检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Sikha Pentyala Mayana Pereira Martine De Cock School of Engineering and Technology University of Washington Tacoma Department of Electrical Engineering Universidade de Brasilia Brazil School of Engineering and Technology University of Washington Tacoma and Department of Applied Mathematics Computer Science and Statistics Ghent University Belgium

Data is the lifeblood of the modern world, forming a fundamental part of AI, decision-making, and research advances. With increase in interest in data, governments have taken important steps towards a regulated data world, drastically impacting data sharing and data usability and resulting in massive amounts of data confined within the walls of organizations. While synthetic data generation (SDG) is an appealing solution to break down these walls and enable data sharing, the main drawback of existing solutions is the assumption of a trusted aggregator for generative model training. Given that many data holders may not want to, or be legally allowed to, entrust a central entity with their raw data, we propose a framework for collaborative and private generation of synthetic tabular data from distributed data holders. Our solution is general, applicable to any marginal-based SDG, and provides input privacy by replacing the trusted aggregator with secure multiparty computation (MPC) protocols and output privacy via differential privacy (DP). We demonstrate the applicability and scalability of our approach for the state-of-the-art select-measure-generate SDG algorithms MWEM+PGM and AIM.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Long-Tail Theory Under Gaussian Mixtures 26

Long-Tail Theory Under Gaussian Mixtures

引用

26th European Conference on Artificial Intelligence, ECAI 2023

作者： Bolatov, Arman Tezekbayev, Maxat Melnykov, Igor Pak, Artur Nikoulina, Vassilina Assylbekov, Zhenisbek Department of Computer Science School of Engineering and Digital Sciences Nazarbayev University Kabanbay Batyr 53 Astana010000 Kazakhstan Department of Mathematics School of Sciences and Humanities Nazarbayev University Kabanbay Batyr 53 Astana010000 Kazakhstan Department of Mathematics and Statistics University of Minnesota Duluth DuluthMN United States Naver Labs Europe 6-8 chemin de Maupertuis Meylan38240 France Department of Mathematical Sciences Purdue University Fort Wayne Fort WayneIN United States

ISBN: (纸本)9781643684369

We suggest a simple Gaussian mixture model for data generation that complies with Feldman's long tail theory (2020). We demonstrate that a linear classifier cannot decrease the generalization error below a certain level in the proposed model, whereas a nonlinear classifier with a memorization capacity can. This confirms that for long-tailed distributions, rare training examples must be considered for optimal generalization to new data. Finally, we show that the performance gap between linear and nonlinear models can be lessened as the tail becomes shorter in the subpopulation frequency distribution, as confirmed by experiments on synthetic and real data. © 2023 The Authors.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

UNSYNCHRONIZED DECENTRALIZED Q-LEARNING: TWO TIMESCALE ANALYSIS BY PERSISTENCE

arXiv

引用

arXiv 2023年

作者： Yongacoglu, Bora Arslan, Gürdal Yüksel, Serdar Department of Electrical and Computer Engineering University of Toronto Canada Department of Electrical Engineering University of Hawaii Manoa United States Department of Mathematics and Statistics Queen’s University Canada

Non-stationarity is a fundamental challenge in multi-agent reinforcement learning (MARL), where agents update their behaviour as they learn. 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. Synchronization enables analysis of many MARL algorithms via multi-timescale methods, but such synchronization is infeasible in many decentralized applications. In this paper, we study an unsynchronized variant of the decentralized Q-learning algorithm, a recent MARL algorithm for stochastic games. We provide sufficient conditions under which the unsynchronized algorithm drives play to equilibrium with high probability. Our solution utilizes constant learning rates in the Q-factor update, which we show to be critical for relaxing the synchronization assumptions of earlier work. Our analysis also applies to unsynchronized generalizations of a number of other algorithms from the regret testing tradition, whose performance is analyzed by multi-timescale methods that study Markov chains obtained via policy update dynamics. This work extends the applicability of the decentralized Q-learning algorithm and its relatives to settings in which parameters are selected in an independent manner, and tames non-stationarity without imposing the coordination assumptions of prior work. © 2023, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Stochastic Approximate Expectation-Maximization for Structure Determination Directly from Cryo-EM Micrographs

arXiv

引用

arXiv 2023年

作者： Kreymer, Shay Singer, Amit Bendory, Tamir The School of Electrical Engineering Tel Aviv University Tel Aviv Israel The Department of Mathematics and Program in Applied and Computational Mathematics Princeton University PrincetonNJ08544 United States

A single-particle cryo-electron microscopy (cryoEM) measurement, called a micrograph, consists of multiple two-dimensional tomographic projections of a three-dimensional molecular structure at unknown locations, taken under unknown viewing directions. All existing cryo-EM algorithmic pipelines first locate and extract the projection images, and then reconstruct the structure from the extracted images. However, if the molecular structure is small, the signal-to-noise ratio (SNR) of the data is very low, and thus accurate detection of projection images within the micrograph is challenging. Consequently, all standard techniques fail in low-SNR regimes. To recover molecular structures from measurements of low SNR, and in particular small molecular structures, we devise a stochastic approximate expectation-maximization algorithm to estimate the three-dimensional structure directly from the micrograph, bypassing locating the projection images. We corroborate our computational scheme with numerical experiments, and present successful structure recoveries from simulated noisy measurements. © 2023, CC BY.

关键词： Molecular structure

来源：评论

学校读者我要写书评

暂无评论

Comparative Study of Classical Machine Learning Models in Credit Scoring 6

Comparative Study of Classical Machine Learning Models in Cr...

引用

6th International Conference on Artificial Intelligence and Advanced Manufacturing, AIAM 2024

作者： Meng, Xiaoqi Jia, Lihan Chen, Shuo Zhou, Yu Liu, Chang School of Mathematics and Statistics Hainan Normal University Haikou571158 China McGill University MontrealH3A 0G4 Canada Miami University Oxford45056 United States College of Information Engineering Shanghai Maritime University Shanghai200135 China Department of Electrical and Computer Engineering University of Illinois at Urbana Champaign IL61801 United States

ISBN: (纸本)9781837242672

Credit scoring is a crucial aspect of financial risk management, enabling lenders to evaluate the creditworthiness of individuals and entities. This paper presents a comprehensive comparison of various classical machine learning models to determine the most effective one for predicting credit scores. The evaluated models include logistic regression, decision trees, random forests, support vector machines, gradient boosting, adaptive boosting (AdaBoost), k-nearest neighbors (KNN), Naive Bayes, eXtreme Gradient Boosting (XGBoost), Light Gradient-Boosting Machine (LightGBM), multi-layer perceptron (MLP) classifiers, deep neural networks (DNN), and convolutional neural networks (CNN). Each model underwent 10-fold cross-validation and was assessed based on metrics such as accuracy, precision, recall, F1 score, and receiver operating characteristic curve (ROC curve). Results indicate that XGBoost and LightGBM outperform other models across multiple metrics, demonstrating their suitability for credit score prediction. The paper concludes with recommendations for selecting models tailored to specific business requirements and data characteristics, highlighting the significant potential of advanced machine learning techniques in enhancing credit scoring accuracy and reliability. © The Institution of engineering & Technology 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Wave Propagation Model in a Human Long Poroelastic Bone under Effect of Magnetic Field and Rotation

引用

Computers, Materials & Continua 2021年第8期68卷 1485-1504页

作者： A.M.Abd-Alla Hanaa Abu-Zinadah S.M.Abo-Dahab J.Bouslimi M.Omri Department of Mathematics Sohag UniversitySohagEgypt Department of Statistics University of JeddahCollege of ScienceJeddahSaudi Arabia Department of Mathematics South Valley UniversityQena83523Egypt Department of Engineering Physics and Instrumentation National Institute of Applied Sciences and TechnologyCarthage UniversityTunisia Department of Physics Taif UniversityTaifSaudi Arabia Deanship of Scientific Research King Abdulaziz UniversityJeddahSaudi Arabia

This article is aimed at describing the way rotation and magnetic field affect the propagation of waves in an infinite poroelastic cylindrical *** offers a solution with an exact closed *** authors got and examined numerically the general frequency equation for poroelastic ***,they calculated the frequencies of poroelastic bone for different values of the magnetic field and *** the results of previous studies,the authors noticed little frequency dispersion in the wet *** proposed model will be applicable to wide-range parametric projects of bone mechanical *** the vibration of surface waves in rotating cylindrical,long human bones under the magnetic field can have an *** findings of the study are offered in ***,a comparison with the results of the literature is conducted to show the effect of rotation and magnetic field on the wave propagation *** is worth noting that the results of the study highly match those of the literature.

关键词： Propagation of waves rotation magnetic field poroelastic wet bone natural frequency magnetic field

来源：评论

学校读者我要写书评

暂无评论

Widely accessible method for 3D microflow mapping at high spatial and temporal resolutions

引用

Microsystems & Nanoengineering 2022年第4期8卷 57-71页

作者： Evan Lammertse Nikhil Koditala Martin Sauzade Hongxiao Li Qiang Li Luc Anis Jun Kong Eric Brouzes Department of Biomedical Engineering Stony Brook UniversityStony BrookNY 11794USA Department of Mathematics and Statistics Department of Computer ScienceGeorgia State UniversityAtlantaGA 30302USA Laufer Center for Physical and Quantitative Biology Stony Brook UniversityStony BrookNY 11794USA Cancer Center Stony Brook School of MedicineStony BrookNY 11794USA Institute for Engineering Driven Medicine Stony Brook UniversityStony BrookNY 11794USA

Advances in microfluidic technologies rely on engineered 3D flow patterns to manipulate samples at the ***,current methods for mapping flows only provide limited 3D and temporal resolutions or require highly specialized optical ***,we present a simple defocusing approach based on brightfield microscopy and open-source software to map micro-flows in 3D at high spatial and temporal *** workflow is both integrated in ImageJ and *** track seed particles in 2D before classifying their Z-position using a reference *** compare the performance of a traditional cross-correlation method and a deep learning model in performing the classification *** validate our method on three highly relevant microfluidic examples:a channel step expansion and displacement structures as single-phase flow examples,and droplet microfluidics as a twophase flow ***,we elucidate how displacement structures efficiently shift large particles across ***,we reveal novel recirculation structures and folding patterns in the internal flow of microfluidic *** simple and widely accessible brightfield technique generates high-resolution flow maps and it will address the increasing demand for controlling fluids at the microscale by supporting the efficient design of novel microfluidic structures.

关键词： flow resolution method

来源：评论

学校读者我要写书评

暂无评论

The technology of rob water treatment becomes clean water by using the combination of electrolysis, filtration, and ozonation methods 10

The technology of rob water treatment becomes clean water by...

引用

10th International Seminar on New Paradigm and Innovation of Natural Sciences and itsApplication, ISNPINSA 2020

作者： Wahyono, Y. Nofanda, T.T. Budiyanto, I. Nugroho, H. Rosyad, M.S. Halim, M.A.R. Riski, I. Amilia, A. Rahmawati, D.F. Nurauliyaa, A.H. Baihaqi, R.A. Syafira, F. Lestari, P. Doctoral Program of Environmental Science School of Postgraduate Studies Diponegoro University Indonesia Department of Physics Faculty of Science and Mathematics Diponegoro University Indonesia Department of Mathematics Faculty of Science and Mathematics Diponegoro University Indonesia Department of Chemistry Faculty of Science and Mathematics Diponegoro University Indonesia Department of Chemical Engineering Faculty of Engineering Diponegoro University Indonesia

Coastal residents have difficulty in fulfilling the needs of clean water due to tidal water and tidal water intrusion. Meanwhile, clean water is a crucial component in carrying out activities of daily life. This initiative aims to provide an alternative solution in the form of technology to treat water polluted by saline water into clean water. This initiative used the method of electrolysis, filtration, and ozonation. Electrolysis has functions to remove salt contained and reduce levels of heavy metals while filtration functions to filter out the impurities in the form of solid particles, and ozonation takes a role to kill bacteria. Moreover, the target output of this work is the creation of a prototype by using this technology as a technology that can treat water polluted by saline water into clean water. The parameters assessed from this initiative are physical, chemical, and biological parameters of clean water. The results of the total test of dissolved solids of water before processing were 7130 ppm while it was 2490 ppm after getting treatment. The decreasing total value of dissolved solids is 65.077%, which indicates that the test was successful. The result of an electrical conductivity test of water before processing is 310 mA while after processing is 88.6 mA. The decrease in the current value of 71.419% indicates that the test was successful. The results of the turbidity test of water before processing are 0 FAU while after getting treatment is 0 FAU. The turbidity value remains 0, which indicates the test was successful. The results of the pH test of water before processing were 5.7 while the result after treatment got 6.1. The increase in pH value towards pH 7 (neutral) indicates that the test was successful. The test results of salt and heavy metals (Fe, Cu, Cl, Na, and Pb) of water before processing were 0.073 ppm, 0.07 ppm, 0.17 ppm, 0.04 ppm, and 4.24 ppm while after processing was 0 ppm, 0 ppm, 0.07 ppm, 0.03 ppm, and 1.36 ppm. There was a decrease

关键词： Ozonization

来源：评论

学校读者我要写书评

暂无评论

Predefined-Time Tracking Control of Second-Order Multi-agent Systems with State Constraints

Predefined-Time Tracking Control of Second-Order Multi-agent...

引用

Mechatronics, Control and Robotics (ICMCR), International Conference on

作者： Chaoqun Guo Jiangping Hu Ju H. Park Bijoy Kumar Ghosh School of Automation Engineering University of Electronic Science and Technology of China Chengdu China Department of Electrical Engineering Yeungnam University Gyeongsan Republic of Korea Department of Mathematics and Statistics Texas Tech University Lubbock USA

The predefined-time tracking control of an uncertain second-order multi-agent system is studied in this paper. To tackle both the unconstrained and constrained situations simultaneously, a uniform nonlinear transformation function is proposed. In addition, a time base generator (TBG) function is constructed to facilitate designing a predefined-time control. Moreover, to simplify the computational complexity of the controller design, a predefined-time characteristic filter is proposed by utilizing the TBG function. And also based on the TBG function, an adaptive law is developed to estimate the unknown norms of the fuzzy weights. Consequently, a predefined-time tracking controller based on the TBG technique is designed to achieve the state constraints and consensus tracking. At last, to validate the predefined-time control strategy for the uncertain nonlinear multi-agent system, the simulations are presented.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Josephson-Anderson relation as diagnostic of turbulent drag reduction by polymers

引用

Physical Review E 2025年第1期111卷 015105-015105页

作者： Samvit Kumar Simon Toedtli Tamer A. Zaki Gregory L. Eyink Department of Applied Mathematics & Statistics Johns Hopkins University Baltimore Maryland 21218 USA Department of Mechanical Engineering Johns Hopkins University Baltimore Maryland 21218 USA Department of Physics & Astronomy The Johns Hopkins University Baltimore Maryland 21218 USA

The detailed Josephson-Anderson relation, which equates instantaneously the volume-integrated vorticity flux and the work by pressure drop, has been the key to drag reduction in superconductors and superfluids. We employ a classical version of this relation to investigate the dynamics of polymer drag-reduced channel flows, particularly in the high-extent drag reduction (HDR) regime which is known to exhibit strong space-time intermittency. We show that high drag is not created instantaneously by near-wall coherent vortex structures as assumed in prior works. These predominantly spanwise near-wall vortex structures can produce a net “up-gradient” flux of vorticity toward the wall, which instead reduces instantaneous drag. Increase of wall vorticity and skin friction due to this up-gradient flux occurs after an apparent lag of several advection times, increasing with the Weissenberg number. This increasing lag appears to be due to polymer damping of up-gradient nonlinear vorticity transport that arises from large-scale eddies in the logarithmic layer. The relatively greater polymer damping of down-gradient transport due to small-scale eddies results in lower net vorticity flux and hence lower drag. The Josephson-Anderson relation thus provides an exact tool to diagnose the mechanism of polymer drag reduction in terms of vorticity dynamics and it explains also prior puzzling observations on transient drag reduction, as for centerline-release experiments in pipe flow.

关键词： Vorticity

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：