality of the proposed framework. © 2024 held by the owner/author(s).

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Complexity of Motion Planning of Arbitrarily Many Robots: Gadgets, Petri Nets, and Counter Machines 2

Complexity of Motion Planning of Arbitrarily Many Robots: Ga...

引用

2nd Symposium on Algorithmic Foundations of Dynamic Networks, SAND 2023

作者： Ani, Joshua Coulombe, Michael Demaine, Erik D. Diomidov, Yevhenii Gomez, Timothy Hendrickson, Dylan Lynch, Jayson Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology CambridgeMA United States

ISBN: (纸本)9783959772754

We extend the motion-planning-through-gadgets framework to several new scenarios involving various numbers of robots/agents, and analyze the complexity of the resulting motion-planning problems. While past work considers just one robot or one robot per player, most of our models allow for one or more locations to spawn new robots in each time step, leading to arbitrarily many robots. In the 0-player context, where all motion is deterministically forced, we prove that deciding whether any robot ever reaches a specified location is undecidable, by representing a counter machine. In the 1-player context, where the player can choose how to move the robots, we prove equivalence to Petri nets, EXPSPACE-completeness for reaching a specified location, PSPACE-completeness for reconfiguration, and ACKERMANN-completeness for reconfiguration when robots can be destroyed in addition to spawned. Finally, we consider a variation on the standard 2-player context where, instead of one robot per player, we have one robot shared by the players, along with a ko rule to prevent immediately undoing the previous move. We prove this impartial 2-player game EXPTIME-complete. © Joshua Ani, Michael Coulombe, Erik D. Demaine, Yevhenii Diomidov, Timothy Gomez, Dylan Hendrickson, and Jayson Lynch;licensed under Creative Commons License CC-BY 4.0.

关键词： Petri nets

来源：评论

学校读者我要写书评

暂无评论

AI learns to write sorting software on its own

引用

NATURE 2023年第7964期618卷 240-241页

作者： Solar-Lezama, Armando Armando Solar-Lezama is in the Computer Science & Artificial Intelligence Laboratory Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA.

Deep reinforcement learning has been used to improve computer code by treating the task as a game-with no special knowledge needed on the part of the player. The result has already worked its way into countless progra... 详细信息

关键词： Machine learning Mathematics and computing

来源：评论

学校读者我要写书评

暂无评论

Enhancing Malware Detection using Deep Learning with SMOTE and Noise-based Augmentation 6

Enhancing Malware Detection using Deep Learning with SMOTE a...

引用

6th International Conference on Advanced Aspects of Software Engineering, ICAASE 2024

作者： Khadidja, Belattar Mehdi Mohamed, Ait Mohamed, Abdellaoui Imene, Akkouche Okba, Zobiri Computer Science Department University of Algiers Algiers Algeria Laboratory of Research in Artificial Intelligence University of Science and Technology Houari Boumediene Algiers Algeria

ISBN: (纸本)9798331522810

The increasing prevalence of malware poses a significant security threat to financial institutions, businesses, and individuals. Tackling this growing issue requires innovative methods for the accurate and rapid identification and classification of malware, facilitating comprehensive behavioral analysis. Deep learning has emerged as a powerful tool for malware detection due to its ability to identify complex patterns and anomalies in data. This study investigates the application of a custom-designed Convolutional Neural Network (CNN) alongside a hybrid CNN-Long Short-Term Memory (LSTM) model, utilizing augmentation techniques for image-based malware classification. The augmentation methods employed include the Synthetic Minority Over-sampling Technique (SMOTE) and the additive noise. Using the MalImg dataset, the custom-built CNN achieved an impressive classification accuracy of 99.89%, and the hybrid CNN-LSTM model reached 99.68%. Both models outperformed existing state-of-The-Art approaches. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Survey in the Use of Deep Learning Techniques in The Open Classroom Approach 6

A Survey in the Use of Deep Learning Techniques in The Open ...

引用

6th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2024

作者： Boutabia, Ines Benmachiche, Abdelmadjid Betouil, Ali Abdelatif Chemam, Chaouki Chadli Bendjedid University Lima Laboratory Faculty of Science And Technology House of Artificial Intelligence Dept of Computer Science El-Tarf36000 Algeria

ISBN: (纸本)9798350350265

In the wake of the COVID-19 pandemic, the educational system found itself in dire need of a reliable tool to continue its activities. Enter the E-Learning system, a tool that had been overlooked or forgotten by many of its users. One of the most promising trends in E-learning is the open classrooms (OCR) approach, which aims to provide an all-encompassing online education that is easily accessible to everyone, thereby enabling numerous individuals to enhance their skills and capabilities. Over the years, there have been many attempts to integrate deep-learning techniques into the OCR approach to enhance performance and improve results. To gain a better understanding of this topic, our paper presents a comprehensive literature review of researchers who have used deep-learning approaches to improve the outcomes of e-learning in Open classrooms. Our primary focus is on individuals who employ neural network-based techniques, such as artificial Neural Networks, Convolutional Neural Networks, Deep Neural Networks, Recurrent Neural Networks, and hybrid approaches. After conducting a thorough evaluation of the selected methods, we have been able to identify their respective strengths and weaknesses. By scrutinizing these approaches, we have successfully outlined the benefits and drawbacks of each, allowing us to provide a confident and informative analysis of this important subject. © 2024 IEEE.

关键词： E-learning

来源：评论

学校读者我要写书评

暂无评论

Ouroboros: Generating Longer Drafts Phrase by Phrase for Faster Speculative Decoding

Ouroboros: Generating Longer Drafts Phrase by Phrase for Fas...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Zhao, Weilin Huang, Yuxiang Han, Xu Xu, Wang Xiao, Chaojun Zhang, Xinrong Fang, Yewei Zhang, Kaihuo Liu, Zhiyuan Sun, Maosong Department of Computer Science and Technology Institute for Artificial Intelligence Beijing National Research Center for Information Science and Technology Tsinghua University Beijing China Shanghai Artificial Intelligence Laboratory Shanghai China ModelBest Inc. China

ISBN: (纸本)9798891761643

Speculative decoding is a widely used method that accelerates the generation process of large language models (LLMs) with no compromise in model performance. It achieves this goal by using an existing smaller model for drafting and then employing the target LLM to verify the draft in a low-cost parallel manner. Under such a drafting-verification framework, drafting efficiency has become a bottleneck in the final speedup of speculative decoding. Therefore, generating longer drafts at less cost can lead to better decoding speedup. To achieve this, we introduce Ouroboros, which can generate draft phrases to parallelize the drafting process and meanwhile lengthen drafts in a training-free manner. The experimental results on various typical text generation tasks show that Ouroboros can achieve speedups of up to 2.4× over speculative decoding and 3.9× over vanilla decoding, without fine-tuning draft and target models. The source code of Ouroboros is available at https://***/thunlp/Ouroboros. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

A new method to analyze visual efficiency in interactive web environment

A new method to analyze visual efficiency in interactive web...

引用

2023 International Conference on computer Graphics, artificial intelligence, and Data Processing, ICCAID 2023

作者： Lv, Wen Li, Tingting Xie, Zhenping School of Artificial Intelligence and Computer Science Jiangnan University China Jiangsu Key Laboratory of Media Design and Software Technology China

ISBN: (纸本)9781510677685

Interactive web interfaces are widely used, yet users often struggle to discover important and useful information. Psychologists have pointed out that in the process of human visual perception, the global topological characteristics of objects will be perceived first. In this study, we propose a cognitive experiment system to explore what object properties affect visual efficiency in interactive web environments. Our experimental results showed that, first, if a small number of colors are allocated to different objects, these objects could be more effectively memorized than no color or too many colors. Second, in static single-page, different topology structures will not cause apparently different visual attraction, however, topological structural differences could be stronger memorized. Third, in dynamic HTML, the motion changes will induce stronger visual attention and memory intensity, than their own topology changes. Subsequently, we verified the feasibility of the experiment system in real webpages and demonstrated that the above results can theoretically guide better web interaction design works with less cognitive load of users, higher operation efficiency and more natural user experience. In addition, some design reference suggestions are also provided in our study. © 2024 SPIE.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Large circuit models: opportunities and challenges

引用

science China(Information sciences) 2024年第10期67卷 25-66页

作者： Lei CHEN Yiqi CHEN Zhufei CHU Wenji FANG Tsung-Yi HO Ru HUANG Yu HUANG Sadaf KHAN Min LI Xingquan LI Yu LI Yun LIANG Jinwei LIU Yi LIU Yibo LIN Guojie LUO Hongyang PAN Zhengyuan SHI Guangyu SUN Dimitrios TSARAS Runsheng WANG Ziyi WANG Xinming WEI Zhiyao XIE Qiang XU Chenhao XUE Junchi YAN Jun YANG Bei YU Mingxuan YUAN Evangeline F.Y.YOUNG Xuan ZENG Haoyi ZHANG Zuodong ZHANG Yuxiang ZHAO Hui-Ling ZHEN Ziyang ZHENG Binwu ZHU Keren ZHU Sunan ZOU Huawei Noah’s Ark Lab School of Integrated Circuits Peking University Faculty of Electrical Engineering and Computer Science Ningbo University Department of Electronic and Computer Engineering Hong Kong University of Science and Technology Department of Computer Science and Engineering The Chinese University of Hong Kong School of Integrated Circuits Southeast University Huawei HiSilicon Peng Cheng Laboratory School of Computer Science Peking University School of Microelectronics State Key Laboratory of Integrated Chips and System Fudan University School of Artificial Intelligence Shanghai Jiao Tong University

Within the electronic design automation(EDA) domain, artificial intelligence(AI)-driven solutions have emerged as formidable tools, yet they typically augment rather than redefine existing methodologies. These solutions often repurpose deep learning models from other domains, such as vision, text, and graph analytics, applying them to circuit design without tailoring to the unique complexities of electronic circuits. Such an “AI4EDA” approach falls short of achieving a holistic design synthesis and understanding,overlooking the intricate interplay of electrical, logical, and physical facets of circuit data. This study argues for a paradigm shift from AI4EDA towards AI-rooted EDA from the ground up, integrating AI at the core of the design process. Pivotal to this vision is the development of a multimodal circuit representation learning technique, poised to provide a comprehensive understanding by harmonizing and extracting insights from varied data sources, such as functional specifications, register-transfer level(RTL) designs, circuit netlists,and physical layouts. We champion the creation of large circuit models(LCMs) that are inherently multimodal, crafted to decode and express the rich semantics and structures of circuit data, thus fostering more resilient, efficient, and inventive design methodologies. Embracing this AI-rooted philosophy, we foresee a trajectory that transcends the current innovation plateau in EDA, igniting a profound “shift-left” in electronic design methodology. The envisioned advancements herald not just an evolution of existing EDA tools but a revolution, giving rise to novel instruments of design-tools that promise to radically enhance design productivity and inaugurate a new epoch where the optimization of circuit performance, power, and area(PPA) is achieved not incrementally, but through leaps that redefine the benchmarks of electronic systems' capabilities.

关键词： AI-rooted EDA large circuit models (LCMs) multimodal circuit representation learning circuit optimization

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：