检索结果-内蒙古大学图书馆

BDAN: Mitigating Temporal Difference Across Electrodes in Cross-Subject Motor Imagery Classification via Generative Bridging Domain

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Zhige Yang, Rui Huang, Mengjie Qin, Chengxuan Wang, Zidong School of Advanced Technology Xi'an Jiaotong-Liverpool University Suzhou215123 China School of Electrical Engineering Electronics and Computer Science University of Liverpool LiverpoolL69 3BX United Kingdom Design School Xi'an Jiaotong-Liverpool University Suzhou215123 China Department of Computer Science Brunel University London Uxbridge MiddlesexUB8 3PH United Kingdom

Because of "the non-repeatability of the experiment settings and conditions" and "the variability of brain patterns among subjects", the data distributions across sessions and electrodes are different in cross-subject motor imagery (MI) studies, eventually reducing the performance of the classification model. Systematically summarised based on the existing studies, a novel temporal-electrode data distribution problem is investigated under both intra-subject and inter-subject scenarios in this paper. Based on the presented issue, a novel bridging domain adaptation network (BDAN) is proposed, aiming to minimise the data distribution difference across sessions in the aspect of the electrode, thus improving and enhancing model performance. In the proposed BDAN, deep features of all the EEG data are extracted via a specially designed spatial feature extractor. With the obtained spatio-temporal features, a special generative bridging domain is established, bridging the data from all the subjects across sessions. The difference across sessions and electrodes is then minimized using the customized bridging loss functions, and the known knowledge is automatically transferred through the constructed bridging domain. To show the effectiveness of the proposed BDAN, comparison experiments and ablation studies are conducted on a public EEG dataset. The overall comparison results demonstrate the superior performance of the proposed BDAN compared with the other advanced deep learning and domain adaptation methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Electrodes

Bridging Gaps in Paper Design Considering Impacts of Switching Speed and Power-loop Layout 12

学校读者我要写书评

暂无评论

Bridging Gaps in Paper Design Considering Impacts of Switchi...

12th Annual IEEE Energy Conversion Congress and Exposition, ECCE 2020

作者： Ren, Ren Dong, Zhou Wang, Fei Fred University of Tennessee at Knoxville Department of Electrical Engineering and Computer Science KnoxvilleTN United States University of Tennessee at Knoxville Department of Electrical Engineering and Computer Science Oak Ridge National Laboratory KnoxvilleTN United States

ISBN: (数字)9781728158266

ISBN: (纸本)9781728158266

In power electronics design, the selection of power semiconductors mainly considers the voltage and current ratings, power loss, and operating frequency capability. However, in the paper design stage, the impacts of switching speed and busbar or PCB layouts related geometries design on the converter system are not considered comprehensively, which may introduce a trial-and-error procedure of selecting the gate resistance in the implementations due to overvoltage and cooling capability issues. To shrink the design gaps between the paper design and implementations, with advanced modeling methods and design automation programs, this paper proposes a systematic design approach for power semiconductors selection and cooling system design considering the impacts of switching speed and power-loop layout. The discrete-time switching behavior model for the switching speed prediction and automatic geometry creation program for the power-loop layout are developed to determine the gate resistance in the paper design stage, and the partial element equivalent circuits (PEEC) numerical modeling method is adopted to extract the parasitic inductance for the created geometries. After the determination of gate resistances for the selected device and topology, a multi-variable switching loss scaling method with the discrete-time switching behavior model is used to get the more accurate device losses for the cooling design. A complete design iteration and related verifications are also given in this paper. © 2020 IEEE.

关键词： Geometry

Mitigating Misleading Chain-of-Thought Reasoning with Selective Filtering

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Yexin Zhang, Zhuosheng Zhao, Hai College of Zhiyuan Shanghai Jiao Tong University China School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University China Department of Computer Science and Engineering Shanghai Jiao Tong University China Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University China

Large language models have manifested remarkable capabilities by leveraging chain-of-thought (CoT) reasoning techniques to solve intricate questions through step-by-step reasoning chains. Despite its success, the efficacy of such reasoning is inherently contingent upon the quality of CoT. However, flawless CoT reasoning cannot be guaranteed due to the presence of indecomposable questions and the potential for erroneous reasoning chains, particularly in the case of small-scale language models. To tackle this challenge, we propose a novel approach called the selective filtering reasoner (SelF-Reasoner) that assesses the entailment relationship between the question and the candidate reasoning chain. Then, we proceed with CoT reasoning when the reasoning chain demonstrates confidence;otherwise, we opt to predict the answer directly. SelF-Reasoner improves the fine-tuned T5 baseline consistently over the scienceQA, ECQA, and LastLetter tasks. Code is available at https://***/LibroWu/SelF-Reasoner. © 2024, CC BY-SA.

关键词： Computational linguistics

Identification of Abnormal Operating Conditions in Fused Magnesium Furnaces Using a Deep Multiview TSK Fuzzy System

学校读者我要写书评

暂无评论

Identification of Abnormal Operating Conditions in Fused Mag...

Chinese Automation Congress (CAC)

作者： Chao Xi Gaochang Wu Zizhu Fan Feng Xu Hui Wang Xiaokun Tang School of Electrical and Automation Engineering East China Jiaotong University Nanchang China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang China College of Computer Science and Technology Shanghai University of Electric Power Shanghai China Key Laboratory of Advanced Control and Optimization of Jiangxi Province East China Jiaotong University Nanchang China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

The smelting process of the fused magnesium furnace(FMF) is prone to semi-molten abnormal conditions, which may lead to low product quality and production efficiency, and is also prone to safety accidents. At present, the operating conditions of FMF mainly rely on manual observation of the furnace shell features for diagnosis, and it is hard to timely and effectively identify abnormalities in the smelting process. Given the difficulty in identifying abnormal operating conditions of FMF, this paper proposes a deep multiview fuzzy classification method(DMFC) that can automatically identify abnormal operating conditions. First, based on the three-phase current data, CNN and Transformer are used to extract features as multiple views. Secondly, the Takagi-Sugeno-Kang(TSK) fuzzy model is used as a single-view model, and the evidence theory is used for dynamic fusion of multiple views. This method can effectively fuse multiple deep features, and has strong scalability and a certain degree of interpretability. Through verification of the real historical data of multiple FMFs, our model has relatively high accuracy, false detection rate(FDR), and missed detection rate(MDR).

关键词： Fuses Heuristic algorithms Furnaces Smelting Production Feature extraction Transformers Magnesium Classification algorithms Unsupervised learning

An I2I Inpainting Approach for Efficient Channel Knowledge Map Construction

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jin, Zhenzhou You, Li Wang, Jue Xia, Xiang-Gen Gao, Xiqi National Mobile Communications Research Laboratory Southeast University Nanjing210096 China Purple Mountain Laboratories Nanjing211100 China School of Information Science and Technology Nantong University Nantong226019 China Nantong Research Institute for Advanced Communication Technologies Nantong226019 China Department of Electrical and Computer Engineering University of Delaware NewarkDE19716 United States

Channel knowledge map (CKM) has received widespread attention as an emerging enabling technology for environment-aware wireless communications. It involves the construction of databases containing location-specific channel knowledge, which are then leveraged to facilitate channel state information (CSI) acquisition and transceiver design. In this context, a fundamental challenge lies in efficiently constructing the CKM based on a given wireless propagation environment. Most existing methods are based on stochastic modeling and sequence prediction, which do not fully exploit the inherent physical characteristics of the propagation environment, resulting in low accuracy and high computational complexity. To address these limitations, we propose a Laplacian pyramid (LP)-based CKM construction scheme to predict the channel knowledge at arbitrary locations in a targeted area. Specifically, we first view the channel knowledge as a 2-D image and transform the CKM construction problem into an image-to-image (I2I) inpainting task, which predicts the channel knowledge at a specific location by recovering the corresponding pixel value in the image matrix. Then, inspired by the reversible and closed-form structure of the LP, we show its natural suitability for our task in designing a fast I2I mapping network. For different frequency components of LP decomposition, we design tailored networks accordingly. Besides, to encode the global structural information of the propagation environment, we introduce self-attention and cross-covariance attention mechanisms in different layers, respectively. Finally, experimental results demonstrate that the proposed scheme outperforms the benchmark, achieving higher reconstruction accuracy while with lower computational complexity. Moreover, the proposed approach has a strong generalization ability and can be implemented in different wireless communication scenarios. © 2024, CC BY-NC-ND.

关键词： Location

A New IoT-Based Smart Greenhouse for Cultivation of Water Convolvulus

学校读者我要写书评

暂无评论

A New IoT-Based Smart Greenhouse for Cultivation of Water Co...

International electrical engineering Congress (iEECON)

作者： Prayoot Akkaraekthalin Tanee Somwong Nonchanutt Chudpooti Nuttawut Hirun Suwaluck. Meesomklin Department of Electrical and Computer Engineering Faculty of Engineering King Mongkut’s University of Technology North Bangkok Bangkok Thailand Department of Electronic and Telecommunications Engineering Faculty of Engineering and Architecture Rajamangala University of Technology Suvarnabhumi Nonthaburi Thailand Department of Industrial Physics and Medical Instrumentation Faculty of Applied Science King Mongkut’s University of Technology North Bangkok Bangkok Thailand Department of Electronics Engineering Technology College of Industrial Technology King Mongkut’s University of Technology North Bangkok Bangkok Thailand

ISBN: (数字)9798331543952

ISBN: (纸本)9798331543969

This paper presents a new IOT system for monitoring and controlling plant cultivation in greenhouse system. The system aims to regulate and monitor the environmental conditions suitable for plant cultivation in real-time. The system utilizes sensors to measure environmental conditions, such as soil moisture sensor, light intensity sensor, temperature sensor, and air humidity sensor. The Node MCU ESP32 is used for processing and controlling these operations. The system is designed to send notifications through Line application and allows farmers to monitor the system's performance and plant growth in real-time via a smartphone application. It can utilize the Blynk IoT application on a smartphone for remotely monitoring and controlling the environmental conditions and provides historical data. This experiment demonstrates that the designed system can successfully cultivate plants and promote better growth compared to traditional cultivation methods with a 13.38% improvement. Additionally, it provides convenience and ease of management for plant cultivation.

关键词： Temperature sensors Temperature measurement Smart agriculture System performance Green products Sensor systems Real-time systems Sensors Internet of Things Monitoring

Embedded Aperture-Based FSS Sensor

学校读者我要写书评

暂无评论

Embedded Aperture-Based FSS Sensor

Antennas and Propagation Society International Symposium

作者： Lauryn Morris Swathi Muthyala Ramesh Logan M. Wilcox Doyle T. Motes Kristen M. Donnell Department of Electrical and Computer Engineering Microwave Sensing (μSense) Laboratory Missouri University of Science and Technology Rolla Missouri USA Nondestructive Evaluation Division Texas Research Institute (TRI) Austin Inc. Austin Texas USA

Frequency selective surfaces (FSSs) are two-dimensional periodic arrays of electrically conductive elements or apertures that have a specific electromagnetic reflection and/or transmission response. In this work, an aperture-based sensor design was proposed that is capable of sensing strain when embedded in a layered dielectric structure (i.e., an embedded sensor). An aperture approach was selected due to the potential for operation in reflection mode without a conductive backplane. The sensor was designed to operate within the K-band to improve the potential strain measurement resolution, as the unit cell dimensions are inversely related to the operating frequency and directly related to strain sensitivity. Simulation results for the small-scale strain range of 0-0.5% are presented, with a sensitivity of 20.4 MHz/O.1 % strain.

关键词：

Pseudo-label Guided Cross-video Pixel Contrast for Robotic Surgical Scene Segmentation with Limited Annotations

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yu, Yang Zhao, Zixu Jin, Yueming Chen, Guangyong Dou, Qi Heng, Pheng-Ann The Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China The Department of Computer Science University College London United Kingdom Zhejiang Lab China

Surgical scene segmentation is fundamentally crucial for prompting cognitive assistance in robotic surgery. However, pixel-wise annotating surgical video in a frame-by-frame manner is expensive and time consuming. To greatly reduce the labeling burden, in this work, we study semi-supervised scene segmentation from robotic surgical video, which is practically essential yet rarely explored before. We consider a clinically suitable annotation situation under the equidistant sampling. We then propose PGV-CL, a novel pseudo-label guided cross-video contrast learning method to boost scene segmentation. It effectively leverages unlabeled data for a trusty and global model regularization that produces more discriminative feature representation. Concretely, for trusty representation learning, we propose to incorporate pseudo labels to instruct the pair selection, obtaining more reliable representation pairs for pixel contrast. Moreover, we expand the representation learning space from previous image-level to cross-video, which can capture the global semantics to benefit the learning process. We extensively evaluate our method on a public robotic surgery dataset EndoVis18 and a public cataract dataset CaDIS. Experimental results demonstrate the effectiveness of our method, consistently outperforming the state-of-the-art semi-supervised methods under different labeling ratios, and even surpassing fully supervised training on EndoVis18 with 10.1% labeling. Our code is available at https://***/yangyu-cuhk/PGVCL. Copyright © 2022, The Authors. All rights reserved.

关键词： Robotic surgery

VCSE: Time-Domain Visual-Contextual Speaker Extraction Network

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Li, Junjie Ge, Meng Pan, Zexu Wang, Longbiao Dang, Jianwu Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China Department of Electrical and Computer Engineering National University of Singapore Singapore Japan Advanced Institute of Science and Technology Ishikawa Japan

Speaker extraction seeks to extract the target speech in a multi-talker scenario given an auxiliary reference. Such reference can be auditory, i.e., a pre-recorded speech, visual, i.e., lip movements, or contextual, i.e., phonetic sequence. References in different modalities provide distinct and complementary information that could be fused to form top-down attention on the target speaker. Previous studies have introduced visual and contextual modalities in a single model. In this paper, we propose a two-stage time-domain visual-contextual speaker extraction network named VCSE, which incorporates visual and self-enrolled contextual cues stage by stage to take full advantage of every modality. In the first stage, we pre-extract a target speech with visual cues and estimate the underlying phonetic sequence. In the second stage, we refine the pre-extracted target speech with the self-enrolled contextual cues. Experimental results on the real-world Lip Reading Sentences 3 (LRS3) database demonstrate that our proposed VCSE network consistently outperforms other state-of-the-art baselines. Copyright © 2022, The Authors. All rights reserved.

关键词： Extraction