检索结果-内蒙古大学图书馆

36th international conference on Algorithmic learning Theory, ALT 2025

作者： Balkanski, Eric Zhu, Cherlin Columbia University United States

An important challenge in robust machine learning is when training data is provided by strategic sources who may intentionally report erroneous data for their own benefit (Caro and Gallien, 2010;Caro et al., 2010). A line of work at the intersection of machine learning and mechanism design aims to deter strategic agents from reporting erroneous training data by designing learning algorithms that are strategyproof (Dekel et al., 2010;Perote and Perote-Pena, 2004;Chen et al., 2018;Cummings et al., 2015;Meir and Rosenschein, 2011;Meir et al., 2008, 2010, 2012;Hardt et al., 2016;Ghalme et al., 2021;Dong et al., 2018;Ahmadi et al., 2021). Strategyproofness is a strong and desirable property, but it comes at a cost in the approximation ratio of even simple risk minimization problems. A recent line of work on mechanism design with advice has shown that side information can be leveraged to overcome worst-case bounds in mechanism design. Strategyproof mechanisms that achieve an improved approximation ratio when the advice is accurate (consistency) and an acceptable approximation ratio when the advice is inaccurate (robustness) have been designed for problems such as strategic facility location (Agrawal et al., 2022;Xu and Lu, 2022;Balkanski et al., 2024a), auction design (Lu et al., 2024;Caragiannis and Kalantzis, 2024;Balkanski et al., 2024b), strategic scheduling (Balkanski et al., 2023), strategic assignment (Colini-Baldeschi et al., 2024), and metric distortion (Berger et al., 2024). In this paper, we study strategyproof learning in two settings: regression and classification;we provide the first non-trivial consistency-robustness tradeoffs for both. In strategic learning, each agent i ∈ {1, ..., n} reports a set Si = {(xi,j, yi,j)}j of labeled data points to the learner. The points xi,j are public information, but the labels yi,j are private information that agent i can potentially misreport. The goal of the mechanism is to learn a function f from a function class F tha

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Concentration prediction of coarse particle two-phase flow in vertical pipe transportation based on machine learning ensemble and numerical simulation methods

引用

POWDER TECHNOLOGY 2025年 458卷

作者： Wan, Chuyi Xiao, Shengpeng Zhou, Dai Zhu, Hongbo Bao, Yan Zhang, Mengmeng Niu, Zirong Han, Zhaolong Shanghai Jiao Tong Univ Sch Ocean & Civil Engn Minhang Campus Shanghai 200240 Peoples R China Minist Educ Key Lab Hydrodynam Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ State Key Lab Ocean Engn Shanghai 200240 Peoples R China Dalian Univ Technol State Key Lab Coastal & Offshore Engn Dalian 116000 Peoples R China Rocket Force Univ Engn Xian 710025 Peoples R China

Timely and long-term predictions of particle concentration changes in pipes are of great significance. The particle concentration outlier data across the pipe cross-section is obtained through numerical simulation and smooth interpolation method. The data is then decomposed using the Ensemble Empirical Mode Decomposition (EEMD), Variational Mode Decomposition (VMD), and VMD-EEMD methods. Finally, the Long Short-Term Memory (LSTM) model is employed to perform multi-step predictions on both the undifferentiated data and the data decomposed by the various models, with a step size of one second. The EEMD-LSTM, LSTM, and VMD-EEMD-LSTM models exhibited the highest accuracy in predicting the first, second, and third steps, respectively. The mean absolute errors (MAE) are 0.00014, 0.00040, and 0.00081;the root mean square errors (RMSE) are 0.00018, 0.00049, and 0.00103;the mean absolute percentage errors (MAPE) are 0.10, 0.27, and 0.56;and the coefficients of determination (R2) are 0.9978, 0.9844, and 0.9316. When selecting this outcome as the ensemble model, it demonstrates greater accuracy compared to other predictive models. This model is applicable to other cross-sections and exhibits values comparable to the MAE, RMSE, MAPE, and R2 of the original cross-section. Additional research is conducted on the formation mechanism of plug flow under various operating conditions using numerical simulations. The lubricating effect of small particles, along with the significant momentum exchange between numerous particles and fluids, can effectively mitigate the plug effect of the pipe flow.

关键词： machine learning Ensemble model Numerical simulation Solid-liquid two-phase flow Pipe transportation

来源：评论

学校读者我要写书评

暂无评论

Proceedings - 2022 7th international conference on data Science and machine learning Applications, CDMA 2022

Proceedings - 2022 7th International Conference on Data Scie...

引用

7th international conference on data Science and machine learning Applications, CDMA 2022

ISBN: (纸本)9781665410144

The proceedings contain 39 papers. The topics discussed include: evaluation of machine learning to early detection of highly cited papers;towards using deep reinforcement learning for better COVID-19 vaccine distribution strategies;an investigation of forecasting Tadawul all share index (TASI) using machine learning;intelligent deep detection method for malicious tampering of cancer imagery;an empirical analysis of health-related campaigns on twitter Arabic hashtags;the accuracy performance of semantic segmentation network with different backbones;a comprehensive evaluation of statistical, machine learning and deep learning models for time series prediction;depression detection in Arabic using speech language recognition;a deep learning framework for temperature forecasting;improving relevance in a recommendation system to suggest charities without explicit user profiles using dual-autoencoders;and the impact of feature selection on different machine learning models for breast cancer classification.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Identification of Disc Cutter Wear via Operation Parameters Combined With Vibration data: A Case Study

引用

international JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS 2025年第1期49卷 256-279页

作者： Wang, Yan-Ning Chen, Han Min, Xin-Hao Zhao, Lin-Shuang Shantou Univ Dept Civil Engn & Smart Cities Shantou Guangdong Peoples R China Shantou Univ Coll Engn Shantou Guangdong Peoples R China

This paper proposed an approach to estimate disc cutter wear utilizing a combination of multiple operational parameters and vibration data collected during shield tunneling operations. The incorporation of vibration signals, notably those originating from acceleration sensors mounted on the back plate of the soil chamber, has markedly enhanced the accuracy of the model. Time-frequency domain features were extracted through analysis methods such as Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT), and Continuous Wavelet Transform (CWT). A predictive model utilizing vibration and shield operation parameters was developed using the XGBoost algorithm, and a deep GoogLeNet Convolutional Neural Network (CNN) was trained on time-frequency graphs from the CWT. In addition, this study also investigated the impact of signal duration on wavelet image information and model accuracy. In the Huang-Shang Intercity Railway Project, the approach effectively assessed disc cutter wear during tunneling operations and dynamically optimized the operational parameters of the shield tunnel machine through predictive analysis.

关键词： cutter life cutter wear deep learning prediction model vibration signal

来源：评论

学校读者我要写书评

暂无评论

Sailing Through data: Automated data Annotation Systems for machine learning Applications in Maritime Navigation 24

Sailing Through Data: Automated Data Annotation Systems for ...

引用

14th international conference on the Internet of Things, IoT 2024

作者： Vibild, Patrick Dominique Hernandez, Jon Kim, Daeyoung School of Computing KAIST Daejeon Korea Republic of

ISBN: (纸本)9798400712852

The maritime industry plays a crucial role in global trade and transport, necessitating effective collision avoidance, navigation security, and reduction in human errors and costs. This paper presents an automated data annotation pipeline for detection and distance estimation using monocular camera systems in maritime applications. Our system integrates AIS data with ship detections from a pre-trained deep learning model, creating comprehensive RGB/IR data with annotations of ship positions, distances, speeds, categories, and sizes. We use sensor data from our developed embedded solution, AISail, and its post-processing techniques to generate this data. We evaluate our system’s performance against manually annotated images, demonstrating its potential to enhance maritime navigation and safety by accurately estimating ship positions and speeds. © 2024 Copyright held by the owner/author(s).

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Refining Multiple Instance learning with Attention Regularization for Whole Slide Image Classification 25th

Refining Multiple Instance Learning with Attention Regulariz...

引用

25th international conference on Intelligent data engineering and Automated learning

作者： Carretero, Ilan Meseguer, Pablo del Amor, Rock Naranjo, Valery Univ Politecn Valencia Inst Univ Invest Tecnol Centrada Ser Humano HUMAN Valencia 46022 Spain Valencian Grad Sch & Res Network Artificial Intel Valencia 46022 Spain

ISBN: (纸本)9783031777301;9783031777318

Histopathological analysis of biopsy sections is crucial for the detection of cancer and the distinction between different tumor subtypes. To this end, pathologists identify certain key regions of the biopsy from which a diagnosis is derived. The classification of whole slide images (WSIs) can be addressed as a multiple instance learning (MIL) problem where only slide-level labels are available. In order to model the relevance scores of the different WSI patches, attention mechanisms are implemented within the MIL framework. However, excessive flexibility in the attention mechanisms for computing attention scores to patches may lead to nearly uniform attention distributions, potentially deteriorating the model's performance. In this paper, we introduce an unsupervised auxiliary loss function to recalibrate the attention mechanism enhancing emphasis on crucial patches and downscaling the influence of less relevant ones. The proposed MIL framework has been evaluated on invasive breast carcinoma (TCGA-BRCA) and renal cell carcinoma (TCGA-RCC) subtyping. The results obtained show that attention regularization not only improves the predictive capacity of the model but also significantly increases its interpretability by identifying regions and patterns of high diagnostic value.

关键词： Hispathology imaging Multiple Instance learning Attention mechanisms Attention regularization Explainable Deep learning

来源：评论

学校读者我要写书评

暂无评论

Selecting the most important variables for crime prediction based on machine learning approaches

Selecting the most important variables for crime prediction ...

引用

2024 international conference on Mechatronics and Intelligent Control, ICMIC 2024

作者： Zhang, Xu Zhu, Weili Liu, Yang Guangdong Police College 500 Binjiang East Road Guangdong Guangzhou510230 China

ISBN: (数字)9781510686847

ISBN: (纸本)9781510686830

Crime prediction has been challenging, partly due to the difficulty in selecting predictive variables. Previous studies theoretically and empirically confirmed the relationship between many variables and crime. This study identified a total of 19 theory-driven variables and applied machine learning tree models for crime prediction. To refine the variable pool, we extracted the subsets of the variables from the variable pool by using feature selection processes, which was based on Gain and Permutated Variable Importance Measure (PVIM) methods. We finally found historical cases, percent nonlocal population, and ambient population(25-44) variables ranked high. Thus, our results call more attentions on these variables for crime prediction. © 2025 SPIE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of machine learning Models for Plant Disease Identification 19th

A Comparative Study of Machine Learning Models for Plant D...

引用

19th international conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2024

作者： Mara, Măcelaru Pop, Petrică Barata, José Department of Mathematics and Computer Science North University Center at Baia Mare Technical University of Cluj-Napoca Cluj-Napoca Romania Electrical and Computer Engineering Department School of Sciences and Technology NOVA University of Lisbon Lisbon Portugal

ISBN: (纸本)9783031750090

Early identification of plant diseases plays an important role in reducing the financial losses for farmers and enhancing the productivity and quality of crops. The present work proposes five machine learning (ML) algorithms: K-Nearest Neighbors, Random Forest, Iterative Dichotomizer 3, Adaptive Boosting and Logistic Regression for plant disease identification. In order to test the proposed ML algorithms, we used a large data set containing 74798 images mainly from PlantVillage dataset. The achieved experimental results show that the proposed ML techniques can be used to reliably identify various plant diseases. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Privacy Assessment Methodology for machine learning Models and data Sources 19th

Privacy Assessment Methodology for Machine Learning Models ...

引用

19th international conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2024

作者： Erdei, Rudolf Pasca, Emil Delinschi, Daniela Avram, Anca Chereja, Ionela Matei, Oliviu Technical University of Cluj Napoca North University Centre of Baia Mare Baia Mare Romania

ISBN: (纸本)9783031750090

machine learning and data Mining imply notable privacy vulnerabilities since they have the potential to expose confidential details about people or collectives that have contributed to the data. This paper proposes a Generalised Privacy Assessment Methodology that can be universally applied to any machine learning model and its corresponding data Source. The suggested technique facilitates the quantification of the Vulnerability Score for any given data source or model. We demonstrate the practicality and efficacy of the concept by implementing it in several scenarios, ranging from a small-scale IoT project to a hypothetical extensive database, like those used by governmental entities. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Overseas Communication Path for Chinese Corporate Image Based on Multimodal Discourse Analysis with machine learning Algorithm 24

Design and Implementation of Overseas Communication Path for...

引用

3rd international conference on Artificial Intelligence and Intelligent Information Processing, AIIIP 2024

作者： Lin, Jiling Lei, Fei Liu, Weiliang School of Foreign languages Guangdong University of Science & Technology Guangdong Dongguan China School of Finance and Economics Guangdong University of Science & Technology Guangdong Dongguan China Huizhou Modern Vocational Technology School Guangdong Huizhou China

ISBN: (纸本)9798400707308

In an increasingly globalized economy, the communication of corporate image plays a pivotal role in shaping perceptions and fostering relationships with international stakeholders. This study explores the design and implementation of effective overseas communication strategies for Chinese corporations, leveraging multimodal discourse analysis and machine learning algorithms. By synthesizing quantitative and qualitative data, we analyze various communication modalities—including visual, textual, and auditory elements—utilized in corporate branding across different cultural contexts. Employing machine learning techniques, we identify patterns and insights from vast datasets, enabling a nuanced understanding of how multimodal elements influence audience reception and engagement. Our findings reveal key factors that contribute to successful corporate image dissemination and highlight the importance of cultural sensitivity in communication strategies. The research provides a framework for Chinese corporations to enhance their global presence by integrating data-driven insights into their communication practices, ultimately fostering a more favorable corporate image and stronger international relationships. This study not only contributes to the field of corporate communication but also offers practical implications for practitioners seeking to navigate the complexities of multicultural environments in an era of rapid technological advancement. © 2024 Copyright held by the owner/author(s).

关键词： machine learning Algorithms

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：