检索结果-内蒙古大学图书馆

2022 Findings of the Association for computational Linguistics: EMNLP 2022

作者： An, Zhenwei Huang, Quzhe Jiang, Cong Feng, Yansong Zhao, Dongyan Wangxuan Institute of Computer Technology Peking University China School of Software & Microelectronics Peking University China School of Intelligence Science and Technology Peking University China Peking University Law School China Institute for Artificial Intelligence Peking University China The MOE Key Laboratory of Computational Linguistics Peking University China

The charge prediction task aims to predict the charge for a case given its fact description. Recent models have already achieved impressive accuracy in this task, however, little is understood about the mechanisms they use to perform the judgment. For practical applications, a charge prediction model should conform to the certain legal theory in civil law countries, as under the framework of civil law, all cases are judged according to certain local legal theories. In China, for example, nearly all criminal judges make decisions based on the Four Elements Theory (FET). In this paper, we argue that trustworthy charge prediction models should take legal theories into consideration, and standing on prior studies in model interpretation, we propose three principles for trustworthy models should follow in this task, which are sensitive, selective, and presumption of innocence. We further design a new framework to evaluate whether existing charge prediction models learn legal theories. Our findings indicate that, while existing charge prediction models meet the selective principle on a benchmark dataset, most of them are still not sensitive enough and do not satisfy the presumption of innocence. Our code and dataset are released at https://***/ZhenweiAn/EXP_LJP. © 2022 Association for computational Linguistics.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

A Two-Step Approach for Trend Detection in Injection Molding: Mann-Kendall Test and Decision Trees 6

A Two-Step Approach for Trend Detection in Injection Molding...

引用

6th International Conference on Industry 4.0 and Smart Manufacturing, ISM 2024

作者： Duy, Du Nguyen Himmelbauer, Johannes Software Competence Center Hagenberg Austria

In the highly complex and dynamic environment of injection molding, detecting trends in process parameters is crucial for maintaining product quality and optimizing manufacturing efficiency. While the Mann-Kendall test, a widely applied trend detection technique, could identify statistically significant trends, it inherently lacks the capacity to determine their practical relevance. In order to address this issue, the present work proposes a novel two-step approach combining the test with machine learning. After the Mann-Kendall test identifies potential trends in the first step, a Decision Trees model, trained on deliberately engineered features, will be utilized in the second step to filters for those with practical implications. To evaluate the capability of the proposed framework, empirical experiments were carried out on real injection pressure data. The results showcase the method's effectiveness in detecting relevant trends, highlighting its potential for proactive process monitoring and control in manufacturing. © 2024 The Authors. Published by Elsevier B.V.

关键词： Process control

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Graph Transfer Learning for Category-aware Cross-Domain Sequential Recommendation 25

Heterogeneous Graph Transfer Learning for Category-aware Cro...

引用

34th ACM Web Conference, WWW 2025

作者： Xu, Zitao Chen, Xiaoqing Pan, Weike Ming, Zhong College of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen University Shenzhen Technology University Shenzhen China

ISBN: (纸本)9798400712746

Cross-domain sequential recommendation (CDSR) is proposed to alleviate the data sparsity issue while capturing users’ sequential preferences. However, most existing methods do not explore the item transition patterns across different domains and can also not be applied to a multi-domain scenario. Moreover, previous methods rely on overlapping users as bridges to transfer knowledge, which struggles to capture the complex associations across domains without sufficient overlapping users. In this paper, we introduce item attributes into CDSR, and propose a heterogeneous graph transfer learning method to address these issues. Specifically, we construct a cross-domain heterogeneous graph to allow the association of user, item, and category nodes from different domains, and enhance the flexibility of the model by enabling message propagation between more nodes through edge expansion based on the semantic similarity and co-occurrence probability. In addition, we devise meta-paths from different perspectives for nodes at item, user and category levels to guide information aggregation, which can transfer knowledge across domains and reduce the reliance on the number of overlapping users. We further design attention modules to capture users’ dynamic preferences from the item sequences they have interacted with in each domain, and explore the transition patterns within category sequences which reflect users’ coarse-grained preferences. Finally, we perform knowledge transfer across different domains, and predict the most likely items that users will interact with in each domain. Extensive empirical studies on three real-world datasets indicate that our HGTL significantly outperforms the state-of-the-art baselines in all cases. © 2025 Copyright held by the owner/author(s).

关键词： Transfer learning

来源：评论

学校读者我要写书评

暂无评论

parallel Electromagnetic Simulations based on Novel Refinement Templates of AMR

Parallel Electromagnetic Simulations based on Novel Refineme...

引用

International Symposium on Antennas, Propagation and EM Theory (ISAPE)

作者： Weijie Wang Jingjie Xu Yannan Liu Shaoliang Hu Software Center for High Performance Numerical Simulation Institute of Applied Physics and Computational Mathematics CAEP Beijing China

ISBN: (数字)9798350353129

ISBN: (纸本)9798350353136

A new adaptive refinement method and its parallel implementation of adaptive mesh refinement is developed by regrouping the tetrahedra into 7 refinement templates. In the parallel refinement process, our method only transfers nonconformity messages and refining requirements between corresponding submeshes. The communication algorithm is utilized to update shared information, eliminating the need for a global index in the refinement process. Numerical experiments of a printed dipole antenna, patch antenna, and printed circuit board are presented to validate the accuracy of our new parallel refinement method.

关键词： Dipole antennas Antenna theory Patch antennas computational modeling Printed circuits Refining Numerical models Indexes Integrated circuit modeling Electromagnetics

来源：评论

学校读者我要写书评

暂无评论

An Empirical Study on Kubernetes Operator Bugs 2024

An Empirical Study on Kubernetes Operator Bugs

引用

33rd ACM SIGSOFT International Conference on software Testing and Analysis (ISSTA)

作者： Xu, Qingxin Gao, Yu Jun Wei Chinese Acad Sci Univ Chinese Acad Sci Inst Software Key Lab Syst Software Beijing Peoples R China

ISBN: (纸本)9798400706127

Kubernetes is the leading cluster management platform, and within Kubernetes, an operator is an application-specific program that leverages the Kubernetes API to automate operation tasks for managing an application deployed on a Kubernetes cluster. Users can declare a desired state for the managed cluster, specifying their configuration preferences. The operator program is responsible for reconciling the cluster's actual state to align with the desired state. However, the complex, dynamic, and distributed nature of the overall system can introduce operator bugs, and lead to severe consequences, e.g., outages and undesired cluster state. In this paper, we conduct the first comprehensive study on 210 operator bugs from 36 Kubernetes operators. For all the studied bugs, we investigate their root causes, manifestations, impacts and fixing. Our study reveals many interesting findings that can guide the detection and testing of operator bugs, as well as the development of more reliable operators.

关键词： Kubernetes operation operator bugs empirical study

来源：评论

学校读者我要写书评

暂无评论

A Static Bi-dimensional Sample Selection for Federated Learning with Label Noise 28th

A Static Bi-dimensional Sample Selection for Federated Lear...

引用

28th International Conference on Database Systems for Advanced Applications, DASFAA 2023

作者： Rong, Qian Yuan, Ling Li, Guohui Li, Jianjun Zhang, Lu Ding, Xuanang School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China School of Software Engineering Huazhong University of Science and Technology Wuhan China School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9783031306365

In real-world Federated learning(FL), client training data may contain label noise, which can harm the generalization performance of the global model. Most existing noisy label learning methods rely on sample selection strategies that treat small-loss samples as correctly labeled ones, and large-loss samples as mislabeled ones. However, large-loss samples may also be valuable hard samples with correct labels. Also, these sample selection strategies have an issue in FL setting: Moreover sample selection strategies require training two models simultaneously and are executed in every mini-batch, which increases the communication cost and client computation. In this paper, we propose an efficient multi-stage federated learning framework to tackle label noise. Firstly, we use self-supervision pre-training to extract category-independent features without using any labels. Secondly, we propose a federated static two-dimensional sample selection(FedSTSS) method, which uses category probability entropy and loss as separation metrics to identify samples. Notably, to improve the accuracy of recognition samples, we add a weight entropy minimization term to the cross-entropy loss function. Finally, we use a semi-supervised method to finetune the global model on identified clean samples and mislabeled samples. Extensive experiments on multiple synthetic and real-world noisy datasets demonstrate that our method outperforms the state-of-the-art methods. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

A New Method of Search Engine Optimization Based on Semantic Kernel Idea

A New Method of Search Engine Optimization Based on Semantic...

引用

作者： Orekhov, Sergey Godlevsky, Michail Malyhon, Hennadiy Goncharenko, Tetiana Computer Science and Software Engineering Department National Technical University Kharkiv Polytechnical Institute Kharkiv61002 Ukraine

The results of more than 30 projects in the field of search engine optimization on the Internet show that the effectiveness of existing approaches is catastrophically reduced. Paid promotion methods no longer guarantee a banal increase in the traffic of a given web resource, and even more so the number of leads. The reason is that competition has moved to the Internet as much as possible. So, we need to look for new alternative methods. Therefore, the paper proposes to consider a new alternative way to promote goods on the Internet. This approach is a new information technology – virtual promotion. It is based on the creation of a virtual image of a product on the Internet. We called this virtual image the semantic kernel. Such a kernel is placed on the Internet in order to attract a potential buyer. It is placed in the Internet nodes according to a special online sales scheme, which is based on a map of the user's travels on the Internet. This scheme is called the virtual promotion map. The map must guarantee maximum traffic to a given web resource. However, it will not be possible to immediately build a map. Therefore, its creation is iterative. We are, as it were, setting ourselves up to find effective ways between the semantic core and the user who is ready to buy our product online. The article offers a description of the virtual promotion technology. The main idea of virtual promotion is that promotion is a logistical channel between a company and a client. We put a special message (semantic core) into the channel. He moves through the logistics channel and tries to attract as many new buyers of the goods as possible. This article presents the results of the work of special software that implements the synthesis, quality assessment and management of the semantic kernel of web content. The program generates and manages two channels: the first is intended for the distribution of the semantic kernel, the second is for assessing the quality of such distribution. T

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Efficient Training Approaches for Performance Anomaly Detection Models in Edge Computing Environments

arXiv

引用

arXiv 2024年

作者： Fernando, Duneesha Rodriguez, Maria A. Arroba, Patricia Ismail, Leila Buyya, Rajkumar Laboratory School of Computing and Information Systems The University of Melbourne Parkville Australia The CCS Center for Computational Simulation Universidad Politécnica de Madrid Madrid Spain Lab Department of Computer Science and Software Engineering United Arab Emirates University United Arab Emirates

Microservice architectures are increasingly used to modularize IoT applications and deploy them in distributed and heterogeneous edge computing environments. Over time, these microservice-based IoT applications are susceptible to performance anomalies caused by resource hogging (e.g., CPU or memory), resource contention, etc., which can negatively impact their Quality of Service and violate their Service Level Agreements. Existing research on performance anomaly detection for edge computing environments focuses on model training approaches that either achieve high accuracy at the expense of a time-consuming and resource-intensive training process or prioritize training efficiency at the cost of lower accuracy. To address this gap, while considering the resource constraints and the large number of devices in modern edge platforms, we propose two clustering-based model training approaches: (1) intra-cluster parameter transfer learning-based model training (ICPTL) and (2) cluster-level model training (CM). These approaches aim to find a trade-off between the training efficiency of anomaly detection models and their accuracy. We compared the models trained under ICPTL and CM to models trained for specific devices (most accurate, least efficient) and a single general model trained for all devices (least accurate, most efficient). Our findings show that ICPTL's model accuracy is comparable to that of the model per device approach while requiring only 40% of the training time. In addition, CM further improves training efficiency by requiring 23% less training time and reducing the number of trained models by approximately 66% compared to ICPTL, yet achieving a higher accuracy than a single general model. Copyright © 2024, The Authors. 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

ST-FVGAN: filling series traffic missing values with generative adversarial network

引用

TRANSPORTATION LETTERS-THE INTERNATIONAL JOURNAL OF TRANSPORTATION RESEARCH 2022年第4期14卷 407-415页

作者： Yang, Bing Kang, Yan Yuan, Yaoyao Li, Hao Wang, Fei Yunnan Univ Sch Software Dept Software Engn Kunming 650091 Yunnan Peoples R China

The imputation of time series traffic flow data is of great significance to the intelligent transportation, urban planning, and road emergency handling. This paper proposes a filling missing time series traffic data with Generative Adversarial Network (ST-FVGAN), which not only considers the spatio-temporal correlation and utilizes the idea of data generating of the Generative Adversarial Network, but also considers the external factors and introduces a more comprehensive loss function. Specifically, the model firstly constructs a Generator network which is composed of convolutional layer, residual block, and pixelshuffle block for the better potential distribution of the existing data, and then use the Discriminator network for the input judging. Experiments are conducted on the open-source TaxiBJ GPS dataset, and evaluated by the root mean square error (RMSE) index. The experimental results show that our model has the better accurate and reasonable performance than the traditional imputation methods

关键词： Generative adversarial networks (GAN) spatio-temporal features processing traffic data imputation urban computing

来源：评论

学校读者我要写书评

暂无评论

Kalman Filter-Mutual Information Models Suggest Improved Training Protocols for Treating Cortical Blindness 59

Kalman Filter-Mutual Information Models Suggest Improved Tra...

引用

59th Annual Conference on Information sciences and Systems, CISS 2025

作者： Pack, Christopher C. Pack, Charles D. McGill University Montreal Neurological Institute Quebec Canada Monmouth University Computer Science/Software Eng. Red BankNJ United States

ISBN: (纸本)9798331513269

Vision training protocols, especially those for subjects with brain impairments, often employ a form of visual discrimination in which specific stimuli must be identified over the course of many days. Trainers can control exercise difficulty by modifying the stimuli in several ways, e.g., Gaussian smoothing, affine transformations, adding pixel noise or varying contrast and brightness. Typically, subjects improve performance in training, but such gains do not generalize afterwards. In a previous paper, we introduced a Kalman Filter model of vision training that, when coupled with a Shannon-type information-theoretic assessment technology, "mutual information", might be used to assess a protocol's potential for lasting acuity gains. Our empirical observation, based on about 300 training protocols with only one image (stimulus) pair, was that including correlated noise, as a training feature, seemed to improve the chances of a good generalization outcome. In this paper, using a sample of more than 3000 vision training protocols, with 5 image pairs of various complexities (entropies), 11 pixel noise levels, 8 pixel noise correlation types and 7 image distortion strategies, we extend and enhance the previous arguments. That is, we believe our models can be used to better understand which protocols may inhibit or enhance perceptual learning and, hence, show the way to improved brain training for individuals with cortical blindness. This would be a significant breakthrough since, currently, vision training, for this condition, is often ineffective. We also discuss how one might apply simple statistical techniques to interpolate, extrapolate or optimize protocol features. Future work, targeted for 2025, will focus on laboratory experiments that will allow us to explore and validate our ideas and observations. © 2025 IEEE.

关键词： Affine transforms

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：