检索结果-内蒙古大学图书馆

Frontiers in Optics + Laser science 2023, FiO, LS 203: Part of Frontiers in Optics + Laser science 2023

作者： Zhu, Chen Huang, Jie Research Center for Optical Fiber Sensing Zhejiang Laboratory Hangzhou311100 China Department of Electrical and Computer Engineering Missouri University of Science and Technology RollaMO65409 United States

An optical fiber curvature sensor based on a no-core fiber (NCF) cascaded with a hollow-core fiber (HCF), realizing simultaneously high sensitivity and a broad dynamic range with the assistance of machine learning ana... 详细信息

ISBN: (纸本)9781957171296

关键词： Curvature sensors Fiber Bragg gratings Fiber optic sensors Long period fiber gratings Machine learning Optical fibers

来源：评论

学校读者我要写书评

暂无评论

Generative Neural Net for Spatial Concept-To-Image

Generative Neural Net for Spatial Concept-To-Image

引用

2023 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2023

作者： Young, Brendan Anderson, Derek T. Keller, James M. Petry, Fred Michael, Chris J. University of Missouri Electrical Engineering and Computer Science Department ColumbiaMO United States Stennis Space Center MS U.S. Naval Research Laboratory United States

ISBN: (纸本)9798350359527

Numerous artificial intelligence (AI) approaches, such as generative AI (GAI), large language models (LLM) and text-To-image networks, necessitate spatial intelligence for effective operation. Yet, a prevailing ideology embedded in contemporary solutions is their inclination to strictly adhere to a data-driven approach when addressing spatial learning. This viewpoint leads to opaque solutions, aka "black boxes", and it presupposes that the intricacies of spatial reasoning will effortlessly surface as an inherent byproduct of extensive data exposure. This simplistic reliance on data overlooks the wealth of established psychology and mathematics governing spatial relations and their inherent uncertainties. In this article, we delve into neural networks and learning strategies for incorporating well-established spatial relations mathematics into an effective and explainable GAI. Specifically, we showcase how to translate abstract concepts, expressed as sets of spatial relations, into visual imagery. This resulting imagery can be immediately used or passed on as a spatial prior to a larger and more sophisticated GAI. We present two synthetic use cases wherein spatial concepts materialize as intricate arrangements of multi-part colored geometric primitives, undergoing a spectrum of diverse planar affine transformations. Our exploration is categorized into two dimensions: discerning spatial placements (aka the "where") and shape depiction (aka the "what"). Overall, we've noted a strength in our proposed neural networks in discerning spatial placements. However, while our nets also exhibit a competence in determining "what"to depict, we recognize the necessity for additional fine-Tuning to achieving visual imagery consistent with state-of-The-Art works like DALL-E and stable diffusion. © 2023 IEEE.

关键词： Abstracting

来源：评论

学校读者我要写书评

暂无评论

Averaging-Based Energy Management Strategy for Hybrid Battery-Ultracapacitor Energy Storage System

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Rezaei, H. Abdollahi, S.E. Abdollahi, Seyedreza Filizadeh, S. Babol Noshirvani University of Technology Energy Conversion Research Laboratory Department of Power Engineering Faculty of Electrical and Computer Engineering Babol*** Iran University of Science and Technology of Mazandaran Electrical and Computer Engineering Department Behshahr48518-78195 Iran University of Manitoba Department of Electrical and Computer Engineering WinnipegMBR3T 5V6 Canada

Considering a battery as the energy source for steady average current and an ultracapacitor as the dynamic energy buffer, this paper proposes a novel control strategy based on average current demand. The methodology averages the demand current over a time window to determine battery current, combined with rules to maintain the ultracapacitor's state of charge above 50%. However, the battery current is significantly affected by the averaging time window. A small window causes fluctuations, while a large one imposes unwanted currents to control the ultracapacitor's state of charge. As so, by optimizing the averaging window, this strategy minimizes battery current stress and power loss. Simulations under an urban driving cycle compare the proposed strategy with a sole-battery configuration, dynamic programming (benchmark), and rule-based approaches (original strategy). The simulation results indicate that the proposed strategy smooths battery currents, achieves over 75% of dynamic programming's effectiveness in extending battery life, and reduces computational burden and memory usage by over 85%. The strategy's low computational burden, and causal nature enable real-time control, which is demonstrated practically on an experimental setup. © 2025 IEEE.

关键词： State of charge

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Reflective Learning Through Self-Distillation and Online Clustering for Speaker Representation Learning

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 1535-1550页

作者： Danwei Cai Zexin Cai Ze Li Ming Li Department of Electrical and Computer Engineering Duke University Durham NC USA Suzhou Municipal Key Laboratory of Multimodal Intelligent Systems Digital Innovation Research Center Duke Kunshan University Kunshan China School of Computer Science Wuhan University Wuhan China

Speaker representation learning is crucial for voice recognition systems, with recent advances in self-supervised approaches reducing dependency on labeled data. Current two-stage iterative frameworks, while effective, suffer from significant computational overhead due to repeated rounds of clustering and training. They also struggle with noisy pseudo labels that can impair model learning. This paper introduces self-supervised reflective learning (SSRL), an improved framework that addresses these limitations by enabling continuous refinement of pseudo labels during training. Through a teacher-student architecture and online clustering mechanism, SSRL eliminates the need for iterative training rounds. To handle label noise, we incorporate noisy label modeling and pseudo label queues that maintain temporal consistency. Experiments on VoxCeleb show SSRL's superiority over current two-stage iterative approaches, surpassing the performance of a 5-round method in just a single training round. Ablation studies validate the contributions of key components like noisy label modeling and pseudo label queues. Moreover, consistent improvements in pseudo labeling and the convergence of cluster counts demonstrate SSRL's effectiveness in deciphering unlabeled data. This work marks an important advancement in efficient and accurate self-supervised speaker representation learning through the novel reflective learning paradigm.

关键词： Training Representation learning Iterative methods Noise Adaptation models Noise measurement Predictive models Speech processing Robustness Convergence

来源：评论

学校读者我要写书评

暂无评论

In situ construction of PtSe_(2)/Ge Schottky junction array with interface passivation for broadband infrared photodetection and imaging

引用

InfoMat 2024年第4期6卷 37-46页

作者： Xue Li Shuo-En Wu Di Wu Tianxiang Zhao Pei Lin Zhifeng Shi Yongtao Tian Xinjian Li Longhui Zeng Xuechao Yu School of Physics and Microelectronics Key Laboratory of Material Physics Ministry of EducationZhengzhou UniversityZhengzhouHenanthe People's Republic of China Department of Electrical and Computer Engineering University of CaliforniaLa JollaCaliforniaUSA National Research Center for Optical Sensors/Communications Integrated Networks School of Electronic Science and EngineeringSoutheast UniversityNanjingJiangsuthe People's Republic of China Key Laboratory of Multifunctional Nanomaterials and Smart Systems Suzhou Institute of Nano-Tech and Nano-BionicsChinese Academy of SciencesSuzhouJiangsuthe People's Republic of China

Infrared(IR)detection is vital for various military and civilian *** research has highlighted the potential of two-dimensional(2D)topological semimetals in IR detection due to their distinctive advantages,including van der Waals(vdW)stacking,gapless electronic structure,and Van Hove singularities in the electronic density of ***,challenges such as large-scale patterning,poor photoresponsivity,and high dark current of photodetectors based on 2D topological semimetals significantly impede their wider applications in low-energy photon ***,we demonstrate the in situ fabrication of PtSe_(2)/Ge Schottky junction by directly depositing 2D PtSe_(2) films with a vertical layer structure on a Ge substrate with an ultrathin AlOx *** to high quality junction,the photodetector features a broadband response of up to 4.6μm,along with a high specific detectivity of�1012 Jones,and operates with remarkable stability in ambient conditions as ***,the highly integrated device arrays based on PtSe_(2)/AlOx/Ge Schottky junction showcases excellent Mid-IR(MIR)imaging capability at room *** findings highlight the promising prospects of 2D topological semimetals for uncooled IR photodetection and imaging applications.

关键词： broadband photodetection imaging platinum diselenide Schottky junction van der Waals integration

来源：评论

学校读者我要写书评

暂无评论

TOWARDS NON-ASYMPTOTIC CONVERGENCE FOR DIFFUSION-BASED GENERATIVE MODELS 12

TOWARDS NON-ASYMPTOTIC CONVERGENCE FOR DIFFUSION-BASED GENER...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Li, Gen Wei, Yuting Chen, Yuxin Chi, Yuejie Department of Statistics The Chinese University of Hong Kong Hong Kong Department of Statistics and Data Science Wharton School University of Pennsylvania PhiladelphiaPA19104 United States Department of Electrical and Systems Engineering University of Pennsylvania PhiladelphiaPA19104 United States Department of Electrical and Computer Engineering Carnegie Mellon University PittsburghPA15213 United States

Diffusion models, which convert noise into new data instances by learning to reverse a Markov diffusion process, have become a cornerstone in contemporary generative modeling. While their practical power has now been widely recognized, the theoretical underpinnings remain far from mature. In this work, we develop a suite of non-asymptotic theory towards understanding the data generation process of diffusion models in discrete time, assuming access to `2-accurate estimates of the (Stein) score functions. For a popular deterministic sampler (based on the probability flow ODE), we establish a convergence rate proportional to 1/T (with T the total number of steps), improving upon past results;for another mainstream stochastic sampler (i.e., a type of the denoising diffusion probabilistic model), we derive a convergence rate proportional to 1/√T, matching the state-of-the-art theory. Imposing only minimal assumptions on the target data distribution (e.g., no smoothness assumption is imposed), our results characterize how 2 score estimation errors affect the quality of the data generation process. In contrast to prior works, our theory is developed based on an elementary yet versatile non-asymptotic approach without resorting to toolboxes for SDEs and ODEs. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Diffusion

来源：评论

学校读者我要写书评

暂无评论

Rapid Monitoring and Defense Approach for Resilience Improvement of Grid Cyber Security

Rapid Monitoring and Defense Approach for Resilience Improve...

引用

2023 IEEE Industry Applications Society Annual Meeting, IAS 2023

作者： Qiu, Wei Dong, Yuqing Yin, He He, Minjun She, Buxin Liu, Yilu University of Bologna Bologna40136 Italy University of Tennessee Department of Electrical Engineering & Computer Science KnoxvilleTN United States Oak Ridge National Laboratory TN United States

ISBN: (纸本)9798350320169

False data injection attacks based on synchrophasor measurement data pose a serious threat to the safe and stable operation of power systems. To mitigate this issue, a rapid monitoring and defense approach is proposed to defend against cyber attacks. First, the Time and Frequency based Convolutional neural Network (TFCN) is proposed to detect different types of attacks. In TFCN, the spectrum layer extracts the frequency information of the input data, and then the time-frequency block can be formed. Next, a comprehensive defense strategy is developed for multiple cyber attacks to ensure the stability and resilience of the power system. To verify the effectiveness of the proposed approach, the high-speed frequency measurements collected from the wide-area monitoring system are used. The results demonstrate that the cyber attack detection performance can reach 95.57 % compared with traditional neural networks. The defense strategy is conducted and verified in a modified IEEE 39 bus system as well, which shows better performance in faster stability restoration. © 2023 IEEE.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Fabry-Perot Interferometer in a Coreless Fiber Fabricated by aFsLaser for High-Temperature Applications

Fabry-Perot Interferometer in a Coreless Fiber Fabricated by...

引用

2023 Conference on Lasers and Electro-Optics, CLEO 2023

An optical fiber in-line Fabry-Perot interferometer (FPI) fabricated in a no-core fiber using the direct femtosecond laser writing technique for high-temperature sensing applications is reported. © Optica Publish... 详细信息

ISBN: (纸本)9781957171258

关键词： Fabry-Perot interferometers

来源：评论

学校读者我要写书评

暂无评论

Fiber Ring Resonator Based on Microwave Photonics and Sensitivity-Enhanced Sensing

Fiber Ring Resonator Based on Microwave Photonics and Sensit...

引用

Optica Sensing Congress 2023, AIS, FTS, HISE, Sensors, ES 2023 - Part of Optical Sensors and Sensing Congress 2023

A new concept of microwave-photonic (MWP) fiber ring resonator is introduced and a method for sensitivity-enhanced sensing based on the system is proposed and experimentally demonstrated. © 2023 The Author (s).

ISBN: (纸本)9781957171241

关键词： Microwave photonics

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Human-Robot Interaction Without Velocity Measurement Using Off-Policy Reinforcement Learning

引用

IEEE/CAA Journal of Automatica Sinica 2022年第1期9卷 47-63页

作者： Yongliang Yang Zihao Ding Rui Wang Hamidreza Modares Donald C.Wunsch Key Laboratory of Knowledge Automation for Industrial Processes Ministry of Education School of Automation&Electrical EngineeringUniversity of Science and Technology BeijingBeijing 100083China School of Automation Beijng Institute TechnologyBeijing 100081China State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China Mechanical Engineering Department Michigan State UniversityEast LansingMI 48824 USA Department of Electrical&Computer Engineering Missouri University of Science&TechnologyRollaMO 65401 USA

In this paper,we present a novel data-driven design method for the human-robot interaction(HRI)system,where a given task is achieved by cooperation between the human and the *** presented HRI controller design is a two-level control design approach consisting of a task-oriented performance optimization design and a plant-oriented impedance controller *** task-oriented design minimizes the human effort and guarantees the perfect task tracking in the outer-loop,while the plant-oriented achieves the desired impedance from the human to the robot manipulator end-effector in the ***-driven reinforcement learning techniques are used for performance optimization in the outer-loop to assign the optimal impedance *** the inner-loop,a velocity-free filter is designed to avoid the requirement of end-effector velocity *** this basis,an adaptive controller is designed to achieve the desired impedance of the robot manipulator in the task *** simulation and experiment of a robot manipulator are conducted to verify the efficacy of the presented HRI design framework.

关键词： Adaptive impedance control data-driven method human-robot interaction(HRI) reinforcement learning velocity-free

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：