检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Chen, Zeming Gao, Qiyue Department of Computer Science and Software Engineering Rose-Hulman Institute of Technology

Dependency parsing is a tool widely used in the field of Natural Language Processing and computational linguistics. However, there is hardly any work that connects dependency parsing to monotonicity, which is an essential part of logic and linguistic semantics. In this paper, we present a system that automatically annotates monotonicity information based on Universal Dependency parse trees. Our system utilizes surface-level monotonicity facts about quantifiers, lexical items, and token-level polarity information. We compared our system's performance with existing systems in the literature, including NatLog and ccg2mono, on a small evaluation dataset. Results show that our system outperforms NatLog and ccg2mono. © 2021, CC BY.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Secure AI and Blockchain-enabled Framework in Smart Vehicular Networks

Secure AI and Blockchain-enabled Framework in Smart Vehicula...

引用

2021 IEEE Globecom Workshops, GC Wkshps 2021

作者： Rabieinejad, Elnaz Yazdinejad, Abbas Dehghantanha, Ali Parizi, Reza M. Srivastava, Gautam K.N.Toosi University of Technology Information Technology Department Tehran Iran University of Guelph Cyber Science Lab School of Computer Science Ontario Canada Kennesaw State University College of Computing and Software Engineering GA United States Brandon University Department of Mathematics and Computer Science Manitoba Canada

ISBN: (纸本)9781665423908

In recent years, Internet of Things (IoT) devices such as drones, smartphones, and smart vehicles have increased, and Smart Vehicular Networks (SVNs) have formed. SVN are one of the vital components in smart cities and improves their functionality and efficiency. Despite the many applications that SVN have shown, they have security vulnerabilities issues, and there are many reports of them being hacked. Most attacks on SVN occur due to wireless communications and information sharing between vehicles. Also, due to the communication among vehicles, failures can spread from the victim device to the entire network and cause widespread damages. SVN features include mobility, decentralization, resource constraints that make traditional security solutions unsuitable for it. We propose a secure framework using blockchain and Deep Neural networks (DNN) to address these challenges. In this framework, we consider cluster-based architecture that vehicles in each cluster can securely communicate using the blockchain. Also, DNN is adopted to detect abnormal vehicles that have been attacked using their network traffic analysis in each zone. We assess blockchain impacts on throughput using different miners with multiple block sizes in our proposed framework. The experimental result indicates the throughput improvement with an increase in miners and block size. In addition, we evaluated the DNN performance for abnormal vehicle detection, which represents 99.82% in terms of accuracy. © 2021 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

On the robustness of median sampling in noisy evolutionary optimization

引用

science China(Information sciences) 2021年第5期64卷 32-44页

作者： Chao BIAN Chao QIAN Yang YU Ke TANG State Key Laboratory for Novel Software Technology Nanjing University Shenzhen Key Laboratory of Computational Intelligence Department of Computer Science and EngineeringSouthern University of Science and Technology

Evolutionary algorithms(EAs) are a sort of nature-inspired metaheuristics, which have wide applications in various practical optimization problems. In these problems, objective evaluations are usually inaccurate, because noise is almost inevitable in real world, and it is a crucial issue to weaken the negative effect caused by noise. Sampling is a popular strategy, which evaluates the objective a couple of times, and employs the mean of these evaluation results as an estimate of the objective value. In this work, we introduce a novel sampling method, median sampling, into EAs, and illustrate its properties and usefulness theoretically by solving OneMax, the problem of maximizing the number of 1 s in a bit string. Instead of the mean, median sampling employs the median of the evaluation results as an estimate. Through rigorous theoretical analysis on OneMax under the commonly used onebit noise, we show that median sampling reduces the expected runtime exponentially. Next, through two special noise models, we show that when the 2-quantile of the noisy fitness increases with the true fitness, median sampling can be better than mean sampling; otherwise,it may fail and mean sampling can be better. The results may guide us to employ median sampling properly in practical applications.

关键词： evolutionary algorithms noisy optimization median sampling computational complexity runtime analysis

来源：评论

学校读者我要写书评

暂无评论

Gradient Imitation Reinforcement Learning for General Low-Resource Information Extraction

arXiv

引用

arXiv 2022年

作者： Hu, Xuming Meng, Shiao Zhang, Chenwei Yang, Xiangli Wen, Lijie King, Irwin Yu, Philip S. The School of Software Tsinghua University Beijing China The Amazon Seattle United States The School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China The Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong The Department of Computer Science University of Illinois at Chicago Chicago United States

Information Extraction (IE) aims to extract structured information from heterogeneous sources. IE from natural language texts include sub-tasks such as Named Entity Recognition (NER), Relation Extraction (RE), and Event Extraction (EE). Most IE systems require comprehensive understandings of sentence structure, implied semantics, and domain knowledge to perform well;thus, IE tasks always need adequate external resources and annotations. However, it takes time and effort to obtain more human annotations. Low-Resource Information Extraction (LRIE) strives to use unsupervised data, reducing the required resources and human annotation. In practice, existing systems either utilize self-training schemes to generate pseudo labels that will cause the gradual drift problem, or leverage consistency regularization methods which inevitably possess confirmation bias. To alleviate confirmation bias due to the lack of feedback loops in existing LRIE learning paradigms, we develop a Gradient Imitation Reinforcement Learning (GIRL) method to encourage pseudo-labeled data to imitate the gradient descent direction on labeled data, which can force pseudo-labeled data to achieve better optimization capabilities similar to labeled data. Based on how well the pseudo-labeled data imitates the instructive gradient descent direction obtained from labeled data, we design a reward to quantify the imitation process and bootstrap the optimization capability of pseudo-labeled data through trial and error. In addition to learning paradigms, GIRL is not limited to specific sub-tasks, and we leverage GIRL to solve all IE sub-tasks (named entity recognition, relation extraction, and event extraction) in low-resource settings (semi-supervised IE and few-shot IE). Experimental results on seven public datasets across three IE sub-tasks demonstrate the effectiveness of GIRL on low-resource information extraction when comparing with strong baselines. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Multi-block-single-probe variance reduced estimator for coupled compositional optimization 22

Multi-block-single-probe variance reduced estimator for coup...

引用

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Wei Jiang Gang Li Yibo Wang Lijun Zhang Tianbao Yang National Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science the University of Iowa Iowa City Department of Computer Science and Engineering Texas A&M University College Station

ISBN: (纸本)9781713871088

Variance reduction techniques such as SPIDER/SARAH/STORM have been extensively studied to improve the convergence rates of stochastic non-convex optimization, which usually maintain and update a sequence of estimators for a single function across iterations. What if we need to track multiple functional mappings across iterations but only with access to stochastic samples of O(1) functional mappings at each iteration? There is an important application in solving an emerging family of coupled compositional optimization problems in the form of ∑mi=1 fi(gi(w)), where gi is accessible through a stochastic oracle. The key issue is to track and estimate a sequence of g(w) = (g1(w),..., gm(w)) across iterations, where g(w) has m blocks and it is only allowed to probe O(1) blocks to attain their stochastic values and Jacobians. To improve the complexity for solving these problems, we propose a novel stochastic method named Multi-block-Single-probe Variance Reduced (MSVR) estimator to track the sequence of g(w). It is inspired by STORM but introduces a customized error correction term to alleviate the noise not only in stochastic samples for the selected blocks but also in those blocks that are not sampled. With the help of the MSVR estimator, we develop several algorithms for solving the aforementioned compositional problems with improved complexities across a spectrum of settings with non-convex/convex/strongly convex/Polyak-Łojasiewicz (PL) objectives. Our results improve upon prior ones in several aspects, including the order of sample complexities and dependence on the strong convexity parameter. Empirical studies on multi-task deep AUC maximization demonstrate the better performance of using the new estimator.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Trust Assessment System for Online Social Networking Environment Using Learning Assisted Classification Model

A Novel Trust Assessment System for Online Social Networking...

引用

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

作者： S. Nithya D. Deepa Vunnava Dinesh Babu Pallavi G Hamed Fawareh R. D. Kayalvizhy Department of Computer Science Engineering Saveetha Engineering College Chennai Department of CSE St. Joseph's College of Engineering Chennai Department of CSE RV Institute of Technology Guntur Department of Civil Engineering Laxminarayan Innovation Technological University Nagpur India Software Engineering Dept Faculty of Information Technology Zarqa University Zarqa Jordan University of Business and Technology Jeddah Saudi Arabia Department of Electronics and Communication Engineering R.M.K. Engineering College Kavaraipettai Tamil Nadu

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543631

Trust is essential in the constantly developing environment of online social networks. The Nexus Terroism Impact and Intro Trust facilitate the communication, decision-making, and maintenance of digital communities among individuals. Trust-based Learning with Classification Principle (TLCP) is introduced in this research to improve the assessment of trust in social networks. To ascertain its efficacy, our proposed methodology undergoes a 10-fold cross-validation against the Trust Assessment Methodology (TAM). The traditional Trust Assessment Model (TAM) evaluates and regulates trust in Internet social networking. This method incorporates trust assessment factors to generate a user dependability judgment that is both discriminatory and dynamic. The Trustworthy Account Management (TAM) challenge is designed to enhance digital trust by addressing disinformation, deception, and hostile conduct. The TLCP is able to accurately evaluate trust by utilizing reputation scoring, community analysis, and user behavior metrics. Trust levels are dynamically adjusted as a result of the identification of trustworthy behavioral patterns and danger indicators by machine learning algorithms and social media data. This research examines the connections between trust management and digital interactions and suggests a general solution to alleviate specific trust evaluation issues across online social networks.

关键词： Measurement Machine learning algorithms Social networking (online) Privacy breach Real-time systems Hazards Maintenance Internet Fake news Trust management

来源：评论

学校读者我要写书评

暂无评论

EEG-Based VR Scene Adaptive Generation System for Regulating Emotion

EEG-Based VR Scene Adaptive Generation System for Regulating...

引用

Virtual Reality (ICVR), IEEE International Conference on

作者： Hui Liang ShiQing Liu Yi Wang Junjun Pan Jialin Fu Yingkai Yuan Department of Software Engineering Zhengzhou University of Lightindustry Zheng zhou China Department of Rehabilitation General Hospital of Pingmei Shenma Group Pingdingshan China Department of Computer Science and Engineering Beihang University Peking China College of Economics and Management Zhengzhou University of LightIndustry Zheng zhou China College of Computer and Communication Engineering Zhengzhou University of LightIndustry Zheng zhou China

Research based on electroencephalogram (EEG) and virtual reality (VR) has developed rapidly in recent years. The integration of EEG and VR technology has been broadly employed in psychological intervention to treat various psychological conditions and alleviate stress. At present, there exist commercial-level EEG emotion recognition products, however, research on the correlation between the content of VR scenarios and emotions remains unavailable. Therefore, this work investigates VR scene construction based on EEG, which incorporates the recovery environment theory and virtual reality technology to construct different three-dimensional dynamic virtual reality interaction scenes. EEG equipment constructs an emotion-adaptive VR system based on EEG signals with the aid of VR interaction equipment. Herein, EEG data feedback was applied to automatically generate VR scenes, and the experimental findings revealed that the system has certain feasibility and application value.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Collaborative Multi-Agent Reinforcement Learning for Automated Feature Transformation with Graph-Driven Path Optimization

arXiv

引用

arXiv 2025年

作者： Huang, Xiaohan Wang, Dongjie Ning, Zhiyuan Qiao, Ziyue Long, Qingqing Zhu, Haowei Du, Yi Wu, Min Zhou, Yuanchun Xiao, Meng Computer Network Information Center Chinese Academy of Sciences University of the Chinese Academy of Sciences China Computer Network Information Center Chinese Academy of Sciences China Department of Electrical Engineering and Computer Science University of Kansas United States School of Computing and Information Technology Great Bay University Dongguan China School of Software Tsinghua University China Institute for Infocomm Research Agency for Science Technology and Research Singapore

Feature transformation methods aim to find an optimal mathematical feature-feature crossing process that generates high-value features and improves the performance of downstream machine learning tasks. Existing frameworks, though designed to mitigate manual costs, often treat feature transformations as isolated operations, ignoring dynamic dependencies between transformation steps. To address the limitations, we propose TCTO, a collaborative multi-agent reinforcement learning framework that automates feature engineering through graph-driven path optimization. The framework’s core innovation lies in an evolving interaction graph that models features as nodes and transformations as edges. Through graph pruning and backtracking, it dynamically eliminates low-impact edges, reduces redundant operation, and enhances exploration stability. This graph also provides full traceability to empower TCTO to reuse high-utility subgraphs from historical transformations. To demonstrate the efficacy and adaptability of our approach, we conduct comprehensive experiments and case studies, which show superior performance across a range of datasets. Copyright © 2025, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Deep learning and artificial intelligence: Project collaboration across classes

Deep learning and artificial intelligence: Project collabora...

引用

2020 ASEE Virtual Annual Conference, ASEE 2020

作者： Kurfess, Franz J. Pantoja, Maria Humer, Irene Computer Science and Software Engineering Department California Polytechnic State University San Luis Obispo United States CalPoly San Luis Obispo Computer Engineering Computer Science and Software Engineering Office: 14-211 United States

Working in collaborative environments is an essential skill for computing professionals. In our program, students have significant team experience from previous classes;almost all of our classes in Cal Poly's computer science, software engineering and computer engineering programs have lab sections, and students start with team-based work early. However, they mostly work within small teams of about five members that are stable throughout a term. To stretch the students' collaborative skills and enhance their experience with more fluid team configurations, we experimented with creating teams across two related senior-level classes within computer science, but with different perspectives and approaches. For one particular project, shark spotting in drone video footage, the teams also collaborated with a group of Marine Biologists from the Shark Lab at CSU Long Beach, who provided the video material and acted as a customer for three teams across the two classes. This exposed our students to collaborators among different fields, with their own terminology, goals, work methods and practical approaches. Our paper reports on the initial experiment during the Fall 2019 term, involving two sections of an Artificial Intelligence class and one section of a Deep Learning class. We are planning to continue this collaboration in the future. © American Society for engineering Education 2020.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

An ensemble neural network for the emotional classification of text 33

An ensemble neural network for the emotional classification ...

引用

33rd International Florida Artificial Intelligence Research Society Conference, FLAIRS 2020

作者： Youngquist, Oscar Department of Computer Science and Software Engineering Rose-Hulman Institute of Technology United States

ISBN: (纸本)9781577358213

In this work, we propose a novel ensemble neural network design that is capable of classifying the emotional context of short sentences. Our model consists of three distinct branches, each of which is composed of a combination of recurrent, convolutional, and pooling layers to capture the emotional context of text. Our unique combination of convolutional and recurrent layers enables our network to extract more emotionally salient information from text than formerly possible. Using this network, experiments classifying the emotional context of short sequences of texts from five distinct datasets, were conducted. Results show that the novel method outperforms all historical approaches across all datasets by 8.31 percentage points on average. Additionally, the proposed work produces results that are on average as accurate as state of the art methods, while using two orders of magnitude less training data. The contribution of this paper is a novel ensemble recurrent convolutional neural network capable of detecting and classifying the emotional context of short sentences. © FLAIRS *** right reserved.

关键词： Convolutional 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：