检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Jiaming Yang Jiuxiang You Ziyue Qiu Guobo Xie Yi Yu Zhenguo Yang School of Computer Science Guangdong University of Technology Guangzhou China Graduate School of Advanced Science and Engineering Hiroshima University Hiroshima Japan

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

External knowledge provides critical clues for knowledge-based visual question answering (KB-VQA), while the implicit knowledge in images is difficult to capture in order to construct effective queries for knowledge bases. To this end, we propose a visual entity-centric prompting for knowledge retrieval (VEPR) to bridge the gap between the implicit and explicit knowledge driven by visual entities via large language models. More specifically, a visual entity question answering (EQ) module is devised to localize the critical entities from the given images and questions to generate entity-centric questions via a large language model. In particular, EQ obtains several entity-centric question-answer pairs via a visual language model. Furthermore, a question-answer-enhanced retrieval (ER) module is devised to construct a query by summarizing the text including question-answer pairs, captions and questions, in order to require explicit knowledge items. Finally, a multi-branch reader (MR) module is designed to encode the given questions, visual content and retrieved knowledge items, which are decoded to make answer predictions. Extensive experiments conducted on two public datasets demonstrate the effectiveness of the VEPR.

关键词： Visualization Large language models Knowledge based systems Signal processing Benchmark testing Question answering (information retrieval) Acoustics Speech processing

来源：评论

学校读者我要写书评

暂无评论

Repurposing transfer learning strategy of computer vision for owl sound classification 7

Repurposing transfer learning strategy of computer vision fo...

引用

7th International Conference on computer science and Computational Intelligence, CSCI 2022

作者： Gunawan, Kevin William Hidayat, Alam Ahmad Cenggoro, Tjeng Wawan Pardamean, Bens Computer Science Department BINUS Graduate Program-Master of Computer Science Bina Nusantara University Jakarta11480 Indonesia Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia

Accurately predicting an owl species based on its sound can be helpful for owl conservation. To build an accurate model for owl sound classification, deep learning is currently the most preferred algorithm, due to its excellent performance for modeling audio data. However, deep learning is generally underperformed for a small dataset, which is the case for recognizing scops owl sound. To overcome the issue, we proposed a transfer learning strategy, which is common for computer vision tasks, that can alleviate overfitting in a deep learning model for the owl sound classification. In our approach, we propose a neural network architecture consisting of the backbone of a EfficientNet model pre-Trained on the massive ImageNet database. The model takes the sound input that has been converted as two image representations: Spectrogram and Mel Frequency Cepstral Coefficients. Our strategy enables the use of a relatively small size of pre-Trained image classification model, which is widely available, for transfer learning in owl sound classification. Deploying the lightweight model in an automatic sound classifier provides a fast and accurate tool for various owl conservation purposes. © 2022 Elsevier B.V.. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

ODRL: A Benchmark for Off-Dynamics Reinforcement Learning 38

ODRL: A Benchmark for Off-Dynamics Reinforcement Learning

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Lyu, Jiafei Xu, Kang Xu, Jiacheng Yan, Mengbei Yang, Jingwen Zhang, Zongzhang Bai, Chenjia Lu, Zongqing Li, Xiu Tsinghua Shenzhen International Graduate School Tsinghua University China Tencent China National Key Laboratory for Novel Software Technology Nanjing University China China Telecom China School of Computer Science Peking University China

We consider off-dynamics reinforcement learning (RL) where one needs to transfer policies across different domains with dynamics mismatch. Despite the focus on developing dynamics-aware algorithms, this field is hindered due to the lack of a standard benchmark. To bridge this gap, we introduce ODRL, the first benchmark tailored for evaluating off-dynamics RL methods. ODRL contains four experimental settings where the source and target domains can be either online or offline, and provides diverse tasks and a broad spectrum of dynamics shifts, making it a reliable platform to comprehensively evaluate the agent's adaptation ability to the target domain. Furthermore, ODRL includes recent off-dynamics RL algorithms in a unified framework and introduces some extra baselines for different settings, all implemented in a single-file manner. To unpack the true adaptation capability of existing methods, we conduct extensive benchmarking experiments, which show that no method has universal advantages across varied dynamics shifts. We hope this benchmark can serve as a cornerstone for future research endeavors. Our code is publicly available at https://***/OffDynamicsRL/off-dynamics-rl. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

List Homomorphisms by Deleting Edges and Vertices: Tight Complexity Bounds for Bounded-Treewidth Graphs 32

List Homomorphisms by Deleting Edges and Vertices: Tight Com...

引用

32nd Annual European Symposium on Algorithms, ESA 2024

作者： Esmer, Barış Can Focke, Jacob Marx, Dániel Rzążewski, Pawel CISPA Helmholtz Center for Information Security Saarbrücken Germany Saarbrücken Graduate School of Computer Science Saarland Informatics Campus Germany Warsaw University of Technology Poland University of Warsaw Poland

ISBN: (纸本)9783959773386

The goal of this paper is to investigate a family of optimization problems arising from list homomorphisms, and to understand what the best possible algorithms are if we restrict the problem to bounded-treewidth graphs. Given graphs G, H, and lists L(v) ⊆ V (H) for every v ∈ V (G), a list homomorphism from (G, L) to H is a function f : V (G) → V (H) that preserves the edges (i.e., uv ∈ E(G) implies f(u)f(v) ∈ E(H)) and respects the lists (i.e., f(v) ∈ L(v)). The graph H may have loops. For a fixed H, the input of the optimization problem LHomVD(H) is a graph G with lists L(v), and the task is to find a set X of vertices having minimum size such that (G− X, L) has a list homomorphism to H. We define analogously the edge-deletion variant LHomED(H), where we have to delete as few edges as possible from G to obtain a graph that has a list homomorphism. This expressive family of problems includes members that are essentially equivalent to fundamental problems such as Vertex Cover, Max Cut, Odd Cycle Transversal, and Edge/Vertex Multiway Cut. For both variants, we first characterize those graphs H that make the problem polynomial-time solvable and show that the problem is NP-hard for every other fixed H. Second, as our main result, we determine for every graph H for which the problem is NP-hard, the smallest possible constant cH such that the problem can be solved in time ctH · nO(1) if a tree decomposition of G having width t is given in the input. Let i(H) be the maximum size of a set of vertices in H that have pairwise incomparable neighborhoods. For the vertex-deletion variant LHomVD(H), we show that the smallest possible constant is i(H) + 1 for every H: Given a tree decomposition of width t of G, LHomVD(H) can be solved in time (i(H)+1)t·nO(1). For any Ε > 0 and H, an (i(H) + 1 − Ε)t · nO(1) algorithm would violate the Strong Exponential-Time Hypothesis (SETH). The situation is more complex for the edge-deletion version. For every H, one can solve LHomED(H) in time

关键词： Polynomial approximation

来源：评论

学校读者我要写书评

暂无评论

A Blockchain-assisted Authentication Protocol for RFID-enabled Supply Chain Management System

引用

IEEE Transactions on Network science and Engineering 2025年

作者： Tariq, Tayyaba Obaidat, Mohammad S. Kuo, Wen-Chung Mahmood, Khalid Ayub, Muhammad Faizan Alenazi, Mohammed J.F National Yunlin University of Science and Technology Graduate School of Engineering Science and Technology Douliu64002 Taiwan The University of Jordan King Abdullah II School of Information Technology Amman11942 Jordan University of Science and Technology Beijing School of Communication and Computing Engineering Beijing100083 China Amity University Amity School of Engineering and Technology Noida201301 India National Yunlin University of Science and Technology Department of Computer Science and Information Engineering Douliu64002 Taiwan National Yunlin University of Science and Technology Graduate School of Intelligent Data Science Douliu64002 Taiwan University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu611731 China Riyadh11451 Saudi Arabia

In the dynamic domain of supply chain management, integrating Radio-Frequency Identification (RFID) technology with blockchain technology represents a significant leap forward. This integration has created a blockchain-assisted RFID-enabled Supply Chain Management System (SCMS). With its ability to utilize electromagnetic fields to identify and track tags attached to objects, RFID technology has revolutionized product management and tracking within supply chains. (SCMS) transmits tracking and product management information through public communication channels. However, (SCMS)'s communication on these channels is vulnerable to various security attacks and privacy challenges. Numerous authentication protocols have recently been proposed to tackle these security and privacy challenges. Unfortunately, most protocols are susceptible to different security attacks, such as tag or reader impersonation, denial of service, physical cloning, desynchronization attacks, etc. Therefore, we have proposed an authentication protocol for an RFID-enabled (SCMS) that also leverages blockchain technology. The integration of blockchain technology ensures data integrity, immutability and transparency across each department involved in (SCMS). In the proposed protocol, we also employ a Physically Unclonable Function (PUF) to secure (SCMS) against physical cloning attacks. We validate the security of the proposed protocol through both informal and formal analysis. The informal analysis confirms that our protocol significantly enhances security and efficiency. Moreover, a performance analysis of the proposed protocol against various competing existing protocols shows its superior performance. Notably, the proposed protocol substantially reduces computational and communication costs by 24.38% and 8.03%, respectively, which underscores its enhanced performance and resource efficiency. © 2013 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Cross-Modal Retrieval for Motion and Text via DropTriple Loss 5

Cross-Modal Retrieval for Motion and Text via DropTriple Los...

引用

5th ACM International Conference on Multimedia in Asia, MMAsia 2023

作者： Yan, Sheng Liu, Yang Wang, Haoqiang Du, Xin Liu, Mengyuan Liu, Hong School of Artificial Intelligence Chongqing University of Technology China College of Computer Science Sichuan University China Key Laboratory of Machine Perception Shenzhen Graduate School Peking University China

ISBN: (纸本)9798400702051

Cross-modal retrieval of image-text and video-text is a prominent research area in computer vision and natural language processing. However, there has been insufficient attention given to cross-modal retrieval between human motion and text, despite its wide-ranging applicability. To address this gap, we utilize a concise yet effective dual-unimodal transformer encoder for tackling this task. Recognizing that overlapping atomic actions in different human motion sequences can lead to semantic conflicts between samples, we explore a novel triplet loss function called DropTriple Loss. This loss function discards false negative samples from the negative sample set and focuses on mining remaining genuinely hard negative samples for triplet training, thereby reducing violations they cause. We evaluate our model and approach on the HumanML3D and KIT Motion-Language datasets. On the latest HumanML3D dataset, we achieve a recall of 62.9% for motion retrieval and 71.5% for text retrieval (both based on R@10). The source code for our approach is publicly available at https://***/eanson023/rehamot. © 2023 Copyright 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

Rotational Regrasping For a Stick-type Object Using Inertial Force Based on High-speed Swing-up Motion

Rotational Regrasping For a Stick-type Object Using Inertial...

引用

2023 IEEE International Conference on Robotics and Biomimetics, ROBIO 2023

作者： Senoo, Taku Konno, Atsushi Otsubo, Hayato Ishii, Idaku Hokkaido University Course of Systems Science and Informatic Graduate School of Information Science and Technology Kita 14 Nishi 9 Kita-ku Hokkaido Sapporo060-0814 Japan Hiroshima University Smart Innovation Program Graduate School of Advanced Science and Engineering 1-4-1 Kagamiyama Hiroshima Higashi-Hiroshima739-8527 Japan

ISBN: (纸本)9798350325706

In this paper, robotic regrasping is considered with the goal of achieving dexterous manipulation. The strategy using quick wrist snap is based on human regrasping, and involves rotating the grasped object due to the inertial force while maintaining the contact state. The posture of the grasped object is measured by high-speed vision in real time and is used for posture control. Experimental results are shown where regrasping of a stick-type object in the direction contrary to gravity can be achieved by a high-speed robot hand. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Internet of Things for Smart Furniture: A Systematic Review 11

Internet of Things for Smart Furniture: A Systematic Review

引用

11th International Conference on ICT for Smart Society, ICISS 2024

作者： Maulana, Fairuz Iqbal Kurniawan, Bambang Kartono Pramono, Andi Purnomo, Agung Abdulah, Muhamad Ezran Bin Zainal Mohamed, Mustafa School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia School of Design Bina Nusantara University Interior Design Department Jakarta11480 Indonesia Binus Business School Undergraduate Program Bina Nusantara University Entrepreneurship Department Jakarta11480 Indonesia Universiti Malaysia Kelantan Faculty of Creative Technology and Heritage Department of Design and Creative Technology Kelantan Bachok16300 Malaysia School of Technology of the Arts Republic Polytechnic Singapore

ISBN: (纸本)9798350389654

The research entitled 'Internet of Things for Smart Furniture: A Systematic Review' explores integrating IoT technology into smart furniture to improve functionality and user experience. By synthesizing existing literature, this research aims to discover how the Internet of Things (IoT) is leveraged to meet specific user needs with smart furniture and what hardware and sensor technologies are proposed for smart furniture design, considering factors such as scalability and need for users. Current research trends in securing user communications and data in the context of IoT-enabled smart furniture. This research involved a literature search which included journal articles and reviews from the Web of science (WOS) and the Scopus database (core collection) from 2014 to 2023, and the final results were 9 documents that would be reviewed. Several impactful means must be used to meet specific user needs, such as health monitoring, enhanced security, enhanced convenience and mobility, integration with smart home systems, and personalization and adaptability. Drawing insights from IoT data quality, machine learning applications in IoT, and smart furniture design for elderly care, this research provides a comprehensive picture of the current IoT-enabled smart furniture landscape. This research can explore existing IoT solutions for smart furniture, with a focus on cutting-edge technologies and design principles that can shape the future of this domain. © 2024 IEEE.

关键词： internet of things literature review scopus smart furniture web of science

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of SAAS Model and NPC Integration for Enhancing VR Shopping Experiences

引用

Applied sciences (Switzerland) 2024年第15期14卷

作者： Doungtap, Surasachai Wang, Jenq-Haur Phanichraksaphong, Varinya International Graduate Program of Electrical Engineering and Computer Science National Taipei University of Technology Taipei10608 Taiwan Department of Computer Science and Information Engineering National Taipei University of Technology Taipei10608 Taiwan

This article examines the incorporation of the Shopping Assistance Automatic Suggestion (SAAS) model into Virtual Reality (VR) environments in order to improve the online shopping experience. The SAAS model employs sophisticated deep learning methods to offer customized product recommendations, which are conveyed by non-player characters (NPCs) via voice-based interactions. Our goal is to develop an interactive shopping experience that replicates real-life interactions by integrating AI-powered recommendations with immersive VR technology. We gather and standardize data from several open commerce databases, such as Amazon Product and Customer Reviews. The SAAS model, in conjunction with GPT-3, BERT, and T5, undergoes training and testing to evaluate its effectiveness across multiple criteria. The results demonstrate that the SAAS model surpasses other models in delivering contextually aware and pertinent recommendations. The integration process outlines the specific steps involved in capturing, processing, and transforming user interactions in virtual reality (VR) into vocal suggestions provided by non-player characters (NPCs). This strategy improves customization and utilizes the immersive features of virtual reality to effectively engage people. The results of our research establish a higher standard for e-commerce, with the goal of enhancing the user experience of online purchasing by making it more instinctive, engaging, and pleasurable. © 2024 by the authors.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Emotion Recognition from EEG with Normalized Mutual Information and Convolutional Neural Network 12

Emotion Recognition from EEG with Normalized Mutual Informat...

引用

12th International Conference on Electrical and computer Engineering, ICECE 2022

作者： Maria, Mahfuza Akter Akhand, M.A.H. Shimamura, Tetsuya Khulna University of Engineering & Technology Department of Computer Science and Engineering Khulna Bangladesh Saitama University Graduate School of Science and Engineeringment Saitama Japan

ISBN: (纸本)9798350398793

Emotion is the fundamental trait of human beings, and brain signals are prospectus for emotion recognition (ER). Electroencephalography (EEG) is a preferable brain signal for ER as it is non-invasive, fast, portable and easy to use. Automatic ER from EEG is a challenging computational intelligence or machine learning task due to the inherited complexity of EEG signals. The aim of this study is to analyze different sub-bands (e.g., Alpha, Beta, and Gamma) of the EEG signals to find the most appropriate band to classify emotions using deep learning. At a glance, this study investigated main EEG signals (i.e., full frequency spectrum) and its sub-bands for connectivity feature map (CFM) construction using normalized mutual information (NMI);and convolutional neural network (CNN) was used for emotion classification from NMI CFMs. Experimental results identified that NMI CFMs from Gamma band showed relatively better emotion classification accuracy by CNN than CFMs with other bands. Finally, the proposed ER method with NMI and CNN is revealed as a potential EEG based ER method showing better than or competitive to existing related methods. © 2022 IEEE.

关键词： Electroencephalography

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：