检索结果-内蒙古大学图书馆

Correntropy-Based Data Selective Adaptive Filtering

IEEE Transactions on Circuits and Systems I: Regular Papers 2024年第2期71卷 754-766页

作者： Chien, Ying-Ren Wu, Sheng-Teng Tsao, Hen-Wai Diniz, Paulo S.R. Adaptive and Autonomous Communication Laboratory Department of Electrical Engineering National Ilan University Yilan City260007 Taiwan College of Electrical Engineering and Computer Science Graduate Institute of Communication Engineering National Taiwan University Taipei10617 Taiwan Program of Electrical Engineering Department of Electronics and Computer Engineering COPPE/Poli Universidade Federal do Rio de Janeiro Rio de Janeiro68504 Brazil

Data selection can be used in conjunction with adaptive filtering algorithms to avoid unnecessary weight updating and thereby reduce computational overhead. This paper presents a novel correntropy-based data selection method as an alternative to conventional data selection mechanisms based on squared error values. We developed a variable correntropy sensing algorithm to maximize the instantaneous correntropy function for the Gaussian kernel function to mitigate the impact of impulse noise and other forms of noise that can be disregarded in data selection. The proposed data selection mechanism can be implemented with any adaptive filtering algorithm. In simulations, the proposed method (implemented with the least mean squared algorithm) outperformed comparable error-based data selection schemes in terms of hit rate and miss rate, and the resulting weight updating ratio was close to the expected weight updating ratio. © 2023 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.

关键词： Adaptive filters

来源：评论

学校读者我要写书评

暂无评论

Efficient Node Reduction Heuristic for GNN-Based Traffic Speed Forecasting 11

Efficient Node Reduction Heuristic for GNN-Based Traffic Spe...

引用

11th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, BDCAT 2024

作者： Inokuchi, Yuto Gonzalez, Pedro Henrique Aritsugi, Masayoshi Mendonca, Israel Kumamoto University Graduate School of Science and Technology Kumamoto Japan Federal University of Rio de Janeiro System Engineering and Computer Science Program Rio de Janeiro Brazil Kumamoto University Faculty of Advanced Science and Technology Kumamoto Japan

ISBN: (纸本)9798350367300

In this paper, we aim to reduce the number of nodes from Graph Neural Networks (GNNs), thereby simplifying models and reducing computational costs. GNNs are highly effective for various tasks, such as prediction, classification, and clustering, due to their ability to learn node and edge attributes and relationships, and they have been utilized for intelligent transportation systems recently by converting sensor networks into graph structures. Deep spatio-temporal neural networks, including Spatio-Temporal Graph Convolutional Networks (STGCNs), capture spatial and temporal dependencies, making them suitable for traffic speed forecasting, traffic demand prediction, and travel time estimation. Despite their success, GNNs face challenges in industrial applications due to significant memory usage and time consumption. In this paper, we propose a new approach to node reduction that outperforms existing methods in computational efficiency. Our experiments on two real-world traffic datasets demonstrate that using the heuristic and edge information to reduce nodes can cut computation time of optimization up to 95% and, by eliminating noise, can even enhance prediction accuracy. © 2024 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

PROM: A Phrase-level Copying Mechanism with Pre-training for Abstractive Summarization 30

PROM: A Phrase-level Copying Mechanism with Pre-training for...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Ma, Xinbei Gong, Yeyun He, Pengcheng Zhao, Hai Duan, Nan Department of Computer Science and Engineering Shanghai Jiao Tong University China Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University China Microsoft Research Asia China

ISBN: (纸本)9782493814104

Based on the remarkable achievements of pre-trained language models in abstractive summarization, the copying mechanism has proved helpful by improving the factuality, stability, and overall performance. This work proposes PROM, a new PhRase-level cOpying Mechanism that enhances attention on n-grams, which can be applied to zero-shot summarization with pre-training. PROM adds an indicator layer to explicitly pick up tokens in n-gram that can be copied from the source, and calculates an auxiliary loss for the copying prediction. Empirical studies show that PROM makes significant improvements in fine-tuning on benchmarks. In the zero-shot setting, PROM is utilized in the self-supervised pre-training on raw corpora and provides new general baselines on a wide range of summarization datasets. Further analysis shows that PROM performs more reasonable copying and contributes to faithfulness. Our code is publicly available at https://***/xbmxb/PROM. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Knowledge of cultural moral norms in large language models

arXiv

引用

arXiv 2023年

作者： Ramezani, Aida Xu, Yang Department of Computer Science University of Toronto Canada Department of Computer Science Cognitive Science Program University of Toronto Canada

Moral norms vary across cultures. A recent line of work suggests that English large language models contain human-like moral biases, but these studies typically do not examine moral variation in a diverse cultural setting. We investigate the extent to which monolingual English language models contain knowledge about moral norms in different countries. We consider two levels of analysis: 1) whether language models capture fine-grained moral variation across countries over a variety of topics such as "homosexuality" and "divorce";2) whether language models capture cultural diversity and shared tendencies in which topics people around the globe tend to diverge or agree on in their moral judgment. We perform our analyses with two public datasets from the World Values Survey (across 55 countries) and PEW global surveys (across 40 countries) on morality. We find that pre-trained English language models predict empirical moral norms across countries worse than the English moral norms reported previously. However, fine-tuning language models on the survey data improves inference across countries at the expense of a less accurate estimate of the English moral norms. We discuss the relevance and challenges of incorporating cultural knowledge into the automated inference of moral norms. © 2023, CC BY.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Towards Efficient and Diverse Generative Model for Unconditional Human Motion Synthesis 24

Towards Efficient and Diverse Generative Model for Unconditi...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Yu, Hua Liu, Weiming Bai, Jiapeng Gui, Xu Hou, Yaqing Ong, Yewsoon Zhang, Qiang Key Laboratory of Social Computing and Cognitive Intelligence Ministry of Education School of Computer Science and Technology Dalian University of Technology Dalian China College of Computer Science Zhejiang University Hangzhou China National and Local Joint Engineering Laboratory of Computer Aided Design School of Software Engineering Dalian University Dalian China College of Computing and Data Science Nanyang Technological University Singapore Singapore Yonsei University Seoul Korea Republic of

ISBN: (纸本)9798400706868

Recent generative methods have revolutionized the way of human motion synthesis, such as Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and Denoising Diffusion Probabilistic Models (DMs). These methods have gained significant attention in human motion fields. However, there are still challenges in unconditionally generating highly diverse human motions from a given distribution. To enhance the diversity of synthesized human motions, previous methods usually employ deep neural networks (DNNs) to train a transport map that transforms Gaussian noise distribution into real human motion distribution. According to Figalli's regularity theory, the optimal transport map computed by DNNs frequently exhibits discontinuities. This is due to the inherent limitation of DNNs in representing only continuous maps. Consequently, the generated human motions tend to heavily concentrate on densely populated regions of the data distribution, resulting in mode collapse or mode mixture. To address the issues, we propose an efficient method called MOOT for unconditional human motion synthesis. First, we utilize a reconstruction network based on GRU and transformer to map human motions to latent space. Next, we employ convex optimization to match the noise distribution with the latent space distribution of human motions through the Optimal Transport (OT) map. Then, we combine the extended OT map with the generator of reconstruction network to generate new human motions. Thereby overcoming the issues of mode collapse and mode mixture. MOOT generates a latent code distribution that is well-behaved and highly structured, providing a strong motion prior for various applications in the field of human motion. Through qualitative and quantitative experiments, MOOT achieves state-of-the-art results surpassing the latest methods, validating its superiority in unconditional human motion generation. © 2024 ACM.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Objective Sleep Quality Measurement based on Fuzzy Logic and Wearable Device 8

Objective Sleep Quality Measurement based on Fuzzy Logic and...

引用

8th International Conference on computer science and Computational Intelligence, ICCSCI 2023

作者： Wijaya, Jason A. Andersen, Pauw D. Surantha, Nico Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta11480 Indonesia Department of Electrical Electronic and Communication Engineering Tokyo City University Japan

Sleep is the natural state of relaxation for human being. Sleep quality is an essential yet frequently neglected aspect of sleep in general. Sleep quality is essential because it allows the body to restore itself and prepare for the next day. The standard method for evaluating sleep quality was subjective evaluation. Actigraphy devices, which can measure the sleep cycle, are now widely available. This study developed a method using Fuzzy Logic and an actigraphy device to measure and classify sleep quality. The fuzzy logic method was developed in several stages, which are determining the sleep quality measurement parameters, constructing the fuzzy set for each input variable, and developing the fuzzy rules. To evaluate the proposed fuzzy model, five individuals were invited to participate in the experiment and required to complete the PSQI subjective sleep questionnaire. The evaluation result shows that our proposed Fuzzy model achieves lower error compared to the existing method. © 2023 The Authors. Published by Elsevier B.V.

关键词： actigraphy artificial intelligences fuzzy logic sleep sleep quality

来源：评论

学校读者我要写书评

暂无评论

Evolution of moral semantics through metaphorization 44

Evolution of moral semantics through metaphorization

引用

44th Annual Meeting of the cognitive science Society: cognitive Diversity, CogSci 2022

作者： Ramezani, Aida Stellar, Jennifer E. Feinberg, Matthew Xu, Yang Department of Computer Science University of Toronto Canada Department of Psychology University of Toronto Canada Rotman School of Management University of Toronto Canada Department of Computer Science Cognitive Science Program University of Toronto Canada

Although language is critical to supporting morality within society, it is not clear how moral language itself evolved. We investigate the evolution of moral semantics, hypothesizing that words evolved to take on moral meanings from concrete experiences through metaphorization. We test this hypothesis by analyzing moral semantic change in words from the Moral Foundations Dictionary and the Historical Thesaurus of English over the past hundreds of years. In contrast with the observation that words become concrete over time, we demonstrate that moral words in the English lexicon undergo concrete-to-abstract shifts, reflecting systematic metaphorical mappings to the moral domain. Our results provide large-scale evidence for the role of metaphor in the historical development of the English moral lexicon. © 2022 The Author(s). This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY)

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

A Secure Cloud-Based Infrastructure for Virtual Sensors in IoT Environments 6

A Secure Cloud-Based Infrastructure for Virtual Sensors in I...

引用

6th International Conference of Women in Data science at Prince Sultan University, WiDS-PSU 2023

作者： Makda, Tahira Jehan Luiza Barros, Ana Dilek, Selma University of Doha for Science and Technology College of Business Management Doha Qatar University of Ceará Computer Science Program State Fortaleza Brazil Hacettepe University Department of Computer Engineering Ankara Turkey

ISBN: (纸本)9781665477239

Recent work has proven the effort of researchers to integrate small sensors and a cloud environment, delivering the Internet of Things (IoT). Sensors as a service are one of the leading research concerns in this context. Nevertheless, security is becoming one of the most significant attributes of the IoT as sensors become more human-independent and are being extensively used to monitor human lives. That way, IoT brings many key security challenges that need attention, some of which we address in this position paper. We present a cloud-based infrastructure that can deliver sensors and actuators as a service, providing secure communication between them and the control nodes on which IoT applications rely while implementing Big Data algorithms. For mapping our proposal, two scenarios related to Health Assistance are discussed, considering secure communications in a sensor network. In conclusion, we propose a scope for future research in this field considering digital twin concepts. Since domains exploiting IoT technologies can benefit from adopting Digital Twin, our goal is to evolve a virtual sensor and actuator system into this technology. © 2023 IEEE.

关键词： Actuators

来源：评论

学校读者我要写书评

暂无评论

Communicative Need Modulates Lexical Precision Across Semantic Domains: A Domain-Level Account of Efficient Communication 44

Communicative Need Modulates Lexical Precision Across Semant...

引用

44th Annual Meeting of the cognitive science Society: cognitive Diversity, CogSci 2022

作者： Bradford, Laurestine Thomas, Guillaume Xu, Yang Department of Linguistics McGill University Canada Department of Linguistics University of Toronto Canada Department of Computer Science Cognitive Science Program University of Toronto Canada

Different domains exhibit different degrees of lexical precision. Existing work has suggested that communicative need may modulate the precision of word meaning in individual domains. We extend this proposal across domains by asking why languages have more precise vocabulary in some domains than others. We hypothesize that lexical precision for a domain reflects how frequently speakers need to refer to it. We test this proposal using a cross-linguistic dataset of word-concept mappings for nine diverse domains from seven languages, and word frequencies from independent corpora. We find that the more frequent domains (except for kinship) tend to be more precise in every language, supporting a domain-level account of efficient communication on the precision of the lexicon. © 2022 The Author(s). This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY)

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

Automated classification of big X-ray diffraction data using deep learning models

引用

npj Computational Materials 2023年第1期9卷 169-180页

作者： Jerardo E.Salgado Samuel Lerman Zhaotong Du Chenliang Xu Niaz Abdolrahim Materials Science program University of RochesterRochesterNew York 14627USA Department of Computer Science University of RochesterRochesterNew York 14627USA Department of Mechanical Engineering University of RochesterRochesterNew York 14627USA Laboratory for Laser Energetics University of RochesterRochesterNew York 14627USA

In current in situ X-ray diffraction(XRD)techniques,data generation surpasses human analytical capabilities,potentially leading to the loss of *** techniques require human intervention,and lack the performance and adaptability required for material *** the critical need for high-throughput automated XRD pattern analysis,we present a generalized deep learning model to classify a diverse set of materials’crystal systems and space *** our approach,we generate training data with a holistic representation of patterns that emerge from varying experimental conditions and crystal *** also employ an expedited learning technique to refine our model’s expertise to experimental *** addition,we optimize model architecture to elicit classification based on Bragg’s Law and use evaluation data to interpret our model’s *** evaluate our models using experimental data,materials unseen in training,and altered cubic crystals,where we observe state-of-the-art performance and even greater advances in space group classification.

关键词： Automate classify cubic

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：