检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Chang, Kai Chieh Hasegawa-Johnson, Mark McElwain, Nancy L. Islam, Bashima Department of Electrical and Computer Engineering University of Illinois Urbana Champaign United States Beckman Institute for Advanced Science and Technology University of Illinois Urbana-Champaign United States Department of Human Development and Family Studies University of Illinois Urbana-Champaign United States Department of Electrical and Computer Engineering Worcester Polytechnic Institute United States

Infant sleep is critical to brain and behavioral development. Prior studies on infant sleep/wake classification have been largely limited to reliance on expensive and burdensome polysomnography (PSG) tests in the laboratory or wearable devices that collect single-modality data. To facilitate data collection and accuracy of detection, we aimed to advance this field of study by using a multi-modal wearable device, LittleBeats (LB), to collect audio, electrocardiogram (ECG), and inertial measurement unit (IMU) data among a cohort of 28 infants. We employed a 3-branch (audio/ECG/IMU) large scale transformer-based neural network (NN) to demonstrate the potential of such multi-modal data. We pretrained each branch independently with its respective modality, then finetuned the model by fusing the pretrained transformer layers with cross-attention. We show that multi-modal data significantly improves sleep/wake classification (accuracy = 0.880), compared with use of a single modality (accuracy = 0.732). Our approach to multi-modal mid-level fusion may be adaptable to a diverse range of architectures and tasks, expanding future directions of infant behavioral research. © 2023, CC BY.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

User-Driven Value Alignment: Understanding Users’ Perceptions and Strategies for Addressing Biased and Discriminatory Statements in AI Companions

arXiv

引用

arXiv 2024年

作者： Fan, Xianzhe Xiao, Qing Zhou, Xuhui Pei, Jiaxin Sap, Maarten Lu, Zhicong Shen, Hong Tsinghua University Beijing China Human-Computer Interaction Institute Carnegie Mellon University PittsburghPA United States Language Technologies Institute Carnegie Mellon University PittsburghPA United States Stanford University StanfordCA United States Department of Computer Science George Mason University FairfaxVA United States

Large language model-based AI companions are increasingly viewed by users as friends or romantic partners, leading to deep emotional bonds. However, they can generate biased, discriminatory, and harmful outputs. Recently, users are taking the initiative to address these harms and re-align AI companions. We introduce the concept of user-driven value alignment, where users actively identify, challenge, and attempt to correct AI outputs they perceive as harmful, aiming to guide the AI to better align with their values. We analyzed 77 social media posts about discriminatory AI statements and conducted semi-structured interviews with 20 experienced users. Our analysis revealed six common types of discriminatory statements perceived by users, how users make sense of those AI behaviors, and seven user-driven alignment strategies, such as gentle persuasion and anger expression. We discuss implications for supporting user-driven value alignment in future AI systems, where users and their communities have greater agency. Copyright © 2024, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

END-TO-END LYRICS RECOGNITION WITH SELF-SUPERVISED LEARNING

arXiv

引用

arXiv 2022年

作者： Zhang, Xiangyu Li, Shuyue Stella He, Zhanhong Togneri, Roberto Garcia, Leibny Paola Center for Language and Speech Processing Johns Hopkins University United States Human Language Technology Center of Excellence Johns Hopkins University United States Department of Computer Science University of Western Australia Australia

Lyrics recognition is an important task in music processing. Despite traditional algorithms such as the hybrid HMM-TDNN model achieving good performance, studies on applying end-to-end models and self-supervised learning (SSL) are limited. In this paper, we first establish an end-to-end baseline for lyrics recognition and then explore the performance of SSL models on lyrics recognition task. We evaluate a variety of upstream SSL models with different training methods (masked reconstruction, masked prediction, autoregressive reconstruction, and contrastive learning). Our end-to-end self-supervised models, evaluated on the DAMP music dataset, outperform the previous state-of-the-art (SOTA) system by 5.23% for the dev set and 2.4% for the test set even without a language model trained by a large corpus. Moreover, we investigate the effect of background music on the performance of self-supervised learning models and conclude that the SSL models cannot extract features efficiently in the presence of background music. Finally, we study the out-of-domain generalization ability of the SSL features considering that those models were not trained on music datasets. Copyright © 2022, The Authors. All rights reserved.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Performance measurement of processes and threads controlling, tracking and monitoring based on shared-memory parallel processing approach 3

Performance measurement of processes and threads controlling...

引用

3rd International Conference on Engineering Technology and its Applications, IICETA 2020

作者： Sharif, Karzan H. Zeebaree, Subhi R. M. Haji, Lailan M. Zebari, Rizgar R. University of Human Development Computer Science Department Sulaymaniyah Iraq Duhok Polytechnic University Duhok Iraq University of Zakho Computer Science Department KRG Iraq

ISBN: (纸本)9781728182315

In this paper, a professional integrated operating system performance measurement system is proposed, designed, and implemented in order to monitor processes and threads. Because monitoring itself is not enough, so this system provides full control of the CPU forcing, Priority-changing, and pausing-resuming-killing of these processes/threads. During these complex operations, it provides the ability to measure/compute the details of Total-execution-time, CPU-time, User-time, Kernel-time, context switching, CPU -usage, and the priority of processes/threads. The proposed system can be used with any architecture of multicore processors. Despite that measuring all these parameters in one system is very difficult and complex especially the Kernel-time and Context-switching, but all of them have been applied and determined successfully. Adding to the above features that never been integrated (by any previous work) in one system, and in order to illustrate the real abilities of this system, a shared-memory parallel processing approach has been proposed and applied. The proposed system prepared to run all possible cases of processes and threads that construct under-tested-program which may be one of the: Single-Process-Single-Thread, Single-Process-Multi- Thread, Multi-Process-Single-Thread, Multi-Process-Multi- Thread, and Multi- Process-Single-Multi-Thread. The algorithms of this system are designed and implemented via C++ programming language(using C++ Standard Library Thread) which is the nearest programming to the OSs provides a more increasing processing speed. © 2020 IEEE.

关键词： Software testing

来源：评论

学校读者我要写书评

暂无评论

Block Chain for Cloud Security: Enhancing Trust and Data Integrity in AI-Based Systems

Block Chain for Cloud Security: Enhancing Trust and Data Int...

引用

Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES), International Conference on

作者： Prakash Chandra Swain Kirti Mahajan Ruchira Rawat Bhavadharani K Pramod Kumar Patjoshi N. Balaraman School of Social Financial & Human Sciences Kalinga Institute of Industrial Technology (KIIT) Bhubaneswar Odisha India Institute of Management & Entrepreneurship Development Bharati Vidyapeeth Deemed University Pune Department of Computer Science & Engineering Graphic Era Deemed to be University Dehradun Uttarakhand India Department of Computer Science and Engineering VSB Engineering College Karur School of Management Centurion University of Technology and Management Department of Computer Science and Engineering St. Martins Engineering College Secunderabad Telangana

The ontology process in research methodology denoted the multiple usage of methodologies in a single study based on various emphases. The concept of multimodal interactions from the perspective of block chain technology in artificial intelligence has successfully secured the security of the cloud infrastructures of various business organizations. With the help of block chain along with the integration of artificial intelligence, large-scale business organizations have integrated server less cloud security. The overall success has provided trustworthiness to business organizations from the core perspectives of the block chain cloud.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing STEM Education with ChatGPT: A Case Study on Developing Practice Assessments for C Programming

Enhancing STEM Education with ChatGPT: A Case Study on Devel...

引用

International Conference on Engineering Education (ICEED)

作者： Shereen Ismail Jenice Lynch Runna Alghazo Merit Network Inc. University of Michigan MI USA School of Electrical Engineering and Computer Science University of North Dakota ND USA Department of Education Health & Behavioral Studies College of Education & Human Development University of North Dakota ND USA

ISBN: (数字)9798350367416

ISBN: (纸本)9798350367423

ChatGPT has proven its effectiveness as an assistance tool in the learning process for students. In this paper, an experiment was designed using ChatGPT to generate practice problems aligned with specific course objectives for a computer science (CS) introductory programming course, focusing on assessment questions in the form of short answer questions (SAQ) and multiple choice questions (MCQ). The paper aims to investigate how ChatGPT can assist educators in generating high quality assessment questions that align with course objectives. The evaluation is done by specialized educators. The findings demonstrate that ChatGPT offers valuable support for teachers in establishing coherent practice exam items that align with course objectives; however, caution must be used when utilizing ChatGPT to create assessment questions to ensure they are error-free and match the evaluation rubric.

关键词： Measurement Electronic learning Codes Focusing Chatbots Encoding Quality assessment Reliability Engineering education Programming profession

来源：评论

学校读者我要写书评

暂无评论

VaccineLies: A Natural language Resource for Learning to Recognize Misinformation about the COVID-19 and HPV Vaccines

arXiv

引用

arXiv 2022年

作者： Weinzierl, Maxwell A. Harabagiu, Sanda M. Human Language Technology Research Institute Department of Computer Science The University of Texas at Dallas 800 W. Campbell Rd. RichardsonTX75080 United States

Billions of COVID-19 vaccines have been administered, but many remain hesitant. Misinformation about the COVID-19 vaccines and other vaccines, propagating on social media, is believed to drive hesitancy towards vaccination. The ability to automatically recognize misinformation targeting vaccines on Twitter depends on the availability of data resources. In this paper we present VACCINELIES, a large collection of tweets propagating misinformation about two vaccines: the COVID-19 vaccines and the human Papillomavirus (HPV) vaccines. Misinformation targets are organized in vaccine-specific taxonomies, which reveal the misinformation themes and concerns. The ontological commitments of the Misinformation taxonomies provide an understanding of which misinformation themes and concerns dominate the discourse about the two vaccines covered in VACCINELIES. The organization into training, testing and development sets of VACCINELIES invites the development of novel supervised methods for detecting misinformation on Twitter and identifying the stance towards it. Furthermore, VACCINELIES can be a stepping stone for the development of datasets focusing on misinformation targeting additional vaccines. © 2022, CC BY.

关键词： Vaccines

来源：评论

学校读者我要写书评

暂无评论

Design of a Hybrid WDMA-Optical-CDMA over Multi-Mode Fiber Transmission System based on LG Modes for Short Haul-Local Area Network 1

Design of a Hybrid WDMA-Optical-CDMA over Multi-Mode Fiber T...

引用

1st International Recent Trends in Technology, Engineering and Computing Conference, IRTTEC 2020

作者： Ghazi, Alaan Aljunid, S.A. Idrus, Syed Zulkarnain Syed Fareed, Alaa Al-Dawoodi, Aras Mohsin, Aram Hewa School of Human Development and Techno Communication University Malaysia Perlis Malaysia School of Computer and Communication Engineering University Malaysia Perlis Malaysia School of Computing University Utara Malaysia Malaysia Computer Science Department College of Computer Science and Information Technology Kirkuk University Iraq Computer Technical Engineering Alkitab University Iraq Criminal Evidence Department Al-Qalam University College Iraq

Various optical communication technologies use Optical-Code Division Multiple Access (Optical-CDMA) as a channel access technique, which allows transmission of different users simultaneously over the optical communication channel. Therefore, Optical-CDMA permits free undue interference transmission of several users. Whereas Wavelength-division multiple access (WDMA) is a technique that can be implemented on how the optical channel can be divided into subchannels. This paper proposes a Hybrid WDMA-Optical-CDMA over multi-mode fiber (MMF) transmission system for short haul-local area network. The WDMA system is carried out by using four Laguerre-Gaussian (LG) modes that operate over four main wavelengths (1551 nm, 1550 nm, 1549 nm and 1548 nm). Additionally, the Optical-CDMA system is carried out by using one dimension Zero Cross-Correlation (ZCC) code to overcome multiple access interference (MAI) between the four ZCC codes. The performance of our system is evaluated based on eye diagrams, Q-Factor and bit-error rates (BER) measurements. © Published under licence by IOP Publishing Ltd.

关键词： Multiple access interference

来源：评论

学校读者我要写书评

暂无评论

LOD-connected offensive language ontology and tagset enrichment

LOD-connected offensive language ontology and tagset enrichm...

引用

2021 Workshops and Tutorials - language Data and Knowledge, LDK 2021

作者： Lewandowska-Tomaszczyk, Barbara Žitnik, Slavko Bączkowska, Anna Liebeskind, Chaya Mitrović, Jelena Oleskeviciene, Giedre Valunaite Department of Language and Communication State University of Applied Sciences Konin Poland University of Ljubljana Faculty for Computer and Information Science Večna pot 113 LjubljanaSI-1000 Slovenia Department of Glottodidactics and Natural Language Processing University of Gdansk Poland Department of Computer Science Jerusalem College of Technology Jerusalem Israel University of Passau Germany The Institute for Artificial Intelligence Research and Development of Serbia Serbia Institute of Humanities Mykolas Romeris University Ateities 20 VilniusLT-08303 Lithuania

The main focus of the paper is the definitional revision and enrichment of offensive language typology, making reference to publicly available offensive language datasets and testing them on available pre-trained lexical embedding systems. We review over 60 available corpora and compare tagging schemas applied there while making an attempt to explain semantic differences between particular concepts of the category OFFENSIVE in English. A finite set of classes that cover aspects of offensive language representation along with linguistically sound explanations is presented, based on the categories originally proposed by Zampieri et al. [1, 2] in terms of offensive language categorization schemata and tested by means of Sketch Engine tools on a large web-based corpus. The schemata are juxtaposed and discussed with reference to non-contextual word embeddings FastText, Word2Vec, and Glove. The methodology for mapping from existing corpora to a unified ontology as presented in this paper is provided. The proposed schema will enable further comparable research and effective use of corpora of languages other than English. It will also be applied in building an enriched tagset to be trained and used on new data, with the application of recently developed LLOD techniques [3]. © 2021 Copyright for this paper by its authors

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Noninvasive Tracking of Every Individual in Unmarked Mouse Groups Using Multi-Camera Fusion and Deep Learning

引用

Neuroscience Bulletin 2023年第6期39卷 893-910页

作者： Feng Su Yangzhen Wang Mengping Wei Chong Wang Shaoli Wang Lei Yang Jianmin Li Peijiang Yuan Dong-Gen Luo Chen Zhang Department of Neurobiology School of Basic Medical SciencesBeijing Key Laboratory of Neural Regeneration and RepairCapital Medical UniversityBeijing100069China Chinese Institute for Brain Research Beijing102206China State Key Laboratory of Translational Medicine and Innovative Drug Development Nanjing210000China Peking-Tsinghua Center for Life Sciences Academy for Advanced Interdisciplinary StudiesPeking UniversityBeijing100871China School of Life Sciences Tsinghua UniversityBeijing100084China School of Biological Science and Medical Engineering Beihang UniversityBeijing100191China The Key Laboratory of Developmental Genes and Human Disease Institute of Life SciencesSoutheast UniversityNanjing210096JiangsuChina Institute for Artificial Intelligence the State Key Laboratory of Intelligence Technology and SystemsBeijing National Research Center for Information Science and TechnologyDepartment of Computer Science and TechnologyTsinghua UniversityBeijing100084China School of Mechanical Engineering and Automation Beihang UniversityBeijing100191China

Accurate and efficient methods for identifying and tracking each animal in a group are needed to study complex behaviors and social *** tracking methods(e.g.,marking each animal with dye or surgically implanting microchips)can be invasive and may have an impact on the social behavior being *** overcome these shortcomings,video-based methods for tracking unmarked animals,such as fruit flies and zebrafish,have been ***,tracking individual mice in a group remains a challenging problem because of their flexible body and complicated interaction *** this study,we report the development of a multi-object tracker for mice that uses the Faster region-based convolutional neural network(R-CNN)deep learning algorithm with geometric transformations in combination with multi-camera/multi-image fusion *** system successfully tracked every individual in groups of unmarked mice and was applied to investigate chasing *** proposed system constitutes a step forward in the noninvasive tracking of individual mice engaged in social behavior.

关键词： Noninvasive tracking Deep learning Multi-camera Mouse group Social interaction

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：