检索结果-内蒙古大学图书馆

Dengue-Infected Mosquito Detection with Uncertainty Evaluation based on Monte Carlo Dropout

IEEE LATIN AMERICA TRANSACTIONS 2025年第2期23卷 135-143页

作者： Torres, Israel Nakano, Mariko Cime-Castillo, Jorge Escamilla-Hernandez, Enrique Lopez-Garcia, Osvaldo Lanz-Mendoza, Humberto Inst Politecn Nacl Culhuacan Coyoacan Mexico Inst Nacl Salud Publ Cuernavaca Morelos Mexico

Considering Aedes mosquitoes are a principal vector of the Dengue virus causing, in the worst case, the death of infected people, accurate detection of infected Aedes mosquitoes is very important to prevent the further spread of the virus. In this paper, we propose a detection algorithm for infected Aedes aegypti mosquitoes by Dengue Virus-2 using their wingbeat signals. The proposed algorithm uses Long Short-Term Memory (LSTM) as a classifier of the input wingbeat signal into healthy mosquitoes and infected mosquitoes. All living beings, even if they are of the same species, have different characteristics depending on the season in which they are born, temperature, humidity, food, etc. This individual difference perhaps influences the level of infection, although it is fed by the same infected blood. Considering these differences between individuals, we introduce an uncertainty measure based on Monte-Carlo dropout. The proposed algorithm detects approximately 5% of uncertainty data from all input wingbeat signals in the test set and provides a classification accuracy of 94.87% without any uncertainty.

关键词： Uncertainty Long short term memory Vectors Viruses (medical) Monte Carlo methods Logic gates Training Accuracy Classification algorithms approximation algorithms dengue fever infected mosquitoes long short-term memory Monte Carlo dropout uncertainty

来源：评论

学校读者我要写书评

暂无评论

Machine Unlearning: Taxonomy, Metrics, Applications, Challenges, and Prospects

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2025年 PP卷 PP页

作者： Li, Na Zhou, Chunyi Gao, Yansong Chen, Hui Zhang, Zhi Kuang, Boyu Fu, Anmin Nanjing Univ Sci & Technol Sch Cyber Sci & Engn Nanjing 210094 Peoples R China Xidian Univ State Key Lab Integrated Serv Networks Xian 710071 Peoples R China Zhejiang Univ Coll Comp Sci & Technol Hangzhou 310058 Peoples R China Univ Western Australia Dept Comp Sci & Software Engn Perth WA 6009 Australia Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Peoples R China

Personal digital data is a critical asset, and governments worldwide have enforced laws and regulations to protect data privacy. Data users have been endowed with the "right to be forgotten" (RTBF) of their data. In the course of machine learning (ML), the forgotten right requires a model provider to delete user data and its subsequent impact on ML models upon user requests. Machine unlearning (MU) emerges to address this, which has garnered ever-increasing attention from both industry and academia. Specifically, MU allows model providers to eliminate the influence of unlearned data without retraining the model from scratch, ensuring the model behaves as if it never encountered this data. While the area has developed rapidly, there is a lack of comprehensive surveys to capture the latest advancements. Recognizing this shortage, we conduct an extensive exploration to map the landscape of MU including the (fine-grained) taxonomy of unlearning algorithms under centralized and distributed settings, debate on approximate unlearning, verification and evaluation metrics, and challenges and solutions across various applications. We also focus on the motivations, challenges, and specific methods for deploying unlearning in large language models (LLMs), as well as the potential attacks targeting unlearning processes. The survey concludes by outlining potential directions for future research, hoping to serve as a beacon for interested scholars.

关键词： Data models Surveys Measurement Training Electronic mail Data privacy Taxonomy General Data Protection Regulation Computational modeling approximation algorithms federated learning (FL) large language model (LLM) machine learning (ML) machine unlearning (MU)

来源：评论

学校读者我要写书评

暂无评论

Architecting Path Selection Method for Incremental Evolution in System-of-Systems

引用

IEEE SYSTEMS JOURNAL 2025年

作者： Fang, Zhemei Chen, Dazhi Ju, Qi Wang, Jianbo Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Natl Key Lab Sci & Technol Multispectral Informat Wuhan 430074 Peoples R China China Ship Dev & Design Ctr Wuhan 430064 Peoples R China

Architecture design for system-of-systems (SoSs) is a complex challenge due to interdependencies, uncertainties, and the large design space. The evolutionary nature of SoSs necessitates a multistage architecting process, adding further complexity. This article, thus, proposes a deep reinforcement learning based evolutionary architecture path selection method that considers uncertainties and interdependency. The approach employs an architecture framework to guide the design and defines SoS architecture decisions as the addition of systems and the allocation of operational architecture to physical architecture across sequential stages. Capability evaluation leverages a capability-activity-system structure, supported by a functional dependency network analysis method. Utilizing a deep neural network as a functional approximator to predict future SoS capability, the article develops a proximal policy optimization (PPO) algorithm that balances immediate and future needs. Applied to a mosaic warfare-oriented naval antisubmarine SoS, the proposed method outperforms heuristic optimization techniques by achieving higher SoS capability, reduced instability, and fewer violations of budget and intermediate requirements constraints in both deterministic and stochastic scenarios. These results highlight the PPO method's effectiveness in addressing SoS architecting path selection challenges under uncertainty.

关键词： Computer architecture Uncertainty Computational modeling Decision making Architecture Standards Resource management Prediction algorithms Planning approximation algorithms Architecture design capability aggregation incremental evolution reinforcement learning (RL) system-of-systems (SoSs)

来源：评论

学校读者我要写书评

暂无评论

A Reconfigurable Floating-Point Division and Square Root Architecture for High-Precision Softmax

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS 2025年

作者： Fang, Xiwei Wang, Yuhan Chen, Lei An, Fengwei Southern Univ Sci & Technol Sch Microelect Shenzhen 518055 Peoples R China PengCheng Lab Shenzhen 518055 Peoples R China

With the advancement of deep learning models, the Softmax function with self-attention has become pervasive in everyday applications. As components of the Softmax function and its inputs, both division and square root operations impact its accuracy. However, these two non-linear operations bring significant area and power consumption for hardware implementation. To address these challenges, this paper proposes a reconfigurable floating-point division and square root (FDSR) architecture that achieves low resource consumption and high accuracy for general-purpose computation. The FDSR enhances the traditional non-restoring algorithm by using shift-registers and optimizing the leading-one detection and shift operations, reducing hardware resource usage while maintaining high accuracy (0.5 ULP). In the mantissa calculation, the division can be converted to a square root operation by simply switching the input to the subtractor through multiplexers. Additionally, a triple-mode reconfigurable iteration unit is introduced, featuring a multi-layer variable pipeline architecture to improve adaptability for different applications. By redesigning the pipeline depth and reusing logical units, the FDSR effectively addresses the issue of lengthy iteration cycles in the non-restoring method. Implementation results using 40nm CMOS technology demonstrate that the proposed design achieves a 76.49% power reduction and a 14.69% area reduction for floating-point division compared to Synopsys Design Ware and an 88.05% power reduction and a 90.57% area reduction for floating-point square root. With 28 nm CMOS technology, the FDSR reduces power consumption by 91.55% and reduces area by 64.39% for floating-point division compared to Synopsys Design Ware. On the FPGA platform, the FDSR significantly reduces hardware resource consumption, achieving an 85.23% reduction for floating-point division and 87.81% for floating-point square root, outperforming state-of-the-art designs.

关键词： Accuracy approximation algorithms Computer architecture Hardware Transformers Pipelines Training Power demand Integrated circuit modeling Heuristic algorithms Floating-point division floating-point square root low-resource consumption and high-accuracy technique reconfigurable iteration units

来源：评论

学校读者我要写书评

暂无评论

Semiglobal Safety-Filtered Extremum Seeking With Unknown CBFs

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2025年第3期70卷 1698-1713页

作者： Williams, Alan Krstic, Miroslav Scheinker, Alexander Univ Calif San Diego Dept Mech & Aerosp Engn San Diego CA 92093 USA Los Alamos Natl Lab Accelerator Operat & Technol Appl Electrodynam AOT Los Alamos NM 87545 USA

We introduce a safe extremum-seeking algorithm that achieves the minimization of an unknown objective function while ensuring that an unknown, yet measured, control barrier function (CBF) remains above an arbitrarily small negative value for all time. In other words, "practical safety" is maintained during the entire period of convergence to the constrained extremum. Our design is based on quadratic program (QP) CBF style filters for safety, which is applied in an average and estimated sense. Using nonsmooth analysis tools, we guarantee semiglobal practical asymptotic (SPA) stability of the global constrained optimum, practical convergence to the safe set if starting in a condition violating the CBF, and practical safety for all time-semiglobally-if starting in safe set. The safety result of this article is analogous with modern notions of SPA stability, guaranteeing that, for any small violation of safety, there exist design coefficients that guarantee that such a small violation is not exceeded. This article outlines a set of sufficient conditions on the barrier and objective functions, and by way of a Lyapunov argument, we demonstrate that nonconvex constrained optimization problems can be solved. We present these results in the setting of a static map and a dynamical system. A simulation example illustrates the results.

关键词： Safety Heuristic algorithms Asymptotic stability Optimization Filtering algorithms Thermal stability Switches Convergence approximation algorithms Stability criteria Constrained optimization extremum seeking (ES) safe control

来源：评论

学校读者我要写书评

暂无评论

A Framework for Energy Efficiency Optimization in IRS-Aided Hybrid MU-MIMO Systems

引用

IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS 2025年第3期43卷 883-898页

作者： Ju, Xin Liu, Heng Gong, Shiqi Xing, Chengwen Zhao, Nan Niyato, Dusit Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China Beijing Inst Technol Sch Cyberspace Sci & Technol Beijing 100081 Peoples R China Dalian Univ Technol Sch Informat & Commun Engn Dalian 116024 Peoples R China Nanyang Technol Univ Coll Comp & Data Sci Singapore Singapore

Energy efficiency (EE) optimization has attracted significant research attention for implementing green communications. With cost-effective and low-power advantages, intelligent reflecting surface (IRS) and hybrid analog-digital transceiver have recently emerged as two promising technologies of next-generation green wireless systems. In this paper, we propose a comprehensive framework for EE optimization in four types of IRS-aided hybrid analog-digital multiuser multiple-input multiple-output communication systems, including the uplink (UL) systems under the sum power and box eigenvalue constraints as well as the per-radio-frequency chain power constraints (PRPCs), and the downlink (DL) systems under the sum power constraint and the PRPCs. This framework proposes a unified design methodology to these four considered systems by separating the optimization of analog and digital matrix variables. Specifically, for the UL EE maximization problems, we firstly propose a channel alignment based algorithm to separately optimize the analog precoders at users, the analog combiner at the base station and the IRS reflecting matrix, whose computational complexity is significantly reduced as compared with the traditional alternating optimization algorithm. Then, by introducing the auxiliary variables and exploiting the Karush-Kuhn-Tucker conditions based algorithm, the optimal digital precoders at users are obtained in closed forms. Furthermore, the intractable DL EE optimization can be equivalently transformed into its virtual UL counterpart using the DL-UL duality, leading to the general applicability of the proposed framework. Extensive simulations reveal that the proposed algorithm attains the almost identical EE performance to the traditional benchmarks with a lower computational complexity.

关键词： Optimization Radio frequency Analog-digital conversion Transceivers Ink Hybrid power systems Linear programming approximation algorithms Massive MIMO Eigenvalues and eigenfunctions Energy efficiency duality hybrid analog-digital system intelligent reflecting surface multiuser multiple-input multiple-output

来源：评论

学校读者我要写书评

暂无评论

algorithms for Cameras View-Frame Placement Problems in the Presence of an Adversary and Distributional Ambiguity

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 353-364页

作者： Park, Seonghun Bansal, Manish Virginia Tech Grado Dept Ind & Syst Engn Blacksburg VA 24061 USA

In this paper, we introduce cameras view-frame placement problem (denoted by CFP) in the presence of an adversary whose objective is to minimize the maximum coverage by p cameras in response to input provided by n autonomous agents in a remote location. We allow uncertainty in the success of attacks, incomplete information of the probability distribution associated with the uncertain data, and varying levels of risk-appetite of the adversary. We present an exact cutting planes based algorithm to solve this problem and provide conditions under which it is finitely convergent. Since this approach solves deterministic CFP in each iteration, we also present improved exact method for CFP with p = 1, approximation algorithm and heuristics for Multi-CFP with p >= 2, and Multi-CFP with fixed tilt of the cameras. To evaluate the effectiveness and performance of the proposed approaches, we conduct computational experiments using randomly generated instances and simulation experiments where these approaches are utilized to find a hidden object in a remote location.

关键词： Cameras approximation algorithms Uncertainty Robot vision systems Random variables Probability distribution Videos Adversarial camera view-frame placement distributionally robust optimization pan-tilt-zoom cameras exact and approximate algorithm cutting planes

来源：评论

学校读者我要写书评

暂无评论

Parameterized inapproximability of Morse matching

引用

COMPUTATIONAL GEOMETRY-THEORY AND APPLICATIONS 2025年 126卷

作者： Bauer, Ulrich Rathod, Abhishek Tech Univ Munich Munich Germany Ben Gurion Univ Negev Beer Sheva Israel

We study the problem of minimizing the number of critical simplices from the point of view of inapproximability and parameterized complexity. We first show inapproximability of MIN-MORSE MATCHING within a factor of 2 log ( 1-r) n . Our second result shows that MIN- MORSE MATCHING is W [ P ]-hard with respect to the standard parameter. Next, we show that MIN-MORSE MATCHING with standard parameterization has no FPT approximation algorithm for any approximation factor rho . The above hardness results are applicable to complexes of dimension >= 2. On the positive side, we provide a factor O( n log n ) approximation algorithm for MIN-MORSE MATCHING on 2-complexes, noting that no such algorithm is known for higher dimensional complexes. Finally, we devise discrete gradients with very few critical simplices for typical instances drawn from a fairly wide range of parameter values of the Costa-Farber model of random complexes. (c) 2024 Published by Elsevier B.V.

关键词： Discrete Morse theory approximation algorithms Parameterized complexity Stochastic topology

来源：评论

学校读者我要写书评

暂无评论

DGLG: A Novel Deep Generalized Legendre-Galerkin Approach to Optimal Filtering Problem

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2025年第4期70卷 2584-2590页

作者： Shi, Ji Jiao, Xiaopei Yau, Stephen S. -T. Capital Normal Univ Acad Multidisciplinary Studies Beijing 100048 Peoples R China Univ Twente Dept Appl Math Enschede Netherlands Beijing Inst Math Sci & Applicat Beijing 101400 Peoples R China Tsinghua Univ Dept Math Sci Beijing 100084 Peoples R China

The optimal filtering problem for general nonlinear and continuous state-observation systems attracts lots of attention in the control theory. The essence of optimal filtering requires solving the Duncan-Mortensen-Zakai (DMZ) equation in a computationally feasible way. Under the pioneering work of Yau-Yau filtering, the DMZ equation is reduced to a pathwise computation of a forward Kolmogorov equation with time-varying initial conditions, which is very challenging. To overcome the computational difficulty, in this article, we proposed a new efficient filtering algorithm consisting of a forward Kolmogorov equation solver based on a physics-informed neural network and a probability density approximator based on generalized Legendre polynomials. By utilizing the advanced deep learning method and classical Galerkin approximation, our developed algorithm not only maintains the high accuracy of the spectral method but also removes massive computational loads in the offline part. Furthermore, the convergence of our method is proved. Numerical experiments have been carried out to verify the feasibility of the new method. Regarding accuracy and efficacy, the newly proposed deep generalized Legendre-Galerkin algorithm outperforms other popular suboptimal methods including the extended Kalman filter and particle filter.

关键词： Filtering algorithms Filtering Filtering theory Kalman filters approximation algorithms Real-time systems Estimation Convergence Probability density function Polynomials filtering generalized Legendre polynomial (GLP) neural networks nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

Approximating average bounded-angle minimum spanning trees

引用

COMPUTATIONAL GEOMETRY-THEORY AND APPLICATIONS 2025年 128卷

作者： Biniaz, Ahmad Bose, Prosenjit Devaney, Patrick Univ Windsor Sch Comp Sci Windsor ON Canada Carleton Univ Sch Comp Sci Ottawa ON Canada Univ British Columbia Dept Comp Sci Vancouver BC Canada

Motivated by the problem of orienting directional antennas in wireless communication networks, we study average bounded-angle minimum spanning trees. Let P be a set of points in the plane and let alpha be an angle. An alpha-spanning tree (alpha-ST) of P is a spanning tree of the complete Euclidean graph induced by P such that all edges incident to each point p is an element of P lie in a fixed wedge of angle alpha with apex p. An alpha-minimum spanning tree (alpha-MST) of P is an alpha-ST with minimum total edge length. An average-alpha-spanning tree (denoted by alpha(-)-ST) is a spanning tree with the relaxed condition that incident edges to all points lie in wedges with average angle alpha. An average-alpha-minimum spanning tree (alpha(-)-MST) is an alpha(-)-ST with minimum total edge length. Let A (alpha) be the smallest ratio of the length of the alpha(-)-MST to the length of the standard MST, over all sets of points in the plane. We investigate bounds for A (alpha). For alpha = 2 pi/3, Biniaz, Bose, Lubiw, and Maheshwari (Algorithmica 2022) showed that 4/3 <= A ( 2 pi/3) <= 3/2. We improve the upper bound and show that A ( 2 pi/3) <= 13/9 . We also study this for alpha = pi/2 and prove that 3 /2 <= A ( pi/ 2 ) <= 4. (c) 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

关键词： Bounded-angle MST Average bounded-angle MST Directional antenna approximation 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：