检索结果-内蒙古大学图书馆

IEEE Conference on Wireless Communications and Networking

作者： Mohamed Benzaghta Giovanni Geraci David López-Pérez Alvaro Valcarce Univ. Pompeu Fabra Barcelona Spain Telefónica Research Barcelona Spain Univ. Politècnica de València Spain Nokia Bell Labs Massy France

ISBN: (数字)9798350368369

ISBN: (纸本)9798350368376

We propose a data-driven approach for large-scale cellular network optimization, using a production cellular network in London as a case study and employing Sionna ray tracing for site-specific channel propagation modeling. We optimize base station antenna tilts and half-power beamwidths, resulting in more than double the 10%-worst user rates compared to a 3GPP baseline. In scenarios involving aerial users, we identify configurations that increase their median rates fivefold without compromising ground user performance. We further demonstrate the efficacy of model generalization through transfer learning, leveraging available data from a scenario source to predict the optimal solution for a scenario target within a similar number of iterations, without requiring a new initial dataset, and with a negligible performance loss.

关键词： Cellular networks Transfer learning Production Ray tracing Predictive models linear programming Data models 3GPP Optimization Antennas

来源：评论

学校读者我要写书评

暂无评论

Accelerating Surrogate-Assisted Multi-Objective Evolutionary Optimization via a Dynamic Surrogate Model with Archive Updating

Accelerating Surrogate-Assisted Multi-Objective Evolutionary...

引用

International Conference on Knowledge and Smart Technology (KST)

作者： Zi Wang Yu Ding Yawen Zou Chao Zhang Jun Yu Graduate School of Science and Technology Niigata University Niigata-city Japan Graduate School of Science and Engineering University of Toyama Toyama-city Japan Faculty of Engineering University of Toyama Toyama-city Japan Institute of Science and Technology Niigata University Niigata-city Japan

ISBN: (数字)9798331520403

ISBN: (纸本)9798331520410

In surrogate-assisted evolutionary computation, a common method for constructing surrogate models is the Gaussian Process (GP), which is widely used because of its flexibility in handling complex objective functions. However, GP can lead to an increased computation time when the sample size is large or when more samples are added to update the surrogate model. Furthermore, limited metrics are available for selecting the optimal surrogate model for a given problem, necessitating experimentation with each method to determine the best choice. In this paper, we propose a multi-objective evolutionary algorithm that addresses these issues by introducing an archive updating strategy and a dynamic surrogate model. In the experiment, by controlling the learning data with archive updating and dynamically selecting appropriate models from multiple surrogate models based on evaluation metrics, the computation time can be reduced without compromising the accuracy.

关键词： Measurement Accuracy Computational modeling Heuristic algorithms Training data Evolutionary computation Gaussian processes linear programming Data models Optimization

来源：评论

学校读者我要写书评

暂无评论

Research on Planting Strategy Based on Simulated Annealing Algorithm and Monte Carlo Modeling

Research on Planting Strategy Based on Simulated Annealing A...

引用

Electronics, Energy Systems and Power Engineering (EESPE), IEEE International Conference on

作者： Yingrou Xu Liming Chen Guangdong University of Technology Guangzhou China

ISBN: (数字)9798350389579

ISBN: (纸本)9798350389586

The aim of this study is to optimize the planting strategy of crops based on simulated annealing algorithm and Monte Carlo model in order to enhance the efficiency of resource utilization and sustainable economic development. By analyzing the crop planting data, a crop planting model was established, and the Spearman correlation coefficient was applied to explore the substitutability and complementarity among different crops and the correlation of economic indicators, so as to optimize the planting scheme. The study is divided into three main parts. First, for different plot types, suitable crops and their related information were analyzed to provide a basis for the establishment of objective functions and constraints. Secondly, the implementation results of the simulated annealing algorithm showed that the optimal planting scheme for the future years was obtained after parameter adjustment, reflecting the impact of different cooling coefficients on the yield. Finally, combining with Spearman coefficient analysis, the substitution and complementary relationship between crops was explored, and the greedy strategy was applied to further optimize the planting combinations to ensure a steady increase in returns under controlled risks. The advantage of this study lies in the adoption of a comprehensive model, which can effectively deal with the uncertainty in crop cultivation and enhance the scientific of decision-making. Through the combination of simulation and optimization methods, it provides a feasible planting strategy for economic development, which has important theoretical and practical significance.

关键词： Correlation coefficient Analytical models Solid modeling Monte Carlo methods Uncertainty Crops Simulated annealing linear programming Data models Resource management

来源：评论

学校读者我要写书评

暂无评论

Frank-Wolfe Method with Proximal Regularization for Constrained Federated Learning with Non-iid Data

Frank-Wolfe Method with Proximal Regularization for Constrai...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Robin Francis Sundeep Prabhakar Chepuri Department of Electrical Communication Engineering Indian Institute of Science Bengaluru India

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Federated constrained learning allows us to learn a global model with some specific structure to enhance performance. Most existing federated learning techniques assume data is homogeneously or independently and identically distributed (iid) across clients. However, this iid assumption rarely holds in practice, and the averaging operation on the server’s side leads to degradation in performance at the client level. In this paper, we propose FedFW-Prox, a novel and computationally efficient federated Frank-Wolfe algorithm that learns personalized models for each client, which guarantees better performance even with heterogeneous data distributions across clients. Instead of modifying the objective function, we introduce a proximal regularization term at the update step at each client, simplifying the update process. This regularization term ensures that local models remain close to the global model, promoting convergence to the global optimal point rather than the local optima. The proposed method guarantees a sublinear convergence of $\mathcal{O}(1/\sqrt k )$ for smooth convex objective functions, where k denotes the iteration number. We empirically evaluate the effectiveness of the proposed approach on various machine-learning tasks.

关键词： Federated learning Computational modeling Signal processing algorithms Signal processing linear programming Data models Computational efficiency Indexes Speech processing Convergence

来源：评论

学校读者我要写书评

暂无评论

High-Speed Resource Allocation for Multi-User NOMA Systems Using a Coherent Ising Machine

High-Speed Resource Allocation for Multi-User NOMA Systems U...

引用

International Conference on Information Networking

作者： Shuho Ishibashi Maki Arai Aohan Li Hiroki Takesue Kensuke Inaba Kazuyuki Aihara Mikio Hasegawa Department of Electrical Engineering Graduate School of Engineering Tokyo University of Science 6-3-1 Niijuku Katsushika-ku Tokyo Japan Graduate School of Informatics and Engineering The University of Electro-Communications 1-5-1 Chofugaoka Chofu-shi Tokyo Japan NTT Basic Research Laboratories NTT Corporation Atsugi Kanagawa Japan International Research Center for Neurointelligence The University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo Japan

ISBN: (数字)9798331506940

ISBN: (纸本)9798331506957

Non-Orthogonal Multiple Access (NOMA) is a nextgeneration communication technology that enables multiple users share the same wireless channels. To optimize the performance of NOMA, appropriate channel allocation for each user is essential. Previous research has applied a fast optimization approach using the coherent Ising machine (CIM) to channel allocation in NOMA, multiplexing up to two users per channel, which can be formulated as a basic Ising Hamiltonian including up to second-order interactions. However, NOMA can support multiplexing more than two users per channel depending on channel conditions. This paper propose a method for optimizing NOMA systems with more than two multiplexed users per channel, which becomes a Higher Order Binary Optimization problem. We apply a method to reduce the order of the objective function, converting it to Quadratic Unconstrained Binary Optimization, and derive the Ising Hamiltonian for higher-order NOMA. We evaluate our proposed method through simulations using the CIM for optimization. Our results show that the proposed method leads to better solutions than conventional methods.

关键词： Multiplexing Wireless communication NOMA Simulation Channel allocation linear programming Resource management Optimization

来源：评论

学校读者我要写书评

暂无评论

Low-Complexity Cramér-Rao Lower Bound and Sum Rate Optimization in ISAC Systems

Low-Complexity Cramér-Rao Lower Bound and Sum Rate Optimiza...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Tianyu Fang Nhan Thanh Nguyen Markku Juntti Centre for Wireless Communications University of Oulu Finland

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

While Cramér-Rao lower bound is an important metric in sensing functions in integrated sensing and communications (ISAC) designs, its optimization usually involves a computationally expensive solution such as semidefinite relaxation. In this paper, we aim to develop a low-complexity yet efficient algorithm for CRLB optimization. We focus on a beamforming design that maximizes the weighted sum between the communications sum rate and the sensing CRLB, subject to a transmit power constraint. Given the non-convexity of this problem, we propose a novel method that combines successive convex approximation (SCA) with a shifted generalized power iteration (SGPI) approach, termed SCA-SGPI. The SCA technique is utilized to approximate the non-convex objective function with convex surrogates, while the SGPI efficiently solves the resulting quadratic subproblems. Simulation results demonstrate that the proposed SCA-SGPI algorithm not only achieves superior tradeoff performance compared to existing method but also significantly reduces computational time, making it a promising solution for practical ISAC applications.

关键词： Measurement Lower bound Array signal processing Simulation Signal processing algorithms Integrated sensing and communication Approximation algorithms linear programming Speech processing Optimization

来源：评论

学校读者我要写书评

暂无评论

Learning Strategy with Barlow Twins Objective for Emotion-Robust Speaker Verification System

Learning Strategy with Barlow Twins Objective for Emotion-Ro...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Dmitrii V. Mikhailovskii Aki Kunikoshi David A. van Leeuwen Jaebok Kim Radboud University Nijmegen ReadSpeaker B.V.

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

A recent trend of the automatic speaker verification (ASV) systems is to use speaker representations extracted from deep learning-based speaker encoders. Those representations are robust against linguistic variation but sensitive to emotional fluctuation that potentially degrades the performance of ASV systems. To tackle this issue, we applied a correlation-based objective function called Barlow Twins objective to speaker representation learning for the ASV task with expressive speech. This helps the learned representations from the same speaker become similar despite their emotional state. Our experiment showed that our representation learning using Barlow Twins objective improves the standard deviation of the EERs between different emotions from 2.43 to 1.52 and the total EER from 10.77% to 6.47% on the CREMA-D dataset. The study also evaluates the system’s generalization performance across out-of-domain datasets, demonstrating improved standard deviation of the EERs and total EER of the baseline system in all datasets while observing the performance drop on synthetic emotional data where emotional mismatch occurs.

关键词： Representation learning Hands Fluctuations Signal processing Linguistics Market research linear programming Speech processing Standards Synthetic data

来源：评论

学校读者我要写书评

暂无评论

HQC-Bend: A Python Package of Hybrid Quantum-Classical Multi-cuts Benders’ Decomposition Algorithm

HQC-Bend: A Python Package of Hybrid Quantum-Classical Multi...

引用

Quantum Communications, Networking, and Computing (QCNC), International Conference on

作者： Zhongqi Zhao Mingze Li Lei Fan Zhu Han Department of Electrical and Computer Engineering University of Houston Houston TX USA Department of Engineering Technology University of Houston Houston TX USA

ISBN: (数字)9798331531591

ISBN: (纸本)9798331531607

This article introduces the Hybrid Quantum-Classical Multi-Cut Benders’ Decomposition (HQC-Bend) algorithm, an efficient, open-source Python script designed to tackle complex Mixed-Binary linear programming (MBLP) problems with a block structure by integrating quantum and classical computing methods. HQC-Bend decomposes large MBLP models into a master problem and smaller, manageable subproblems, iteratively refining solutions to accelerate convergence. By leveraging quantum computing for specific computational tasks and classical methods for others, the algorithm significantly enhances efficiency. Supported by D-Wave for quantum solving and Gurobi for classical solving, the demonstration includes two primary modes of problem solving: a purely classical iteration and a hybrid quantum-classical approach. Experimental validations on real-world cases demonstrate the algorithm’s effectiveness and offer practical insights, highlighting its potential for broad adoption across various topics and fields.

关键词： Knowledge engineering Quantum computing Computational modeling Refining linear programming Problem-solving Optimization Python Quantum communication Convergence

来源：评论

学校读者我要写书评

暂无评论

Can LIME Make Interactive Multiobjective Optimization Methods Explainable?

Can LIME Make Interactive Multiobjective Optimization Method...

引用

Trustworthy, Explainable and Responsible Computational Intelligence (CITREx), IEEE Symposium on

作者： Afsar Bekir Giomara Lárraga Kaisa Miettinen Faculty of Information Technology University of Jyvaskyla Finland

ISBN: (数字)9798331520151

ISBN: (纸本)9798331520168

Real-world optimization problems often involve multiple (typically conflicting) objective functions that must be balanced since they cannot reach their optima simultaneously. Interactive multiobjective optimization methods are powerful in solving such problems because they incorporate the preference information of a domain expert, a decision maker (DM), in the solution process to find the most preferred trade-offs. However, the connection between the preferences and the solutions generated based on them may remain unclear, limiting a DM’s understanding and confidence in the solution process. Therefore, we propose a novel approach leveraging local interpretable model-agnostic explanations (LIME) to make interactive methods *** generate local linear approximations of the set of solutions that reflect different trade-offs to provide a DM with insights into the behavior of objective functions near the solutions considered. We can explain how changes in one objective function value affect others, assisting a DM in adjusting their preferences during the interactive solution process to get more preferred solutions. We utilize Karush-Kuhn-Tucker multipliers to construct the linear model analogous to LIME. This is convenient since the multipliers can readily be obtained as a byproduct of applying an appropriate numerical optimization *** our knowledge, this is the first application of LIME-inspired ideas to the domain of interactive multiobjective optimization. The proposed approach can significantly enhance the transparency and trustworthiness of interactive methods, ultimately leading to better decisions.

关键词： Limiting Computational modeling Decision making Optimization methods linear approximation linear programming Numerical models Computational efficiency Reliability Computational intelligence

来源：评论

学校读者我要写书评

暂无评论

L₂,₁-Norm Regularized Robust and Sparse linear Discriminant Analysis via an Alternating Direction Method of Multipliers

引用

IEEE Access 2023年 11卷 34250-34259页

作者： Chun-Na Li Yi Li Yan-Hui Meng Pei-Wei Ren Yuan-Hai Shao Management School Hainan University Haikou China School of Science Hainan University Haikou China Clinical Laboratory Guangdong Provincial Fertility Hospital Guangzhou China

Recently, an absolute value inequalities discriminant analysis criterion with robustness and sparseness for supervised dimensionality reduction was studied. However, it obtains discriminant directions one by one through greedy search, which makes the sparseness of multiple discriminant directions unexplainable. In addition, it relaxes the original problem into a series of linear programming problems which makes it time consuming. In this paper, we construct a novel linear discriminant analysis with robustness and sparseness jointly through the $L_linear programming$ -norm and $L_{2,1}$ -norm. The proposed approach obtains all the discriminant directions simultaneously, and rather than solving linear programming problems, it is solved by a more effective alternating direction method of multipliers. The effectiveness of the proposed method is supported by preliminary experimental results on two artificial datasets, some benchmark datasests and two face image datasets.

关键词： linear discriminant analysis Robustness Feature extraction Optimization linear programming Principal component analysis linear matrix inequalities

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：