检索结果-内蒙古大学图书馆

IEEE Conference on Decision and Control

作者： Bora Yongacoglu Gürdal Arslan Serdar Yüksel Department of Electrical and Computer Engineering University of Toronto Department of Electrical Engineering University of Hawaii at Manoa Department of Mathematics and Statistics Queen's University

Non-stationarity is a fundamental challenge in multi-agent reinforcement learning (MARL), where agents update their behaviour as they learn. In multi-agent settings, individual agents may have an incomplete view of the actions of others, which can complicate the learning process. Many theoretical advances in MARL avoid the challenge of non-stationarity by coordinating the policy updates of agents in various ways, including synchronizing times at which agents are allowed to revise their policies. In this paper, we study an asynchronous variant of the decentralized Q-learning algorithm, a recent MARL algorithm for stochastic games. We provide sufficient conditions under which the asynchronous algorithm drives play to equilibrium with high probability. In this generalization, players need not agree on the schedule of policy update times, and may change their policies at their own separately selected times. This work extends the applicability of the decentralized Q-learning algorithm to settings in which parameters are selected in an independent manner, and tames non-stationarity without imposing the coordination assumptions of prior work.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Output Feedback Reinforcement Learning for Temperature Control in a Fused Deposition Modelling Additive Manufacturing System 9

Output Feedback Reinforcement Learning for Temperature Contr...

引用

9th International Conference on Control, Decision and Information Technologies, CoDIT 2023

作者： Zavrakli, Eleni Parnell, Andrew Dey, Subhrakanti Maynooth University Co. Department of Mathematics and Statistics Kildare Ireland I-Form Advanced Manufacturing Research Centre Ireland Hamilton Institute Maynooth University Co. Kildare Ireland Uppsala University Division of Signals and Systems Department of Electrical Engineering Sweden

ISBN: (纸本)9798350311402

The development of effective closed-loop control algorithms is one of the main challenges in Additive Man-ufacturing (AM). Many parameters of AM processes need continuous monitoring and regulation, with temperature being one of the most important. We investigate the design of an output-feedback controller of the temperature process within the extruder of a Fused Deposition Modelling (FDM) AM system. Based on a state space approach, and using input-output measurements, we first design a model-based linear quadratic tracking controller, followed by an equivalent model-free, data-driven version. We demonstrate these approaches using a simulator of the temperature evolution in the extruder of the AM system, based on a model validated and identified in recent literature. Our findings show that a comparable performance to the model-based case is possible using only measured data, generated through probing control explorations during the simulations. © 2023 IEEE.

关键词： Additives

来源：评论

学校读者我要写书评

暂无评论

Engaging engineering mathematics students online: Tutoring using MS Teams

Engaging Engineering Mathematics students online: Tutoring u...

引用

2022 IEEE IFEES World engineering Education Forum - Global engineering Deans Council, WEEF-GEDC 2022

作者： Padayachee, Pragashni Campbell, Anita Fredericks, Ebrahim Academic Support Program in Engineering University of Cape Town Cape Town South Africa Centre for Research in Engineering Education University of Cape Town Cape Town South Africa Mathematics University of Cape Town Department of Mathematics and Applied Cape Town South Africa

ISBN: (纸本)9781665475280

Uncommon in the South African higher education landscape, online learning came to the fore during the global pandemic. We present an account of the use of Microsoft Teams for hybrid mathematics tutorials in a one-semester Vector Calculus course at a South African university in 2022. Interviews with the lecturer, analysed through a Community of Inquiry lens, showed the lecturer's perspective of the design and experience for tutors and students. Our aim is to improve the design of future hybrid tutorials to ensure engineering students' capacity in mathematics is well developed, their communication skills are improved, and that they experience working in a team. Future research will evaluate students' and tutors' accounts of their experiences. Our findings raise awareness of the possibilities and potential difficulties when using Microsoft Teams for communication, teaching and learning mathematics. © 2022 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Data-based nonlinear and stochastic dynamics

引用

The European Physical Journal Special Topics 2025年 1-5页

作者： Xu, Yong Kurths, Jürgen Li, Yongge Lenci, Stefano School of Mathematics and Statistics Northwestern Polytechnical University Xi’an China Department of Physics Humboldt University Berlin Berlin Germany Department of Civil and Building Engineering and Architecture Polytechnic University of Marche Ancona Italy

This special issue compiles 20 contributions, covering a wide range of latest achievements on dynamical modeling, data-driven algorithms, response predictions, multiple practical applications, and inverse problems. Data science plays a crucial role, helping us constructing more accurate dynamical models that capture and reflect the true dynamical changes of a system. At the same time, data science is also a powerful tool for deriving exact solutions of a system. By integrating it with deep learning algorithms, we are able to effectively predict the system responses and successfully apply this in several applications, such as airfoil flutter, arm musculoskeletal system, financial market, and epidemiology. In addition, inverse problems also occupy a pivotal position. Faced with the existing rich data, how to use data to identify the parameters of the model is still a challenging topic worthy of our continued attention and in-depth exploration.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Explainable AI: Leveraging SHAP for Transparent Decision-Making in Machine Learning

Enhancing Explainable AI: Leveraging SHAP for Transparent De...

引用

International Conference on Informatics and Computing (ICIC)

作者： Marsani Asfi Budi Warsito Adi Wibowo Doctoral Program Of Information System School of Postgraduate Studies Diponegoro University Semarang Indonesia Department Of Statistics Faculty of Science and Mathematics Diponegoro University Semarang Indonesia Departement Of Informatics Faculty Of Science and Mathematics Diponegoro University Semarang Indonesia

ISBN: (数字)9798331517601

ISBN: (纸本)9798331517618

In recent years, the swift progress in Artificial Intelligence (AI) has resulted in to impressive performance across various domains, but the opacity of complex models, often referred to as “black-box” models, has raised concerns regarding trust and interpretability. This paper addresses these challenges through Explainable Artificial Intelligence (XAI), focusing on Shapley Additive Explanations (SHAP) to interpret a predictive model built on an online retail dataset. The research employs Gradient Boosting Machines (GBM) to predict customer repeat purchases and uses SHAP to provide clear insights into the model’s decision-making process. SHAP effectively identifies key features such as TotalPrice, UnitPrice, and Quantity as the most significant factors driving the model’s predictions. TotalPrice was found to be the most influential feature, demonstrating its strong association with repeat purchases. The analysis reveals that the total price of items purchased is the most significant predictor of repeat purchases, with a SHAP value of 0.45, indicating a strong positive correlation. Additionally, unit price and quantity also play important roles, contributing SHAP values of 0.32 and 0.20, respectively The model achieved an accuracy of 85%, with a precision of 0.78 and a recall of 0.80. These results indicate a robust performance in predicting repeat purchases. Furthermore, the research highlights that consumers from certain countries demonstrate distinct purchasing patterns, influencing overall sales performance. These insights empower retail managers to identify critical variables and adapt marketing strategies accordingly. The analysis not only enhances model transparency but also ensures that the predictions align with business expectations, allowing stakeholders to trust and act on the model’s insights.

关键词： Analytical models Adaptation models Ethics Explainable AI Decision making Focusing Predictive models Boosting Stakeholders Informatics

来源：评论

学校读者我要写书评

暂无评论

An investigation into water consumption by university students: actions, beliefs and perceptions

引用

Discover Water 2024年第1期4卷 1-15页

作者： Broering, Leonardo André Michel, Julia Henning, Elisa Kalbusch, Andreza Konrath, Andrea Cristina Mathematics Department Santa Catarina State University Joinville Brazil Civil Engineering Department Santa Catarina State University Joinville Brazil Computing and Statistics Department Federal University of Santa Catarina Florianópolis Brazil

Sustainable development requires water resources, and the supply of water is directly linked to economic growth, political and social issues. This article presents the findings of a research study conducted to identify actions, beliefs and perceptions regarding water consumption among students and the possible associations between actions related to water conservation and demographic, socioeconomic and cultural variables. In addition, we aimed at finding any changes in water use habits due to the COVID-19 pandemic. The study was carried out at a university in the city of Joinville (southern Brazil). Non-probabilistic quota sampling was implemented and a total of 246 responses were obtained. Exploratory data analysis was performed by means of descriptive and graphical measures. Classical statistical tests were applied to identify possible associations. The results provided information on the use of water for personal hygiene and cleaning, among others. Based on the understanding of the consumption habits, more targeted awareness-raising actions can be prioritized.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dielectric Properties of Ground Beef for Microwave Sensing and Imaging Phantoms: Measurement Considerations and Effect of Dehydration

Dielectric Properties of Ground Beef for Microwave Sensing a...

引用

URSI International Symposium on Electromagnetic Theory (EMTS)

作者： Brendon C. Besler Ryan Baker Hua Shen Elise Fear Department of Electrical and Software Engineering University of Calgary Calgary Canada Department of Mathematics and Statistics University of Calgary Calgary Canada

ISBN: (数字)9798350399288

ISBN: (纸本)9798350399295

Ground meat may be used to mimic the dielectric properties of human tissue when validating microwave sensors and algorithms. Accurate measurements of the dielectric properties of ground meat are important for this validation. Based on a pilot study, key considerations for taking reliable measurements of ground beef samples and the effect of hydration on ground beef properties are reported.

关键词： Microwave measurement Protocols Microwave sensors Microwave theory and techniques Dielectric measurement Dielectrics Sensors Reliability Probes Microwave imaging

来源：评论

学校读者我要写书评

暂无评论

Diagnostics for categorical response models based on quantile residuals and distance measures

arXiv

引用

arXiv 2023年

作者： Araripe, Patrícia Peres de Lara, Idemauro Antonio Rodrigues Palma, Gabriel Rodrigues Cahill, Niamh de Andrade Moral, Rafael "Luiz de Queiroz" College Postgraduate Program in Statistics and Agronomic Experimentation University of São Paulo Brazil Department of Mathematics and Statistics Maynooth University Ireland

Polytomous categorical data, in a nominal or ordinal scale, are frequent in many studies in different areas of knowledge. Depending on experimental design, these data can be obtained with an individual or grouped structure. In both structures, the multinomial distribution may be suitable to model the response variable and, in general, the generalized logit model is commonly used to relate the covariates’ potential effects on the response variable. After fitting a multi-categorical model, one of the challenges is the definition of an appropriate residual and choosing diagnostic techniques to assess goodness-of-fit, as well as validate inferences based on the model. Since the polytomous variable is multivariate, raw, Pearson, or deviance residuals are vectors and their asymptotic distribution is generally unknown, which leads to potential difficulties in graphical visualization and interpretation. Therefore, the definition of appropriate residuals, as well as the choice of the correct analysis in diagnostic tools is very important, especially for nominal categorical data, where a restriction of methods is observed. This paper proposes the use of randomized quantile residuals associated with individual and grouped nominal data, as well as Euclidean and Mahalanobis distance measures associated with grouped data only, as an alternative method to reduce the dimension of the residuals and to study outliers. To show the effectiveness of the proposed methods, we developed simulation studies with individual and grouped categorical data structures associated with generalized logit models. Parameter estimation was carried out by maximum likelihood and half-normal plots with simulation envelopes were used to assess model performance using residuals and distance metrics. These studies demonstrated a good performance of the quantile residuals and, also, the distance measurements allowed a better interpretation of the graphical techniques. We illustrate the proposed procedures with

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

New Solitary Wave and Computational Solitons for Kundu Eckhaus Equation

SSRN

引用

SSRN 2022年

作者： Jaradat, Mohammed M.M. Batool, Amna Raza, Nauman Butt, Asma Rashid Mathematics Program Department of Mathematics Statistics and Physics College of Arts and Sciences Qatar University Doha2713 Qatar Department of Mathematics University of the Punjab Quaid-e-Azam Campus Lahore Pakistan Department of Mathematics University of Engineering and Technology Lahore Pakistan

The goal of this research is to nd novel optical solutions to the Kundu-Eckhaus equation,which possess crucial role in the eld of nonlinear optics. Collective variable (CV) strategyis adopted to solve governing equation including Raman e ect and quintic nonlinearity. Theparameters employed in this approach are chirp, temporal position, phase, amplitude, frequencyand width, namely collective variables. The fourth order Runge-Kutta technique is a renownednumerical scheme that aims towards the solution of the resulting system of ordinary di erentialequations representing the variables involved in the pulse ansatz. This techniques presents theevolution of pulse parameters with regard to propagation variable. The graphical pro les atsuitable values of pulse parameters are also provided. The results developed in this article showthat CV approach can be applied to solve a variety of nonlinear problems in mathematicalsciences. © 2022, The Authors. All rights reserved.

关键词： Solitons

来源：评论

学校读者我要写书评

暂无评论

Leveraging LLMs for Multimodal Medical Time Series Analysis 9

Leveraging LLMs for Multimodal Medical Time Series Analysis

引用

9th Machine Learning for Healthcare Conference, MLHC 2024

作者： Chan, Nimeesha Parker, Felix Bennett, William Wu, Tianyi Jia, Mung Yao Fackler, James Ghobadi, Kimia Department of Civil and Systems Engineering Johns Hopkins University BaltimoreMD United States Department of Anesthesiology and Critical Care Medicine Johns Hopkins University BaltimoreMD United States Department of Applied Mathematics and Statistics Johns Hopkins University BaltimoreMD United States Department of Computer Science Johns Hopkins University BaltimoreMD United States

The complexity and heterogeneity of data in many real-world applications pose significant challenges for traditional machine learning and signal processing techniques. For instance, in medicine, effective analysis of diverse physiological signals is crucial for patient monitoring and clinical decision-making and yet highly challenging. We introduce MedTsLLM, a general multimodal large language model (LLM) framework that effectively integrates time series data and rich contextual information in the form of text to analyze physiological signals, performing three tasks with clinical relevance: semantic segmentation, boundary detection, and anomaly detection in time series. These critical tasks enable deeper analysis of physiological signals and can provide actionable insights for clinicians. We utilize a reprogramming layer to align embeddings of time series patches with a pretrained LLM’s embedding space and make effective use of raw time series, in conjunction with textual context. Given the multivariate nature of medical datasets, we develop methods to handle multiple covariates. We additionally tailor the text prompt to include patient-specific information. Our model outperforms state-of-the-art baselines, including deep learning models, other LLMs, and clinical methods across multiple medical domains, specifically electrocardiograms and respiratory waveforms. MedTsLLM presents a promising step towards harnessing the power of LLMs for medical time series analysis that can elevate data-driven tools for clinicians and improve patient outcomes. © 2024 Authors.

关键词： Semantics

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：