检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Saito, Daichi Kanehira, Atsushi Sasabuchi, Kazuhiro Wake, Naoki Takamatsu, Jun Koike, Hideki Ikeuchi, Katsushi Department of Computer Science Tokyo Institute of Technology Tokyo Japan Applied Robotics Research Microsoft RedmondWA United States

In-hand tool manipulation is an operation that not only manipulates a tool within the hand (i.e., in-hand manipulation) but also achieves a grasp suitable for a task after the manipulation. This study aims to achieve an in-hand tool manipulation skill through deep reinforcement learning. The difficulty of learning the skill arises because this manipulation requires (A) exploring long-term contact-state changes to achieve the desired grasp and (B) highly-varied motions depending on the contact-state transition. (A) leads to a sparsity of a reward on a successful grasp, and (B) requires an RL agent to explore widely within the state-action space to learn highly-varied actions, leading to sample inefficiency. To address these issues, this study proposes Action Primitives based on Contact-state Transition (APriCoT). APriCoT decomposes the manipulation into short-term action primitives by describing the operation as a contact-state transition based on three action representations (detach, crossover, attach). In each action primitive, fingers are required to perform short-term and similar actions. By training a policy for each primitive, we can mitigate the issues from (A) and (B). This study focuses on a fundamental operation as an example of in-hand tool manipulation: rotating an elongated object grasped with a precision grasp by half a turn to achieve the initial grasp. Experimental results demonstrated that ours succeeded in both the rotation and the achievement of the desired grasp, unlike existing studies. Additionally, it was found that the policy was robust to changes in object shape. © 2024, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

The Least Square QR Method Improves Extreme Learning Machine

The Least Square QR Method Improves Extreme Learning Machine

引用

2023 IEEE World Conference on applied Intelligence and Computing, AIC 2023

作者： Subramani, Chinnamuthu Kuppili, Venkatanareshbabu Keshavamurthy, B.N. Jagannath, Ravi Prasad K. Prashanth, Gurusiddappa R. National Institute of Technology Goa Department of Applied Sciences Goa Ponda403401 India National Institute of Technology Goa Department of Computer Science and Engineering Goa Ponda403401 India National Institute of Technology Goa Department of Electronics and Communication Engineering Goa Ponda403401 India

ISBN: (纸本)9798350310061

The Extreme Learning Machine (ELM) is a fast and straightforward learning algorithm designed for single hidden layer feed-forward neural networks (SLFNs). The ELM randomly assigns input weights and calculates the output weights analytically using the Moore-Penrose generalized inverse (MP) (also called Pseudo inverse). However, the MP is not preferable for ill-posed systems as it is sensitive to perturbation and may produce inaccurate solutions. Furthermore, it is computationally expensive as it involves matrix inversion. Regularized ELM (RELM) was proposed to address the limitations of the MP. In RELM, the regularization parameter is the primary factor that controls the classifier's performance. Though the RELM is stable and produces accurate results, finding a n a ppropriate regularization parameter value is tedious and increases the computation. This work proposes a variant of ELM named LSQR-ELM algorithm, which balances the ELM and RELM by maintaining stability and accuracy and avoiding the regularization entirely. The LSQR-ELM is based on the Least Squares QR (LSQR) iterative technique. Since the LSQR finds p a rtial Q R rather than full QR, it reduces unnecessary computation. To verify the proposed algorithm's ability, comparisons are made with several existing methods including ELM, Minimum Residual Method (MT-ELM), SVD-based SG-ELM, Optimally Pruned ELM (OP-ELM), and K-nearest neighbor (KNN). Extensive experimen-tation on various datasets demonstrates that the LSQR-ELM algorithm outperforms ELM, SG-ELM, OP-ELM, and KNN on most datasets. Moreover, it exhibits comparable performance to MT-ELM while requiring significantly f e wer computations. © 2023 IEEE.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis in Social Media Marketing: Leveraging Natural Language Processing for Customer Insights 8th

Sentiment Analysis in Social Media Marketing: Leveraging Nat...

引用

8th International Conference on Information and Communication technology for Competitive Strategies, ICTCS 2023

作者： Singh, Kamred Udham Kumar, Ankit Kumar, Gaurav Choudhury, Tanupriya Singh, Teekam Kotecha, Ketan School of Computing Graphic Era Hill University Dehradun India Department of Computer Engineering and Applications GLA University Uttar Pradesh Mathura India CSE Department Symbiosis Institute of Technology Symbiosis International University Lavale Campus Pune411045 India Department of Computer Science and Engineering Graphic Era Deemed to Be University Dehradun India Symbiosis Centre for Applied Artificial Intelligence Symbiosis Institute of Technology Symbiosis International University Pune411045 India

ISBN: (纸本)9789819994885

In this day and age of online marketing, social media platforms have evolved into essential tools for companies to use in order to communicate with their consumers, as well as to advertise the goods and services they provide. Understanding client emotion, tastes, and views may be significantly aided by perusing the vast amounts of user-generated information that can be found on these many platforms. This study investigates the use of sentiment analysis, a branch of natural language processing (NLP), as a potent instrument for gleaning useful consumer insights from data collected from social media platforms like as Facebook and Twitter. The application of computer methods in the process of automatically determining and categorising the sentiment that is represented in textual data, such as tweets, Facebook posts, and online reviews, is what is known as sentiment analysis. In this article, the primary methodology and techniques used in sentiment analysis, such as lexicon-based, machine learning, and deep learning approaches, are dissected and discussed. In addition to this, it sheds light on the difficulties and factors to take into account that are unique to sentiment analysis when applied to the context of data from social media, such as the management of noisy and unstructured language, the management of sarcasm, and the management of context-dependent sentiment. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Real-time testing of vision-based systems for AGVs with ArUco markers

Real-time testing of vision-based systems for AGVs with ArUc...

引用

2022 IEEE International Conference on Big Data, Big Data 2022

作者： Filus, Katarzyna Sobczak, Lukasz Domanska, Joanna Domanski, Adam Cupek, Rafal Institute of Theoretical and Applied Informatics Polish Academy of Sciences Gliwice Poland Silesian University of Technology Faculty of Automatic Control Electronics and Computer Science Gliwice Poland

ISBN: (纸本)9781665480451

Automated Ground Vehicles (AGVs) use deep-learning-based vision systems to perceive the surrounding environment and extract relevant information about it. Although deep learning models offer high capabilities, they require large amounts of data to be properly trained and tested. Testing is especially important when off-the-shelf models are used by the AGVs - to examine whether they can meet the demands of complex environments such as the production halls of automated factories. One area of such perception algorithms is object recognition. To test such systems, we propose a solution based on ArUco fiducial markers used for automatic labeling of objects. Our solution can be used to test deep learning systems in real time directly on a robot. Our solution requires minimal interference with the environment and additional infrastructure - the desired objects only need to be marked with a marker printed on a home printer. Therefore, the presented testing procedure can be used for testing of AGVs in real-life environments during a real ride from an actual robot perspective. Data gathered during the online testing can be used for the offline comparison of the accuracy of different deep learning models. Although we focus on the online and offline testing in our study, we also incorporated a marker masking procedure. Therefore, the resulting datasets may also be used for training. © 2022 IEEE.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Intelligent Intrusion Detection System Based on Blockchain technology 1

Intelligent Intrusion Detection System Based on Blockchain T...

引用

1st IEEE Afro-Mediterranean Conference on Artificial Intelligence, AMCAI 2023

作者： Bessaad, Ferdaws Ktata, Farah Barika Kalboussi, Khalil Ben Higher Institute of Computer Science and Communication Techniques H.Sousse Sousse Tunisia MIRACL HIGHER INSTITUTE of APPLIED SCIENCES and TECHNOLOGY of SOUSSE Sousse Tunisia LAMIH POLYTECHNIC UNIVERSITY of HAUT de FRANCE Valenciennes France

ISBN: (纸本)9798350394689

computer networks are increasingly vulnerable to security disruptions such as congestion, malicious access, and attacks. Intrusion Detection Systems (IDS) play a crucial role in identifying and mitigating these threats. However, many IDSs have limitations, including reduced performance in terms of scalability, configurability, and fault tolerance. In this context, we aim to enhance intrusion detection through a cooperative approach. To achieve this, we propose the implementation of ICIDS-BB (Intelligent Cooperative Intrusion Detection System based on Blockchain). This system leverages Blockchain technology to secure data exchange among collaborative components. Internally, we employ two machine learning algorithms, the decision tree and random forest, to improve attack detection. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

The Optimum Switching Bilateral of Various Mix Noise Filters to reduce Medical X-Ray Images

The Optimum Switching Bilateral of Various Mix Noise Filters...

引用

International Electrical Engineering Congress (iEECON)

作者： Kriengkri Langampol Akara Prayote Wilaiporn Lee Kanabadee Srisomboon Department of Computer Engineering Faculty of Engineering Rajamangala University of Technology Phra Nakhon Thailand Department of Computer and Information Science Faculty of Applied Science King Mongkut's University of Technology North Bangkok Thailand Department of Electrical and Computer Engineering Faculty of Engineering King Mongkut's University of Technology North Bangkok Thailand

ISBN: (数字)9798350383591

ISBN: (纸本)9798350383607

In the past few years, image processing has been widely adopted for symptom diagnosis of medical application. To achieve accurate analysis, the medical applications require high quality image for applying to the symptom diagnosis applications. However, the medical image quality may drop due to the imperfection of devices and the environment. Noise is one of the most critical factors that reduces the image quality where the noise may form by combining two types of noises. In this paper, we considered four well-known mixed noise filters including TF, BM3D, NCSR and OSBF filters. Since the OSBF filter is the alternative version of SBF filter that requires to determine the optimal weights, therefore we first determine the optimal weights for the medical image under the combination of Gaussian noise and salt-and-pepper noise. Then, we evaluate the filtering performance under three metrics including PSNR, SSIM and filtering time. As a result, OSBF filter achieves the best filtering performance since it can reduce the noise effect, preserve image details and consumes the short filtering time. On the other hand, the filtering performance of the TF filter suffers from the combining of salt-and-pepper noise.

关键词： Image quality Biomedical equipment Filters Three-dimensional displays Sparse approximation Noise Switches

来源：评论

学校读者我要写书评

暂无评论

Embedding Color Watermark by Adjusting DCT using RGB Gray Scale Watermarking 1

Embedding Color Watermark by Adjusting DCT using RGB Gray Sc...

引用

1st IEEE International Conference on Emerging Trends in Industry 4.0, ETI 4.0 2021

作者： Gupta, Neha Bhansali, Ashok Symbiosis University of Applied Sciences School of Computer Science and Information Technology Indore India OP Jindal University Computer Science Engineering Raigarh India

ISBN: (纸本)9781665422376

Digital Watermarking with PSO and Fuzzy Logic is an attempt to find suitable location using PSO and Fuzzy Logic by looking at the surrounding pixels and adaptively adjusting the pixel intensity values to encode the watermark bits. The result obtained using this technique indicates that following adaptive insertion on the pixels after finding location by PSO is more effective to obtain better fidelity and robustness. The inverse tradeoff between robustness and fidelity is also demising. The method proposed includes color watermarking as an extension of gray scale watermarking by dividing color image into RGB gray scale and then uniting them together. © 2021 IEEE.

关键词： Fuzzy logic

来源：评论

学校读者我要写书评

暂无评论

Web Extension for Phishing URL Identification 3

Web Extension for Phishing URL Identification

引用

3rd International Conference on Intelligent Computing, Instrumentation and Control Technologies, ICICICT 2022

作者： Kiruthi Whasan, W.R. Khiran Khumhaar, W.R. Reddy, Kalluri Raviteja Dhanalakshmi, R. Kapilan Radhakrishnan, P. Kcg College of Technology Department of Computer Science and Engineering Tamil Nadu Chennai India Applied Computing University of Wales Trinity Saint David United Kingdom

ISBN: (数字)9781665410052

ISBN: (纸本)9781665410052

Phishing attacks have been the most consistent and emerging cyber-attack among other social engineering attacks. In spite of several strict mechanisms and policies to identify and avoid phishing websites, new techniques and tools are used by the attackers to exploit the security measures. This paper proposes a system that uses machine learning algorithm to identify phishing URL based on web parameters. Among several machine learning algorithms, Random Forest algorithm is chosen as it provides more accuracy and performance for the identification process. This system is implemented as a web extension in order to increase the performance of the identification process and to decrease the overhead of the system caused when using a standalone software. © 2022 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Techniques to Detect Learning Disabilities Among children using Handwriting 3

Deep Learning Techniques to Detect Learning Disabilities Amo...

引用

3rd International Conference on Intelligent Computing, Instrumentation and Control Technologies, ICICICT 2022

作者： Vilasini, V. Banu Rekha, B. Sandeep, V. Charan Venkatesh, Vishnu Psg Institute of Technology and Applied Research Department of Computer Science and Engineering Tamil Nadu Coimbatore India Psg College of Technology Department of Biomedical Engineering Tamil Nadu Coimbatore India

ISBN: (数字)9781665410052

ISBN: (纸本)9781665410052

In today's world, we come across children facing certain disabilities which pose as obstacles and hinder their academic growth. Some of these disabilities are explicitly visible to the common eye, whereas some are hard to find and need extra attention. One such condition is Motor Dysgraphia which challenges an individual's ability to write. The common practice that is followed to identify such a condition among children is quite expensive and creates a mental strain on them. There are many intelligent computational methods that have been proposed with bearing a wide range of performances, however they are not quite standardized for assessment. Fortunately the advancements in Deep Learning techniques have been proven beneficial in automating this identification task. In this study, Learning Disability Detection system is built using Deep Learning techniques. The project's application is mainly focused on the pre-school and primary school children. This model analyses the child's handwriting and classifies whether the child is subjected to such a disorder or not. Deep Learning models - Convolutional Neural Networks (CNN) and Vision Transformers are adapted and their Disability Detection performances are analyzed and compared. © 2022 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Recurrent Autoencoder with Sequence-Aware Encoding 21st

Recurrent Autoencoder with Sequence-Aware Encoding

引用

21st International Conference on Computational science, ICCS 2021

作者： Susik, Robert Institute of Applied Computer Science Lódź University of Technology Lódź Poland

ISBN: (纸本)9783030779634

Recurrent Neural Networks (RNN) received a vast amount of attention last decade. Recently, the architectures of Recurrent AutoEncoders (RAE) found many applications in practice. RAE can extract the semantically valuable information, called context that represents a latent space useful for further processing. Nevertheless, recurrent autoencoders are hard to train, and the training process takes much time. This paper proposes a new recurrent autoencoder architecture with sequence-aware encoding (RAES), and its second variant which employs a 1D Convolutional layer (RAESC) to improve its performance and flexibility. We discuss the advantages and disadvantages of the solution and prove that the recurrent autoencoder with sequence-aware encoding outperforms a standard RAE in terms of model training time in most cases. The extensive experiments performed on a dataset of generated sequences of signals shows the advantages of RAES(C). The results show that the proposed solution dominates over the standard RAE, and the training process is the order of magnitude faster. © 2021, Springer Nature Switzerland AG.

关键词： Recurrent neural networks

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：