检索结果-内蒙古大学图书馆

Uncertainty quantification in large language models through convex hull analysis

Discover Artificial Intelligence 2024年第1期4卷 1-14页

作者： Catak, Ferhat Ozgur Kuzlu, Murat Department of Electrical Engineering and Computer Science University of Stavanger Rogaland Norway Department of Engineering Technology Old Dominion University NorfolkVA United States

Uncertainty quantification approaches have been more critical in large language models (LLMs), particularly high-risk applications requiring reliable outputs. However, traditional methods for uncertainty quantification, such as probabilistic models and ensemble techniques, face challenges when applied to the complex and high-dimensional nature of LLM-generated outputs. This study proposes a novel geometric approach to uncertainty quantification using convex hull analysis. The proposed method leverages the spatial properties of response embeddings to measure the dispersion and variability of model outputs. The prompts are categorized into three types, i.e., ’easy’, ’moderate’, and ’confusing’, to generate multiple responses using different LLMs at varying temperature settings. The responses are transformed into high-dimensional embeddings via a BERT model and subsequently projected into a two-dimensional space using Principal Component Analysis (PCA), Isomap, Multidimensional Scaling (MDS). The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is utilized to cluster the embeddings and compute the convex hull for each selected cluster. The experimental results indicate that the uncertainty of the model for LLMs depends on the prompt complexity, the model, and the temperature setting. © The Author(s) 2024.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Robust Place Recognition Using Fusion of Thermal Infrared and RGB Cameras

引用

Journal of Institute of Control, Robotics and Systems 2024年第12期30卷 1414-1421页

作者： Ma, Seungjun Cho, Younggun Department of Electrical and Computer Engineering Inha University Korea Republic of

The performance of camera-based place recognition has significantly improved with the rapid advancement of deep learning. However, RGB cameras still face challenges in handling variations in lighting conditions due to their inherent limitations. Recent research has explored the use of thermal infrared (TIR) cameras to achieve illumination-invariant scene recognition. Nevertheless, TIR cameras can also struggle in scenarios with uniform heat distribution or when objects with extreme temperatures are present. We herein propose a robust place recognition methodology that leverages RGB-TIR feature fusion. We developed a novel network architecture that can dynamically extract information weights from each RGB and TIR image and then generate a weighted descriptor. This approach is particularly effective in scenarios in which one of the sensor modalities is degraded. The proposed method was validated using a dataset covering a subterranean environment, and it demonstrated superior performance compared with existing visual place recognition (VPR) methods. Additionally, we conducted an ablation study to evaluate the effectiveness of the proposed weight extraction network, providing qualitative insights into its performance. © ICROS 2024.

关键词： SLAM robotics

来源：评论

学校读者我要写书评

暂无评论

Improving the Performance of Universal Traffic Light Recognition Through Dataset Construction and Selection in Semi-supervised Learning

引用

Journal of Institute of Control, Robotics and Systems 2024年第8期30卷 787-792页

作者： Kim, Dayoung Li, Xingyou Kim, Hakil Department of Electrical and Computer Engineering Inha University Korea Republic of

Traffic light recognition in autonomous driving is an essential but very challenging task because its performance is affected by unpredictable environmental conditions. Moreover, the shapes and installations of traffic lights in various countries require a lot of cost and time to collect a huge dataset and train a deep learning model. This study develops a method for constructing training datasets with minimal resource expenditure for more efficient and universally applicable traffic light recognition. It aims to develop a deep learning scheme that can universally recognize various international traffic light configurations by replacing the initial supervised learning stage with a phase that can recognize multiple types of traffic lights and improve performance through self-supervised learning on new datasets. Additionally, this study introduces a data selection algorithm that enables robust recognition under changes in road conditions by enhancing adaptability across diverse environments. This research not only demonstrates improved recognition performance in hazardous road situations but also concludes with potential for efficient application worldwide. Experimental results showed 24.2% improvement in mAP50, even with the same proportion of labeled data, and similar performance with half the labeled data. © ICROS 2024.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

A 4-Port Broadband High-Isolated MIMO Antenna for Wireless Communication

引用

Progress In Electromagnetics Research C 2024年 142卷 119-130页

作者： Ali, Ayyaz Rasool, Maryam Zahid, Muhammad Zeeshan Rashid, Imran Siddique, Adil Masood Maqsood, Moazzam Bhatti, Farooq Ahmad Department of Electrical Engineering Military College of Signals National University of Sciences and Technology Islamabad Pakistan Department of Electrical and Computer Engineering Pak–Austria Fachhochschule: Institute of Applied Sciences and Technology KPK Haripur Pakistan

This article proposes a versatile Multiple Input Multiple Output (MIMO) antenna designed for contemporary wireless systems spanning frequencies from 3 to 20 GHz. It serves applications such as 5G mobile, WiFi, WiFi-6E, X-band, partial Ku, and K-band. The original single-element antenna evolves into a 4 × 4 MIMO configuration with optimized ground plane modifications for enhanced performance. A decoupling structure achieves over 20 dB isolation between inter-elements. The feeding structure, featuring a gradually changing design connected to the antenna’s radiating structure, achieves wide bandwidth characteristics. This is further improved by a partial ground structure and slots on the radiating element. The lower frequency band of 3 to 7 GHz is attained with a rectangle-shaped radiator, while semi-circular microstrip lines atop the radiator enable the higher frequency bands of 8 to 15.4 and 18.7 to 20 GHz. The slots and ground structure enhance impedance bandwidth, and semicircles improve the radiation pattern. The MIMO antenna demonstrates measured peak gains of 6.23 dBi at 11.5 GHz, maintaining a radiation efficiency exceeding 80%. Validation through metrics like ECC, DG, CCL, and TARC confirms strong agreement between simulated and experimental results, positioning the MIMO antenna as a robust choice for various wireless communication applications. © 2024, Electromagnetics Academy. All rights reserved.

关键词： Radiators

来源：评论

学校读者我要写书评

暂无评论

A new multi-input two-output switched-source multilevel inverter based on harvest of maximum energy for photovoltaic applications

引用

International Journal of Power Electronics 2024年第2期19卷 167-192页

作者： Salehi, Seyed Jafar Shamsi-Nejad, Mohammad Ali Najafi, Hamid Reza Department of Electrical and Computer Engineering University of Birjand Birjand Iran

In multilevel inverters, unused energies are created due to the asynchronous use of the input DC-sources. This means that when the input DC-sources are replaced by renewable systems such as photovoltaic arrays, some of the input energies remain unused by the multilevel inverter. This paper presents a new multi-input two output multilevel inverter topology that can harvest the unused energies and return them to another load which leads to the harvest of the maximum input energy. The harvest of maximum energy (HME)-based multilevel inverter structure consists of two independent terminals both of which can be connected to the AC load. Another merit of the proposed multilevel inverter is that the number of its switches is comparable to other structures where unused energies cannot be harvested. The nearest level control (NLC) has been used as the switching strategy in the proposed multilevel structure. In order to verify the performance of the HME-based multilevel inverter topology, the simulation and experimental results for a type seven-level inverter are provided by MATLAB and TMS320F28379D DSP, respectively. © 2024 Inderscience Enterprises Ltd.

关键词： Level control

来源：评论

学校读者我要写书评

暂无评论

Recent progress of artificial intelligence for liquid-vapor phase change heat transfer

引用

npj Computational Materials 2024年第1期10卷 2552-2565页

作者： Youngjoon Suh Aparna Chandramowlishwaran Yoonjin Won Department of Mechanical and Aerospace Engineering University of CaliforniaIrvineCAUSA Department of Electrical Engineering and Computer Science University of CaliforniaIrvineCAUSA

Artificial intelligence(AI)is shifting the paradigm of two-phase heat transfer *** innovations in AI and machine learning uniquely offer the potential for collecting new types of physically meaningful features that have not been addressed in the past,for making their insights available to other domains,and for solving for physical quantities based on first principles for phasechange thermofluidic *** review outlines core ideas of current AI technologies connected to thermal energy science to illustrate how they can be used to push the limit of our knowledge boundaries about boiling and condensation *** technologies for meta-analysis,data extraction,and data stream analysis are described with their potential challenges,opportunities,and alternative ***,we offer outlooks and perspectives regarding physics-centered machine learning,sustainable cyberinfrastructures,and multidisciplinary efforts that will help foster the growing trend of AI for phase-change heat and mass transfer.

关键词： transfer meaningful shifting

来源：评论

学校读者我要写书评

暂无评论

DSVPWM-Based SSO-PIDR Control Scheme for Indirect Matrix Converter with Unity Power Factor Under Unbalanced Grid Voltage Condition

引用

Journal of The Institution of Engineers (India): Series B 2024年第5期105卷 1241-1258页

作者： Das, Sarat Kumar Moharana, Jayakrushna Syam, Prasid Department of Electrical Engineering Indian Institute of Engineering Science and Technology Shibpur711103 India Department of Electrical Engineering Hi-Tech Institute of Technology Odisha752057 India

Indirect matrix converter (IMC) is a direct ac-ac conversion system that not requires electrolytic or bulky capacitors. This system has a four-quadrant operation with input power factor control and also generates sinusoidal waveforms for output voltage with variable frequency and amplitude. For ideal operation, the rectification stage must be able to produce + ive dc-link voltage and sinusoidal input current with unity Grid Power Factor (GPF) even under an unbalanced voltage condition. Therefore, Shark Smell Optimization (SSO)-based Proportional Integral Derivative Resonant (PIDR) controller is proposed to operate the IMC switches to attain unity GPF and to maintain a balanced load voltage. The incorporation of derivative control with the existing Proportional Integral Resonant controller reduces the overshoot of the system and thus improves the transient response. Furthermore, the SSO algorithm helps to optimally tune the gain parameters of the PIDR controller. Finally, the 3-phase balanced output voltages with the preferred frequency and magnitude is produced using Double Space Vector Pulse Width Modulation technique. The results from analysis and experimental results illustrate that the system stability and the performance are improved. © The Institution of Engineers (India) 2024.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

A new content-aware image resizing based on Rényi entropy and deep learning

引用

Neural Computing and Applications 2024年第15期36卷 8885-8899页

作者： Ayubi, Jila Chehel Amirani, Mehdi Valizadeh, Morteza Department of Electrical and Computer Engineering Urmia University Urmia Iran

One of the most popular techniques for changing the purpose of an image or resizing a digital image with content awareness is the seam-carving method. The performance of image resizing algorithms based on seam machining shows that these algorithms are highly dependent on the extraction of importance map techniques and the detection of salient objects. So far, various algorithms have been proposed to extract the importance map. In this paper, a new method based on Rényi entropy is proposed to extract the importance map. Also, a deep learning network has been used to detect salient objects. The simulator results showed that combining Rényi’s importance map with a deep network of salient object detection performed better than classical seam-carving and other extended seam-carving algorithms based on deep learning. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Bifurcation analysis and control study of improved full-speed differential model in connected vehicle environment

引用

Chinese Physics B 2024年第7期33卷 245-266页

作者：艾文欢雷正清李丹洋方栋梁刘大为 College of Computer Science and Engineering Northwest Normal UniversityLanzhou 730070China College of Electrical Engineering Lanzhou Institute of TechnologyLanzhou 730050China

In recent years, the traffic congestion problem has become more and more serious, and the research on traffic system control has become a new hot spot. Studying the bifurcation characteristics of traffic flow systems and designing control schemes for unstable pivots can alleviate the traffic congestion problem from a new perspective. In this work, the full-speed differential model considering the vehicle network environment is improved in order to adjust the traffic flow from the perspective of bifurcation control, the existence conditions of Hopf bifurcation and saddle-node bifurcation in the model are proved theoretically, and the stability mutation point for the stability of the transportation system is found. For the unstable bifurcation point, a nonlinear system feedback controller is designed by using Chebyshev polynomial approximation and stochastic feedback control method. The advancement, postponement, and elimination of Hopf bifurcation are achieved without changing the system equilibrium point, and the mutation behavior of the transportation system is controlled so as to alleviate the traffic congestion. The changes in the stability of complex traffic systems are explained through the bifurcation analysis, which can better capture the characteristics of the traffic flow. By adjusting the control parameters in the feedback controllers, the influence of the boundary conditions on the stability of the traffic system is adequately described, and the effects of the unstable focuses and saddle points on the system are suppressed to slow down the traffic flow. In addition, the unstable bifurcation points can be eliminated and the Hopf bifurcation can be controlled to advance, delay, and disappear,so as to realize the control of the stability behavior of the traffic system, which can help to alleviate the traffic congestion and describe the actual traffic phenomena as well.

关键词： bifurcation analysis vehicle queuing bifurcation control Hopf bifurcation

来源：评论

学校读者我要写书评

暂无评论

Active Voltage Balancing of Series-Connected SiC MOSFETs During Body-Diode Turn-Off Using an Active Gate Driver

引用

IEEE Transactions on Power Electronics 2025年第8期40卷 10609-10622页

作者： Rai, Ajay Ganesan, P. Hatua, Kamalesh Bhattacharya, Subhashish Indian Institute of Technology Madras Department of Electrical Engineering Chennai600036 India Thiruvananthapuram695033 India NC State University Department of Electrical and Computer Engineering RaleighNC27606 United States

The series connection of fast-switching SiC mosfets presents a promising approach to enhance the efficiency, compactness, and simplicity of medium voltage (MV) power converters. However, managing voltage imbalances during the turn-offof mosfets and their body diodes introduces significant challenges. While existing literature covers passive and active voltage balancing strategies for switch turn-off (e.g., mosfets/IGBTs), the issue of uneven voltage distribution during body diode turn-off remains largely unexplored. This gap limits the widespread adoption of series-connected devices in MV power converters, as achieving reliable performance with minimal additional components and low switching losses remains difficult. This article presents an active voltage balancing technique for handling body diode turn-off events by regulating the gate charge of parallel mosfets during voltage rise transients in the closed-loop using active gate drivers (AGD). This approach ensures uniform voltage distribution among the body diodes of series-connected mosfets. A comprehensive analysis and modeling of gate charge control during voltage transients is presented. The proposed method is experimentally validated on two series-connected SiC mosfets operating in buck mode (dc-dc) and half-bridge inverter mode (dc-ac) at a 1.5 kV dc bus voltage, with switching frequencies up to 40 kHz. © 1986-2012 IEEE.

关键词： Diodes

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：