检索结果-内蒙古大学图书馆

Semi-Path: An interactive semi-supervised learning framework for gigapixel pathology image analysis

学校读者我要写书评

暂无评论

Smart Health 2024年 32卷

作者： Lai, Zhengfeng Chauhan, Joohi Chen, Dongjie Dugger, Brittany N. Cheung, Sen-Ching Chuah, Chen-Nee Department of Electrical and Computer Engineering University of California Davis Davis 95616 CA United States Department of Pathology and Laboratory Medicine University of California Davis Sacramento 95817 CA United States Department of Electrical and Computer Engineering University of Kentucky Lexington 40506 KY United States

The efficacy of supervised deep learning in medical image analyses, particularly in pathology, is hindered by the necessity for extensive manual annotations. Annotating images at the gigapixel level manually proves to be a highly labor-intensive and time-consuming task. Semi-supervised learning (SSL) has emerged as a promising approach that leverages unlabeled data to reduce labeling efforts. In this work, we introduce Semi-Path, a practical SSL framework enhanced with active learning (AL) for gigapixel pathology tasks. Unlike existing methods that treat SSL and AL as independent components where AL incurs significant computational complexity to SSL, we propose a deep fusion of SSL and AL into a unified framework. Our framework introduces Informative Active Annotation (IAA) that employs a SSL-AL iterative structure to effectively extract knowledge from unlabeled pathology data. This structure significantly minimizes labeling efforts and computational complexity. Then, we propose Adaptive Pseudo-Labeling (APL) to address heterogeneity in class distribution, and prediction difficulty that are often observed in real-world pathology tasks. We evaluate Semi-Path on pathology image classification and segmentation tasks over three datasets that include WSIs from breast, colorectal, and brain tissues. The experimental results demonstrate the consistent superiority of Semi-Path over state-of-the-art methods. © 2024 The Authors

关键词： Active learning Pathology image analysis Semi-supervised learning

Addressing Neon Gas Field Ion Source Instability Through Online Beam Current Estimation

学校读者我要写书评

暂无评论

Microscopy and Microanalysis 2022年第S1期28卷 36-39页

作者： Sheila W. Seidel Luisa Watkins Minxu Peng Akshay Agarwal Christopher Yu Vivek K Goyal Boston University Department of Electrical and Computer Engineering Boston MA United States Charles Stark Draper Laboratory Cambridge MA United States Massachusetts Institute of Technology Department of Electrical and Computer Engineering Cambridge MA United States

Features Mining and Machine Learning for Home Appliance Identification by Processing Smart meter Data

学校读者我要写书评

暂无评论

Features Mining and Machine Learning for Home Appliance Iden...

Advanced Innovations in Smart Cities (ICAISC), International Conference on

作者： Rabab Al Talib Saeed Mian Qaisar Hala Fatayerji Asad Waqar Electrical and Computer Engineering Department Effat University Jeddah Saudi Arabia LINEACT Laboratory CESI Engineering School Villeurbanne France Department of Electrical Engineering Bahria School of Engineering & Applied Sciences Islamabad Campus Islamabad Pakistan

The energy sector is changing as a result of digitalization and IoT advancements. The Internet of Energy (IoE) is developing to link many smart grid components and shareholders effectively. The use of smart meters is becoming more popular in this context. The automatic identification of appliances is one of the most important applications of smart meter data. Enumerated billing and dynamic load management are possible outcomes. This process is complicated due to the usage of many brands and types of equipment. For the purpose of automatically identifying significant home appliances based on their usage patterns, this study presents a novel hybridization of segmentation, time-domain feature extraction, and machine learning algorithms. While automatically categorizing six key household appliances of various manufacturers, the developed technique achieves 96.2 percent accuracy, 97.7 percent specificity, and 98 percent AUC values.

关键词： Support vector machines Home appliances Technological innovation Machine learning algorithms Smart cities Feature extraction Load management

Tighter Lower Bounds on Quantum Annealing Times

学校读者我要写书评

暂无评论

arXiv 2024年

作者： García-Pintos, Luis Pedro Sahasrabudhe, Mrunmay Arenz, Christian Los Alamos National Laboratory New Mexico87545 United States School of Electrical Computer and Energy Engineering Arizona State University TempeAZ85287 United States

We derive lower bounds on the time needed for a quantum annealer to prepare the ground state of a target Hamiltonian. These bounds do not depend on the annealing schedule and can take the local structure of the Hamiltonian into account. Consequently, the bounds are computable without knowledge of the annealer’s dynamics and, in certain cases, scale with the size of the system. We discuss spin systems where the bounds are polynomially tighter than existing bounds, qualitatively capturing the scaling of the exact annealing times as a function of the number of spins. © 2024, CC BY-NC-ND.

关键词： Hamiltonians

REBEL: Rule-based and Experience-enhanced Learning with LLMs for Initial Task Allocation in Multi-Human Multi-Robot Teaming

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Gupte, Arjun Wang, Ruiqi Venkatesh, Vishnunandan L.N. Kim, Taehyeon Zhao, Dezhong Yuan, Ziqin Min, Byung-Cheol SMART Laboratory Department of Computer and Information Technology Purdue University West LafayetteIN United States College of Mechanical and Electrical Engineering Beijing University of Chemical Technology Beijing China

Multi-human multi-robot teams are increasingly recognized for their efficiency in executing large-scale, complex tasks by integrating heterogeneous yet potentially synergistic humans and robots. However, this inherent heterogeneity presents significant challenges in teaming, necessitating efficient initial task allocation (ITA) strategies that optimally form complementary human-robot pairs or collaborative chains and establish well-matched task distributions. While current learning-based methods demonstrate promising performance, they often incur high computational costs and lack the flexibility to incorporate user preferences in multi-objective optimization (MOO) or adapt to last-minute changes in dynamic real-world environments. To address these limitations, we propose REBEL, an LLM-based ITA framework that integrates rule-based and experience-enhanced learning to enhance LLM reasoning capabilities and improve in-context adaptability to MOO and situational changes. Extensive experiments validate the effectiveness of REBEL in both single-objective and multi-objective scenarios, demonstrating superior alignment with user preferences and enhanced situational awareness to handle unexpected team composition changes. Additionally, we show that REBEL can complement pre-trained ITA policies, further boosting situational adaptability and overall team performance. Website at https://***/view/ita-rebel. Copyright © 2024, The Authors. All rights reserved.

关键词： Multipurpose robots

Nonparametric Sparse Online Learning of the Koopman Operator

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Hou, Boya Sanjari, Sina Dahlin, Nathan Koppel, Alec Bose, Subhonmesh Carl R. Woese Institute for Genomic Biology University of Illinois Urbana-Champaign United States Department of Mathematics and Computer Science Royal Military College of Canada Canada Department of Electrical and Computer Engineering University at Albany SUNY United States Artificial Intelligence Research JP Morgan Chase & Co United States Department of Electrical and Computer Engineering Coordinated Science Laboratory University of Illinois Urbana-Champaign United States

The Koopman operator provides a powerful framework for representing the dynamics of general nonlinear dynamical systems. Data-driven techniques to learn the Koopman operator typically assume that the chosen function space is closed under system dynamics. In this paper, we study the Koopman operator via its action on the reproducing kernel Hilbert space (RKHS), and explore the mis-specified scenario where the dynamics may escape the chosen function space. We relate the Koopman operator to the conditional mean embeddings (CME) operator and then present an operator stochastic approximation algorithm to learn the Koopman operator iteratively with control over the complexity of the representation. We provide both asymptotic and finite-time last-iterate guarantees of the online sparse learning algorithm with trajectory-based sampling with an analysis that is substantially more involved than that for finite-dimensional stochastic approximation. Numerical examples confirm the effectiveness of the proposed algorithm. © 2024, CC BY-NC-ND.

关键词： Dynamical systems

Matrix Manifold Precoder Design for User-Centric Network Massive MIMO

学校读者我要写书评

暂无评论

Matrix Manifold Precoder Design for User-Centric Network Mas...

IEEE Conference on Global Communications (GLOBECOM)

作者： Rui Sun Li You An-An Lu Chen Sun Ziyu Xiang Xiqi Gao Xiang-Gen Xia National Mobile Communications Research Laboratory Southeast University Nanjing China Purple Mountain Laboratories Nanjing China Department of Electrical and Computer Engineering University of Delaware Newark USA

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

In this paper, we investigate the precoder design for user-centric network (UCN) massive multiple-input multiple-output (mMIMO) downlink with matrix manifold optimization. In UCN mMIMO systems, each user terminal (UT) is served by a subset of the base stations (BSs) instead of all BSs, lowering the dimension of the precoders to be designed. Each BS in the system has a power constraint. By proving that the precoder set satisfying the constraints forms a Riemannian submanifold, we transform the constrained precoder design problem in Euclidean space as an unconstrained one on the Riemannian submanifold. Riemannian ingredients, including orthogonal projection, Riemannian gradient, retraction and vector transport, of the problem on the Riemannian submanifold are further derived, with which the Riemannian conjugate gradient (RCG) design method is proposed for solving the unconstrained problem. The proposed method avoids the inverses of large dimensional matrices. The complexity analyses show the high efficiency of RCG precoder design. Simulation results demonstrate the superiority of the proposed precoder design and the high efficiency of the UCN mMIMO system.

关键词： Manifolds Simulation Precoding Massive MIMO Transforms Performance gain Vectors Complexity theory Global communication Optimization

Transforming Electroluminescence Images to Current-Voltage (IV) Curves Using Deep Learning

学校读者我要写书评

暂无评论

Transforming Electroluminescence Images to Current-Voltage (...

IEEE Conference Photovoltaic Specialists (PVSC)

作者： Brandon K. Byford Jennifer L. Braid Laura E. Boucheron Bruce King Klipsch School of Electrical and Computer Engineering New Mexico State University Las Cruces NM USA Photovoltaic Systems Evaluation Laboratory Sandia National Laboratories Albuquerque NM

ISBN: (数字)9781665464260

ISBN: (纸本)9781665475822

System health monitoring is an essential task in the operation and maintenance of any photovoltaic (PV) system. Typically, electroluminescence (EL), thermal imaging, and current-voltage (IV) curve analyses are used to analyze PV modules with each providing unique insights into system health. While it is relatively easy to acquire an EL or thermal image of a panel in-situ, acquisition of IV curves requires electrical disconnection of the panel from the array and either removal to a solar simulator or characterization and correction for the incident solar spectrum and intensity. In this work we show that, with the use of a transfer-learned Swin transformer model, we can predict accurate IV curves from EL images. Extracting single diode equation parameters from the predicted IV curves yielded an error less than 1% ± 1% for the maximum power point (MPP), short-circuit current $I_{\mathbf{sc}}$ , open-circuit voltage $V_{\mathbf{oc}}$ and photocurrent I. The series resistance $R_{\mathbf{s}}$ and number of series cells $nN_{\mathbf{s}}V_{\mathbf{th}}$ denoted as $N$ were predicted with errors of ~5% ±7% and ~3% ±2%, respectively. Prediction of the shunt resistance $R_{\mathbf{sh}}$ and dark current $\boldsymbol{I}_{\mathbf{o}}$ yielded larger errors, likely due to sensitivity to small changes in the IV curve.

关键词： Resistance Photovoltaic systems Short-circuit currents Noise Imaging Voltage Transformers Mathematical models Electroluminescence Thermal analysis

CodeEnhance: A Codebook-Driven Approach for Low-Light Image Enhancement

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Xu Hou, XianXu Lai, Zhihui Zhou, Jie Zhang, Ya-Nan Pedrycz, Witold Shen, Linlin The Computer Vision Institute College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518060 China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518060 China School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Shenzhen518060 China The Department of Electrical & Computer Engineering University of Alberta University of Alberta Canada

Low-light image enhancement (LLIE) aims to improve low-illumination images. However, existing methods face two challenges: (1) uncertainty in restoration from diverse brightness degradations;(2) loss of texture and color information caused by noise suppression and light enhancement. In this paper, we propose a novel enhancement approach, CodeEnhance, by leveraging quantized priors and image refinement to address these challenges. In particular, we reframe LLIE as learning an image-to-code mapping from low-light images to discrete codebook, which has been learned from high-quality images. To enhance this process, a Semantic Embedding Module (SEM) is introduced to integrate semantic information with low-level features, and a Codebook Shift (CS) mechanism, designed to adapt the pre-learned codebook to better suit the distinct characteristics of our low-light dataset. Additionally, we present an Interactive Feature Transformation (IFT) module to refine texture and color information during image reconstruction, allowing for interactive enhancement based on user preferences. Extensive experiments on both real-world and synthetic benchmarks demonstrate that the incorporation of prior knowledge and controllable information transfer significantly enhances LLIE performance in terms of quality and fidelity. The proposed CodeEnhance exhibits superior robustness to various degradations, including uneven illumination, noise, and color distortion. Copyright © 2024, The Authors. All rights reserved.

关键词： Image enhancement