检索结果-内蒙古大学图书馆

41st International Conference on Machine Learning, ICML 2024

作者： Nguyen, Duc Zhang, Anderson Ye Department of Computer & Information Science University of Pennsylvania United States Department of Statistics & Data Science Wharton School University of Pennsylvania United States

The Partial Credit Model (PCM) of Andrich (1978) and Masters (1982) is a fundamental model within the psychometric literature with wide-ranging modern applications. It models the integer-valued response that a subject gives to an item where there is a natural notion of monotonic progress between consecutive response values, such as partial scores on a test and customer ratings of a product. In this paper, we introduce a novel, time-efficient and accurate statistical spectral algorithm for inference under the PCM model. We complement our algorithmic contribution with in-depth non-asymptotic statistical analysis, the first of its kind in the literature. We show that the spectral algorithm enjoys the optimal error guarantee under three different metrics, all under reasonable sampling assumptions. We leverage the efficiency of the spectral algorithm to propose a novel EM-based algorithm for learning mixtures of PCMs. We perform comprehensive experiments on synthetic and real-life datasets covering education testing, recommendation systems, and financial investment applications. We show that the proposed spectral algorithm is competitive with previously introduced algorithms in terms of accuracy while being orders of magnitude faster. Copyright 2024 by the author(s)

关键词：

来源：评论

学校读者我要写书评

暂无评论

LATTE: Layer Algorithm-aware Training Time Estimation for Heterogeneous Federated Learning 24

LATTE: Layer Algorithm-aware Training Time Estimation for He...

引用

30th International Conference on Mobile Computing and Networking, ACM MobiCom 2024

作者： Wang, Kun Zhou, Zimu Li, Zhenjiang Department of Computer Science City University of Hong Kong Hong Kong School of Data Science City University of Hong Kong Hong Kong

ISBN: (纸本)9798400704895

Accurate estimation of on-device model training time is increasingly required for emerging learning paradigms on mobile edge devices, such as heterogeneous federated learning (HFL). HFL usually customizes the model architecture according to the different capabilities of mobile edge devices to ensure efficient use of local data from all devices for training. However, due to oversimplification of latency modeling, existing methods rely on a single coefficient to represent computational heterogeneity, resulting in sub-optimal HFL efficiency. We find that existing methods ignore the important impact of runtime optimization of deep learning frameworks, which we call development-chain diversity. Specifically, layers of a model may have different algorithm implementations, and deep learning frameworks often have different strategies for selecting the algorithm they believe is the best based on a range of runtime factors, resulting in different training latencies and invalid predictions from existing methods. In this paper, in addition to considering this diversity to ensure synchronized completion time of model training, we also study how to select the best algorithm each time to reduce the latency of the per-round training, thereby further improving the overall efficiency of federated training. To this end, we propose LATTE, which consists of a novel selector that identifies the best algorithm at runtime based on relative runtime factors. By further integrating it into our training latency model, LATTE provides accurate training time estimation. We develop LATTE as middleware, compatible with different deep learning frameworks. Extensive results show significantly improved training convergence speed and model accuracy compared to state-of-the-art methods. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Transparency in Translation: A Deep Dive into Explainable AI Techniques for Bias Mitigation

Transparency in Translation: A Deep Dive into Explainable AI...

引用

2024 Asia Pacific Conference on Innovation in Technology, APCIT 2024

作者： Deokar, Ruchira Nanjundan, Preethi Mohanty, Sachi Nandan Christ University Department of Data Science Pune Lavasa India School of Computer Science & Engineering VIT -AP University Andhra Pradesh India

ISBN: (纸本)9798350361537

In an era dominated by artificial intelligence (AI), concerns about bias and discrimination loom large. The quest for fairness and equity in AI-driven decision-making has led to the exploration of Explainable AI (XAI) as a viable solution. This paper undertakes a thorough examination of the bias ingrained within AI systems and posits XAI as a potent antidote. Beginning with an exploration of the origins and aftermath of bias in AI, the analysis traverses the evolution of XAI techniques, including SHAP, LIME, and counterfactual explanations, clearly stating their advantages and drawbacks. With each XAI method thoroughly inspected, the study unravels their applicability across diverse AI models and domains. Furthermore, a compelling case study is presented, showcasing XAI's practical application in a language translation app, where it guarantees transparency and equity in the translation process. This tangible example serves as a testament to XAI's efficacy in mitigating bias within real-world applications. As the analysis concludes, it underscores the pivotal role XAI plays in fostering accountability and trustworthiness in AI systems. By shedding light on how XAI mitigates bias and offering concrete examples of its utility, the paper advocates for its widespread adoption as an imperative step towards the development of ethically robust AI systems. In a landscape filled with concerns about bias, XAI emerges as a beacon of hope, promising a future where AI decisions are transparent, fair, and equitable for all. © 2024 IEEE.

关键词： Sensitivity analysis

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Effectiveness of Deep Convolutional Neural Networks for Picture Quality Prediction 27

Evaluating the Effectiveness of Deep Convolutional Neural Ne...

引用

27th International Symposium on Wireless Personal Multimedia Communications, WPMC 2024

作者： Jhingran, Sushant Chaturvedi, Rekha Bansal, Nidhi Malik, Suraj Sharda University Department of Computer Science and Engineering Greater Noida India Manipal University Jaipur School of Information Security and Data Science Department of Data Science and Engineering Rajasthan Jaipur India Pune India IIMT University Department of Computer Science and Engineering Uttar Pradesh Meerut India

ISBN: (纸本)9798350392319

Since the advent of smartphones, capturing images has become deeply embedded in human behavior, evolving into a fundamental part of daily life. Research into human perception of image quality is crucial as people frequently acquire and interpret images. Automation in image processing systems is vital for quantifying image quality, addressing issues like blurriness, noise, and compression that can significantly degrade the visual experience. While image processing professionals can quickly identify such distortions, integrating human visual perception into automated system design remains challenging. Consequently, machine-based evaluation of image quality is a critical research area. Image quality assessment algorithms aim to approximate human visual judgments. Although deep learning has shown remarkable success in various real-world tasks, such as object detection and image classification, its application in image quality assessment is still evolving. The primary challenge is assessing image quality without explicitly considering human perception, making it a dynamic area of ongoing research. A significant hurdle is the lack of images with quality scores, particularly differential average feedback scores provided by image quality experts on a 1-100 scale. This limitation hinders the development of robust models for image quality assessment, highlighting the need for innovative solutions in this evolving research domain. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

ScanTrimmer: an Online Dynamic Objects Removal Framework in Laser Scan for Robust Localization

ScanTrimmer: an Online Dynamic Objects Removal Framework in ...

引用

2023 IEEE International Conference on Robotics and Biomimetics, ROBIO 2023

作者： Wang, Jing Duan, Yifan Ji, Jianmin University of Science and Technology of China School of Data Science Hefei230026 China University of Science and Technology of China School of Computer Science and Technology Hefei230026 China

ISBN: (纸本)9798350325706

The Simultaneous Localization and Mapping (SLAM) technique is often employed in robotic localization tasks. For lidar-based SLAM, point cloud registration (PCR) is one of the crucial factors for overall localization precision. However, the quality of point-to-point correspondence between two distinct point clouds obtained by PCR is environment dependent. Particularly in dynamic environments, the movement of objects can disrupt this correspondence, resulting in a degradation of localization accuracy. Therefore, previous SLAM methods often fail in environments with ample dynamic objects. Intuitively, we can remove dynamic points from the input laser scan to degenerate the influence of environment dynamicity. However, identifying dynamic objects precisely from point clouds which is challenging for the SLAM running in robotic systems with high real-time requirements. This paper proposes a new online lidar-based SLAM framework, i.e., ScanTrimmer, which significantly enhances the localization accuracy of the SLAM systems by detecting and deleting potential dynamic points through space-based comparison in point cloud distribution between scans taken at a specific interval. The experiments demonstrate that ScanTrimmer can improve the localization accuracy by 44.69% on the simulated dataset and 25.99% on the real-world dataset compared with the previous method. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Group Buying Recommendation Model Based on Multi-task Learning 39

Group Buying Recommendation Model Based on Multi-task Learni...

引用

39th IEEE International Conference on data Engineering, ICDE 2023

作者： Zhai, Shuoyao Liu, Baichuan Yang, Deqing Xiao, Yanghua Fudan University School of Data Science Shanghai200433 China Fudan University School of Computer Science Shanghai200433 China Shanghai Key Laboratory of Data Science Shanghai200433 China

ISBN: (纸本)9798350322279

In recent years, group buying has become one popular kind of online shopping activities, thanks to its larger sales and lower unit price. Unfortunately, seldom research focuses on the recommendations specifically for group buying by now. Although some recommendation models have been proposed for group recommendation, they can not be directly used to achieve the real-world group buying recommendation, due to the essential difference between group recommendation and group buying recommendation. In this paper, we first formalize the task of group buying recommendation into two sub-tasks. Then, based on our insights into the correlations and interactions between the two sub-tasks, we propose a novel recommendation model for group buying, namely MGBR, which is built mainly with a multi-task learning module. To improve recommendation performance further, we devise some collaborative expert networks and adjusted gates in the multi-task learning module, to promote the information interaction between the two sub-tasks. Furthermore, we propose two auxiliary losses corresponding to the two sub-tasks, to refine the representation learning in our model. Our extensive experiments not only demonstrate that the augmented representations learned in our model result in better performance than previous recommendation models, but also justify the impacts of the specially designed components in our model. To reproduce our model's recommendation results conveniently, we have provided our model's source code and dataset on https://***/DeqingYang/MGBR. © 2023 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Iterative Android automated testing

引用

Frontiers of computer science 2023年第5期17卷 15-26页

作者： Yi ZHONG Mengyu SHI Youran XU Chunrong FANG Zhenyu CHEN State Key Laboratory for Novel Software Technology Nanjing UniversityNanjing 210093China School of Big Data and Computer Science Chongqing College of Mobile CommunicationChongqing 400065China

With the benefits of reducing time and workforce,automated testing has been widely used for the quality assurance of mobile applications(APPs).Compared with automated testing,manual testing can achieve higher coverage in complex interactive *** the effectiveness of manual testing is highly dependent on the user operation process(UOP)of experienced *** on the UOP,we propose an iterative Android automated testing(IAAT)method that automatically records,extracts,and integrates UOPs to guide the test logic of the tool across the complex Activity *** feedback test results can train the UOPs to achieve higher coverage in each *** extracted 50 UOPs and conducted experiments on 10 popular mobile APPs to demonstrate IAAT’s effectiveness compared with Monkey and the initial automated *** experimental results show a noticeable improvement in the IAAT compared with the test logic without human *** the 60 minutes test time,the average code coverage is improved by 13.98%to 37.83%,higher than the 27.48%of Monkey under the same conditions.

关键词： quality assurance automated testing UOP test coverage

来源：评论

学校读者我要写书评

暂无评论

A Background Knowledge Revising and Incorporating Dialogue Model

A Background Knowledge Revising and Incorporating Dialogue M...

引用

作者： Zhao, Xinyan Feng, Xiao Chen, Huanhuan University of Science and Technology of China School of Data Science Hefei230027 China University of Science and Technology of China School of Computer Science and Technology Hefei230027 China

Currently, dialogue systems have attracted increasing research interest. In particular, background knowledge is incorporated to improve the performance of dialogue systems. Existing dialogue systems mostly assume that the background knowledge is correct and comprehensive. However, low-quality background knowledge is common in real-world applications. On the other hand, dialogue datasets with manual labeled background knowledge are often insufficient. To tackle these challenges, this article presents an algorithm to revise low-quality background knowledge, named background knowledge revising transformer (BKR-Transformer). By innovatively formulating the knowledge revising task as a sequence-to-sequence (Seq2Seq) problem, BKR-Transformer generates the revised background knowledge based on the original background knowledge and dialogue history. More importantly, to alleviate the effect of insufficient training data, BKR-Transformer introduces the ideas of parameter sharing and tensor decomposition, which could significantly reduce the number of model parameters. Furthermore, this work presents a background knowledge revising and incorporating dialogue model that combines the background knowledge revision with response selection in a unified model. Empirical analyses on real-world applications demonstrate that the proposed background knowledge revising and incorporating dialogue system (BKRI) could revise most low-quality background knowledge and substantially outperforms previous dialogue models. © 2012 IEEE.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Teaching Evaluation Index of College Students Based on Random Forest 3

Teaching Evaluation Index of College Students Based on Rando...

引用

3rd International Conference on Educational Technology, ICET 2023

作者： Jiang, Mingchi Huang, Xiao Liu, Dandan Hu, Shengbo School of Big Data and Computer Science Guizhou Normal University Guiyang China

ISBN: (纸本)9798350302332

In order to solve the problem of unscientific setting of evaluation indicators due to the non-standardized evaluation scale of college students, we propose to use random forest algorithm to model and analyze the evaluation indicators of students. Taking the course "Programming Fundamentals"as an example, the teaching evaluation questionnaire is designed, and the random forest algorithm is used to model, compare the measured value and predicted value of the comprehensive scores, and obtain the importance ranking of the teaching evaluation indicators. The experimental results show that teaching effect, Language expression and other teaching evaluation indicators occupy an important position in the scale of university students' evaluation, and have important reference value for the design of the scale. © 2023 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Optimizing Coverage in Wireless Sensor Networks Using the Tumbleweed Algorithm

Journal of Network Intelligence

引用

Journal of Network Intelligence 2025年第1期10卷 250-261页

作者： Zhang, Mengfei Chu, Shu-Chuan Liu, Nengxian Huang, Kuan-Chun Pan, Jeng-Shyang College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China College of Computer and Data Science Fuzhou University Fuzhou350108 China Department of Industrial and Systems Engineering Chung Yuan Christian University Taoyuan320314 Taiwan School of Artificial Intelligence Nanjing University of Information Science and Technology Nanjing210044 China Department of Information Management Chaoyang University of Technology Taichung41349 Taiwan

Wireless sensor networks (WSNs) have found extensive applications across various fields, significantly enhancing the convenience in our daily lives. Hence, an in-creasing number of researchers are directing their attention toward the performance of sensor nodes, with a particular emphasis on addressing the coverage challenges in WSNs. Node coverage plays a crucial role in determining the detection performance of WSNs, making it imperative to attain maximum coverage with a given number of sensor nodes. This paper applies the Tumbleweed Algorithm (TA) to the coverage problem in WSNs and validates its effectiveness through simulation experiments. The results demonstrate its unique advantages in global search and avoiding local optima, providing a new solution for the WSN coverage problem and offering valuable references for future research and practical applications. The TA algorithm simulates the lifecycle of tumbleweed from seedling to mature plant and includes the concept of growth cycles. This paper compares the TA algorithm with PSO, GWO, DE, BOA, SCA, and AOA, setting different numbers of wireless sensor nodes and varying sensing radii for the experiments. The experimental results show that the TA algorithm significantly improves coverage rates under various configurations and consistently outperforms other commonly used algorithms. © 2025, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Sensor nodes

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：