检索结果-内蒙古大学图书馆

Exploring Hyperdimensional Computing Robustness Against Hardware Errors

IEEE Transactions on computers 2025年第6期74卷 1963-1977页

作者： Zhang, Sizhe Juretus, Kyle Jiao, Xun Villanova University Department of Electrical and Computer Engineering VillanovaPA19085 United States

Brain-inspired hyperdimensional computing (HDC) is an emerging machine learning paradigm leveraging high-dimensional spaces for efficient tasks like pattern recognition and medical diagnostics. As a lightweight alternative to deep neural networks, HDC offers smaller model sizes, reduced computation, and memory-centric processing. However, deploying HDC in safety-critical applications, such as healthcare and robotics, is challenged by hardware-induced errors. This paper investigates HDC's robustness to memory errors via extensive bit-flip injection experiments on item and associative memories. Results reveal that certain bit-flips severely degrade accuracy. To address this, we introduce the Hyperdimensional Bit-Flip Search (HD-BFS), a similarity-guided method for identifying vulnerabilities and crafting efficient attacks, where flipping just 6 critical bits—3.9% of random bit-flips—reduces accuracy to chance levels. We further propose Hyperdimensional Accelerated Bit-Flip Search (HD-ABFS), which narrows the search space by targeting critical dimensions and most significant bits (MSBs), achieving up to 282x speedup over HD-BFS. Finally, we develop an effective protection mechanism to enhance model safety. These insights highlight HDC's resilience to random errors, offer robust defenses against targeted attacks, and advance the security and reliability of HDC systems. © 1968-2012 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning Data Collection for Bayesian Inference of Hidden Markov Models

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第5期6卷 1217-1232页

作者： Alali, Mohammad Imani, Mahdi Northeastern University Department of Electrical and Computer Engineering BostonMA02115 United States

Hidden Markov models (HMMs) are a powerful class of dynamical models for representing complex systems that are partially observed through sensory data. Existing data collection methods for HMMs, typically based on active learning or heuristic approaches, face challenges in terms of efficiency in stochastic domains with costly data. This article introduces a Bayesian lookahead data collection method for inferring HMMs with finite state and parameter spaces. The method optimizes data collection under uncertainty using a belief state that captures the joint distribution of system states and models. Unlike traditional approaches that prioritize short-term gains, this policy accounts for the long-term impact of data collection decisions to improve inference performance over time. We develop a deep reinforcement learning policy that approximates the optimal Bayesian solution by simulating system trajectories offline. This pretrained policy can be executed in real time, dynamically adapting to new conditions as data is collected. The proposed framework supports a wide range of inference objectives, including point based, distribution based, and causal inference. Experimental results across three distinct systems demonstrate significant improvements in inference accuracy and robustness, showcasing the effectiveness of the approach in uncertain and data-limited environments. © 2020 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Noninvasive estimation of blood potassium concentration using ECG and FCM-ANFIS model

引用

Research on Biomedical engineering 2024年第3-4期40卷 647-658页

作者： Samandari, Zeynab Molaeezadeh, Seyyedeh Fatemeh Department of Electrical and Computer Engineering Jundi-Shapur University of Technology Dezful Iran

Purpose: Potassium imbalance, often symptomless but potentially fatal, is prevalent in patients with kidney or heart conditions. Traditional laboratory tests for potassium measurement are costly and require skilled technicians. Additionally, phlebotomy is challenging and painful for high-risk patients. To address these issues, this study proposes a noninvasive and bloodless approach that leverages the electrocardiogram (ECG) signal, which may be sensitive to potassium changes. The research introduces a data-driven framework for ECG-based potassium measurement. Methods: The proposed framework comprises five main components: Cohort Selection—Among 461,178 patients in the ECG-ViEW II database, those with simultaneous ECG, Age-adjusted Charlson Comorbidity Index (ACCI), and potassium measurements within a 5-min interval are chosen;Data Labeling—Three labels—hypokalemia, normal, and hyperkalemia—are defined;Feature Extraction—Relevant features such as RR interval, PR interval, QRS duration, QT interval, QTc interval, P axis, QRS axis, T axis, and ACCI are extracted;Feature Selection—The Kruskal–Wallis technique evaluates feature importance and selects discriminative ones;and Model Design—An ANFIS model based on FCM clustering (FCM-ANFIS) is developed using the selected features. Results: Among the 42 patients selected during cohort filtering, T axis demonstrated significant association with potassium levels (P © The Author(s), under exclusive licence to The Brazilian Society of Biomedical engineering 2024.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Theoretical Analysis of Detection of Flying Vehicles Based on the Passive Radiometric Detection of Microwave-Millimeter-Terahertz Wavelength Electromagnetic Emissions from Exhaust Plasma Gases

引用

Progress in Electromagnetics Research C 2025年 154卷 267-275页

作者： Chatziathanasiou, Thomas N. Uzunoglu, Nikolaos Microwave Laboratory Department of Electrical and Computer Engineering National Technical University of Athens Greece

The feasibility of using passive radiometric detection of chaotic electromagnetic signals emanating from low density plasma plumes of the jet exhaust gases to detect low radar cross section aircrafts is analyzed for the frequency band 3.108–3.1012 Hz. The aircraft exhaust plume gas formation is modelled with a number of discrete different dimensions ellipsoids, with each having different electron densities and temperatures. Electromagnetic radiation analysis of emitting signals is computed applying fluctuation-dissipation theorem and geometric considerations. The detection range of characteristic military jet aircraft is computed numerically for the whole frequency band, from UHF to 3 THz. It is shown that high range detection can be achieved at upper microwave frequencies. © 2025, Electromagnetics Academy. All rights reserved.

关键词： Terahertz waves

来源：评论

学校读者我要写书评

暂无评论

A Scalable Unsupervised and Back Propagation Free Learning With SACSOM: A Novel Approach to SOM-Based Architectures

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第4期6卷 955-967页

作者： Hirani, Gaurav R. Wang, Kevin I-Kai Abdulla, Waleed H. The University of Auckland Department of Electrical Computer and Software Engineering Auckland1010 New Zealand

The field of computer vision is predominantly driven by supervised models, which, despite their efficacy, are computationally expensive and often intractable for many applications. Recently, research has expedited alternative avenues such as self-organizing maps (SOM)-based architectures, which offer significant advantages such as tractability, the absence of back-propagation, and feed-forward unsupervised learning. However, these SOM-based approaches frequently suffer from lower accuracy and limited generalization capabilities. To address these shortcomings, we propose a novel model called split and concur SOM (SACSOM). SACSOM overcomes the limitations of closely related SOM-based algorithms by utilizing multiple parallel branches, each equipped with its own SOM modules that process data independently with varying patch sizes. Furthermore, by creating groups of classes and using respective training samples to train independent subbranches in each branch, our approach accommodates datasets with a large number of classes. SACSOM employs a simple yet effective labeling technique requiring minimal labeled samples. The outputs from each branch, filtered by a threshold, contribute to the final prediction. Experimental validation on MNIST-digit, Fashion-MNIST, CIFAR-10, and CIFAR-100 demonstrates that SACSOM achieves competitive accuracy with significantly reduced computation time. Furthermore, it exhibits superior performance and generalization capabilities, even in high-noise scenarios. The weights of the single-layered SACSOM provide meaningful insights into the patch-based learning pattern, enhancing its tractability and making it ideal from the perspective of explainable AI. This study addresses the limitations of current clustering techniques, such as K-means and traditional SOMs, by proposing a lightweight, manageable, and fast architecture that does not require a GPU, making it suitable for low-powered devices. © 2024 IEEE.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Model emotion recognition from speech using StarGAN, DCNN, and SVM

引用

Multimedia Tools and Applications 2025年 1-18页

作者： Banihosseini, Najmeh Sadat Ghods, Vahid Department of Electrical and Computer Engineering Semnan Branch Islamic Azad University Semnan Iran

Emotion recognition from speech is a significant research area in human–computer interaction and psychological assessments. This study proposes a novel three-stage process for emotion recognition from speech signals. In the first stage, pre-processing operations, including noise reduction and normalization, are applied to improve data quality. Then, StarGAN is employed for data augmentation, addressing the challenge of limited emotional speech data. The second stage involves extracting meaningful features from high-dimensional data using deep convolutional neural networks (DCNN). Finally, support vector machines (SVM) are used to classify the extracted features into distinct emotional categories. The main advantage of the proposed method over traditional deep learning techniques is the use of StarGAN for data augmentation, which improves performance on limited datasets. Additionally, the combination of DCNN and SVM enhances accuracy and efficiency in feature extraction and classification. This method has been evaluated using four different datasets, RAVDESS, Emo-DB, SAVEE, and CASIA, achieving accuracy rates of 98.25%, 95.5%, 96.8%, and 96.45%, respectively. The results indicate that the proposed method significantly outperforms existing methods and offers improved capabilities for practical application. Despite achieving high accuracy, challenges such as increased computational complexity and variability in performance across different datasets are discussed. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Finite-Sample Bounds for Adaptive Inverse Reinforcement Learning using Passive Langevin Dynamics

引用

IEEE Transactions on Information Theory 2025年第6期71卷 4637-4670页

作者： Snow, Luke Krishnamurthy, Vikram Cornell University Department of Electrical and Computer Engineering IthacaNY14853 United States

This paper provides a finite-sample analysis of a passive stochastic gradient Langevin dynamics (PSGLD) algorithm. This algorithm is designed to achieve adaptive inverse reinforcement learning (IRL). Adaptive IRL aims to estimate the cost function of a forward learner performing a stochastic gradient algorithm (e.g., policy gradient reinforcement learning) by observing their estimates in real-time. The PSGLD algorithm is considered passive because it incorporates noisy gradients provided by an external stochastic gradient algorithm (forward learner), of which it has no control. The PSGLD algorithm acts as a randomized sampler to achieve adaptive IRL by reconstructing the forward learner’s cost function nonparametrically from the stationary measure of a Langevin diffusion. This paper analyzes the non-asymptotic (finite-sample) performance;we provide explicit bounds on the 2-Wasserstein distance between PSGLD algorithm sample measure and the stationary measure encoding the cost function, and provide guarantees for a kernel density estimation scheme which reconstructs the cost function from empirical samples. Our analysis uses tools from the study of Markov diffusion operators. The derived bounds have both practical and theoretical significance. They provide finite-time guarantees for an adaptive IRL mechanism, and substantially generalize the analytical framework of a line of research in passive stochastic gradient algorithms. © 1963-2012 IEEE.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

Incorporation of plasmonic metal-metal and metal-oxide configurations in a polymer solar cell: introducing inorganic features to organic photovoltaics

引用

Optical and Quantum Electronics 2025年第4期57卷 1-16页

作者： Heidarzadeh, Hamid Department of Electrical and Computer Engineering University of Mohaghegh Ardabili Ardabil Iran

In thicker polymer active layers charge collection efficiency suffers due to low carrier mobility and increased recombination losses. In thin absorber polymer solar cell to increase absorption, light-trapping techniques and plasmonic structures are essential. This study investigates the effect of shell thickness on the photocurrent density of a poly(3-hexylthiophene): phenyl-C61- butyric acid methyl ester (P3HT:PCBM) polymer based solar cell incorporating core–shell nanoparticles with configurations of Au–Ag and Ag-Au core–shell nanoparticles. Through a series of simulation, the photocurrent density was calculated as a function of shell thickness. The results demonstrate that, for both nanoparticle configurations, the photocurrent density generally increases with shell thickness, reaching an optimal point before stabilizing or slightly decreasing. Additionally, the effects of dielectric shells made of SiO₂ and Al₂O₃ on its performance parameters were analyzed. The study also found that the photocurrent decreases with increasing shell thickness for both SiO₂ and Al₂O₃ shells, with a more pronounced decrease for SiO₂ due to its smaller refractive index and greater change in shorter wavelengths. The photocurrent density of 13.74 mA/cm2 is achieved for a cell with a thickness of 80 nm without nanoparticles. This value increases to 16.62 mA/cm2 for a cell incorporating Ag nanoparticles and reaches 19.3 mA/cm2 for a cell with Au–Ag core–shell nanoparticles at the optimal shell thickness. The power conversion efficiency of the polymer solar cell increases from 7.02% without nanoparticles to 8.67% with Ag, 8.45% with Au, and reaches the highest value of 10.26% with Au–Ag core–shell nanoparticles, highlighting the superior performance of the core–shell configuration. This superior performance is attributed to the enhanced plasmonic effects of the Au–Ag combination, which facilitates better light trapping and absorption. These findings underscore the importance of optimizing

关键词： Plasmonics

来源：评论

学校读者我要写书评

暂无评论

Nonlinear observer-based state-of-charge balancing of networked battery energy storage systems

引用

Journal of Control and Decision 2025年第1期12卷 49-64页

作者： Tingyang Meng Zongli Lin Yan Wan Yacov A.Shamash Charles L.Brown Department of Electrical and Computer Engineering University of VirginiaCharlottesvilleVAUSA Department of Electrical Engineering University of Texas at ArlingtonFort WorthTXUSA Department of Electrical and Computer Engineering Stony Brook UniversityStony BrookNYUSA

In this paper,we propose an observer-based algorithm for balancing the state-of-charge(SoC)among battery units in a battery energy storage system(BESS).The dynamical behaviour of a battery unit is approximated by an equivalent circuit model,based on which a nonlinear SoC observer can be *** distribution laws are designed for the battery units according to the states of the battery units,the average battery state,and the average power *** estimation algorithms for the average battery state and the average power demand,as well as SoC observers,are constructed to implement *** BESS is shown to achieve SoC balancing among all its battery units while satisfying the power demand,as long as mild conditions on the underlying communication network and on the power demand are *** results are presented to demonstrate the effectiveness of the proposed algorithm.

关键词： Microgrids networked battery energy storage systems power distribution state-of-charge balancing equivalent circuit model nonlinear observer

来源：评论

学校读者我要写书评

暂无评论

Enabling ILP-Based DSE for Multigranularity, Unified Domain Platforms With DmTSAR-ILP

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2025年第3期44卷 911-922页

作者： Morais, Bruno Jiang, Qucheng Schirner, Gunar Northeastern University Department of Electrical and Computer Engineering BostonMA02115 United States

Domain-specific HWACC-rich platforms blend high performance and efficiency presenting an opportunity to recover nonrecurring engineering costs through wider deployment for many applications. However, the design of such platforms is immensely challenging, partly due to the design space size. MG-DmDSE (Zhang et al., 2023) allocates domain platforms catering to multiple applications. To cope with complexity, it employs a genetic algorithm and heuristics. While this is a typical approach in design space exploration (DSE), it cannot guarantee optimality. Exact solutions could be obtained with integer linear programming (ILP). Yet, the multiapplication DSE presents aggregation challenges as applications might have widely different performance, leading to nonlinear aggregation techniques in previous work that ILPs cannot employ. This work introduces DmTSAR-ILP, an ILP-based platform allocation method that simultaneously considers all applications in a domain. DmTSAR-ILP captures the same underlying analytical model as MG-DmDSE and can explore its domain application set. To enable our linear domain-level formulation, we introduce the average performance achievement metric (APA), offering an ILP-friendly, fair aggregation across domain applications. To highlight benefits of DmTSAR-ILP, a domain platform for 40 OpenVX apps is generated (across varying area budgets). For all area budgets, DmTSAR-ILP platforms perform better or identical to MG-DmDSE even in their original metric. The proposed APA aggregation is fair, allowing more applications (55% vs 37.5% of applications in MG-DmDSE) to be fully accelerated on the generated platform. For the 0.1 mm2 area budget, DmTSAR-ILP's platform increases application throughput by 22.5% over MG-DmDSE's platform. DmTSAR-ILP is 70x faster on average than the heuristic-based MG-DmDSE by leveraging an ILP formulation that considers all applications simultaneously and recent advances in mixed-integer programming solver performance. © 1982-201

关键词： Budget control

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：