检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Wang, Alan Q. Karaman, Batuhan K. Kim, Heejong Rosenthal, Jacob Saluja, Rachit Young, Sean I. Sabuncu, Mert R. School of Electrical and Computer Engineering Cornell University Cornell Tech New YorkNY10044 United States Department of Radiology Weill Cornell Medical School New YorkNY10065 United States Weill Cornell/Rockefeller/Sloan Kettering Tri-Institutional MD-PhD Program New YorkNY10065 United States Martinos Center for Biomedical Imaging Harvard Medical School BostonMA02129 United States Computer Science and Artificial Intelligence Laboratory MIT CambridgeMA02139 United States

Interpretability for machine learning models in medical imaging (MLMI) is an important direction of research. However, there is a general sense of murkiness in what interpretability means. Why does the need for interpretability in MLMI arise? What goals does one actually seek to address when interpretability is needed? To answer these questions, we identify a need to formalize the goals and elements of interpretability in MLMI. By reasoning about real-world tasks and goals common in both medical image analysis and its intersection with machine learning, we identify five core elements of interpretability: localization, visual recognizability, physical attribution, model transparency, and actionability. From this, we arrive at a framework for interpretability in MLMI, which serves as a step-by-step guide to approaching interpretability in this context. Overall, this paper formalizes interpretability needs in the context of medical imaging, and our applied perspective clarifies concrete MLMI-specific goals and considerations in order to guide method design and improve real-world usage. Our goal is to provide practical and didactic information for model designers and practitioners, inspire developers of models in the medical imaging field to reason more deeply about what interpretability is achieving, and suggest future directions of interpretability research. © 2023, CC0.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Template-free try-on image synthesis via semantic-guided optimization

arXiv

引用

arXiv 2021年

作者： Chou, Chien-Lung Chen, Chieh-Yun Hsieh, Chia-Wei Shuai, Hong-Han Liu, Jiaying Cheng, Wen-Huang Department of Electrical and Computer Engineering National Chiao Tung University Taiwan Institute of Electronics National Chiao Tung University Taiwan Artificial Intelligence and Data Science Program National Chung Hsing University Taiwan Wangxuan Institute of Computer Technology Peking University China

The virtual try-on task is so attractive that it has drawn considerable attention in the field of computer vision. However, presenting the three-dimensional (3D) physical characteristic (e.g., pleat and shadow) based on a 2D image is very challenging. Although there have been several previous studies on 2D-based virtual try-on work, most 1) required user-specified target poses that are not user-friendly and may not be the best for the target clothing, and 2) failed to address some problematic cases, including facial details, clothing wrinkles and body occlusions. To address these two challenges, in this paper, we propose an innovative template-free try-on image synthesis (TF-TIS) network. The TF-TIS first synthesizes the target pose according to the user-specified in-shop clothing. Afterward, given an in-shop clothing image, a user image, and a synthesized pose, we propose a novel model for synthesizing a human try-on image with the target clothing in the best fitting pose. The qualitative and quantitative experiments both indicate that the proposed TF-TIS outperforms the state-of-the-art methods, especially for difficult cases. Copyright © 2021, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Asymptotic analysis of an ensemble of randomly projected linear discriminants

arXiv

引用

arXiv 2020年

作者： Niyazi, Lama B. Kammoun, Abla Dahrouj, Hayssam Alouini, Mohamed-Slim Al-Naffouri, Tareq Y. Electrical Engineering Program King Abdullah University of Science and Technology Thuwal Saudi Arabia Department of Electrical and Computer Engineering Effat University Jeddah Saudi Arabia

Datasets from the fields of bioinformatics, chemometrics, and face recognition are typically characterized by small samples of high-dimensional data. Among the many variants of linear discriminant analysis that have been proposed in order to rectify the issues associated with classification in such a setting, the classifier in [1], composed of an ensemble of randomly projected linear discriminants, seems especially promising;it is computationally efficient and, with the optimal projection dimension parameter setting, is competitive with the state-of-the-art. In this work, we seek to further understand the behavior of this classifier through asymptotic analysis. Under the assumption of a growth regime in which the dataset and projection dimensions grow at constant rates to each other, we use random matrix theory to derive asymptotic misclassification probabilities showing the effect of the ensemble as a regularization of the data sample covariance matrix. The asymptotic errors further help to identify situations in which the ensemble offers a performance advantage. We also develop a consistent estimator of the misclassification probability as an alternative to the computationally-costly cross-validation estimator, which is conventionally used for parameter tuning. Finally, we demonstrate the use of our estimator for tuning the projection dimension on both real and synthetic data. Copyright © 2020, The Authors. All rights reserved.

关键词： Random variables

来源：评论

学校读者我要写书评

暂无评论

Graph Laplacian Learning with Exponential Family Noise

引用

IEEE Transactions on Signal and Information Processing over Networks 2025年

作者： Shi, Changhao Mishne, Gal UC San Diego Electrical and Computer Engineering Department CA92093 United States UC San Diego Hal.c.o.glu Data Science Institute and the Neurosciences Graduate Program CA92093 United States

Graph signal processing (GSP) is a prominent framework for analyzing signals on non-Euclidean domains. The graph Fourier transform (GFT) uses the combinatorial graph Laplacian matrix to reveal the spectral decomposition of signals in the graph frequency domain. However, a common challenge in applying GSP methods is that in many scenarios the underlying graph of a system is unknown. A solution in such cases is to construct the unobserved graph from available data, which is commonly referred to as graph or network inference. Although different graph inference methods exist, they are restricted to learning from either smooth graph signals or simple additive Gaussian noise. Other types of noisy data, such as discrete counts or binary digits, are rather common in real-world applications, yet are underexplored in graph inference. In this paper, we propose a versatile graph inference framework for learning from graph signals corrupted by exponential family noise. Our framework generalizes previous methods from continuous smooth graph signals to various data types. We propose an alternating algorithm that jointly estimates the graph Laplacian and the unobserved smooth representation from the noisy signals. We also extend our approach to include an offset variable which models different levels of variation of the nodes. Since real-world graph signals are frequently non-independent and temporally correlated, we further adapt our original setting to a timevertex formulation. We demonstrate on synthetic and real-world data that our new algorithms outperform competing Laplacian estimation methods that suffer from noise model mismatch. © 2025 IEEE.

关键词： Laplace transforms

来源：评论

学校读者我要写书评

暂无评论

3D OCT characterization and quantification of refractive indices of bacteria and biofilms with antibiotic interventions

3D OCT characterization and quantification of refractive ind...

引用

Optical Molecular Probes, Imaging and Drug Delivery, OMP 2021 - Part of Biophotonics Congress: Optics in the Life sciences 2021

作者： Zaki, Farzana R. Sudhir, Kavya Won, Jungeun Monroy, Guillermo L. Choi, Honggu Chaney, Eric J. Spillman, Darold R. Boppart, Stephen A. Beckman Institute for Advanced Science and Technology 405 North Mathews Avenue UrbanaIL61801 United States University of Illinois at Urbana–Champaign Department of Bioengineering 1304 West Springfield Avenue UrbanaIL61801 United States University of Illinois at Urbana–Champaign Department of Electrical and Computer Engineering 306 North Wright Street UrbanaIL61801 United States Carle Illinois College of Medicine University of Illinois at Urbana-Champaign 807 S. Wright St. ChampaignIL61820 United States

Using 3D OCT, we have quantified the refractive indices of ear-infection causing bacteria and biofilms with antibiotics interventions. This approach reveals changing optical properties of bacteria in biofilms followin...

ISBN: (纸本)9781557528209

关键词：

来源：评论

学校读者我要写书评

暂无评论

RSAC: Regularized subspace approximation classifier for lightweight continuous learning

arXiv

引用

arXiv 2020年

作者： Ho, Chih-Hsing Tsai, Shang-Ho Department of Electrical Engineering and Computer Science Honor Program National Chiao Tung University Hsinchu Taiwan Department of Electrical Engineering National Chiao Tung University Hsinchu Taiwan

Continuous learning seeks to perform the learning on the data that arrives from time to time. While prior works have demonstrated several possible solutions, these approaches require excessive training time as well as memory usage. This is impractical for applications where time and storage are constrained, such as edge computing. In this work, a novel training algorithm, regularized subspace approximation classifier (RSAC), is proposed to achieve lightweight continuous learning. RSAC contains a feature reduction module and classifier module with regularization. Extensive experiments show that RSAC is more efficient than prior continuous learning works and outperforms these works on various experimental settings. Copyright © 2020, The Authors. All rights reserved.

关键词： Approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Ferroelectric transistors based on shear-transformation-mediated rhombohedral-stacked molybdenum disulfide

引用

Nature Electronics 2024年第1期7卷 29-38页

作者： Yang, Tilo H. Liang, Bor-Wei Hu, Hsiang-Chi Chen, Fu-Xiang Ho, Sheng-Zhu Chang, Wen-Hao Yang, Liu Lo, Han-Chieh Kuo, Tzu-Hao Chen, Jyun-Hong Lin, Po-Yen Simbulan, Kristan Bryan Luo, Zhao-Feng Chang, Alice Chinghsuan Kuo, Yi-Hao Ku, Yu-Seng Chen, Yi-Cheng Huang, You-Jia Chang, Yu-Chen Chiang, Yu-Fan Lu, Ting-Hua Lee, Min-Hung Li, Kai-Shin Wu, Menghao Chen, Yi-Chun Lin, Chun-Liang Lan, Yann-Wen Department of Physics National Taiwan Normal University Taipei Taiwan Taiwan Semiconductor Research Institute National Applied Research Laboratories Hsinchu Taiwan Department of Electrophysics National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Physics National Cheng Kung University Tainan Taiwan School of Physics Huazhong University of Science and Technology Wuhan China Institute of Cellular and Organismic Biology Academia Sinica Taipei Taiwan Department of Mathematics and Physics University of Santo Tomas Manila Philippines Program for Semiconductor Devices Materials and Hetero-integration Graduate School of Advanced Technology National Taiwan University Taipei Taiwan Hsinchu Taiwan Department of Physics National Taiwan University Taipei Taiwan Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA United States

To develop low-power, non-volatile computing-in-memory device using ferroelectric transistor technologies, ferroelectric channel materials with scaled thicknesses are required. Two-dimensional semiconductors, such as molybdenum disulfide (MoS2), equipped with sliding ferroelectricity could provide an answer. However, achieving switchable electric polarization in epitaxial MoS2 remains challenging due to the absence of mobile domain boundaries. Here we show that polarity-switchable epitaxial rhombohedral-stacked (3R) MoS2 can be used as a ferroelectric channel in ferroelectric memory transistors. We show that a shear transformation can spontaneously occur in 3R MoS2 epilayers, producing heterostructures with stable ferroelectric domains embedded in a highly dislocated and unstable non-ferroelectric matrix. This diffusionless phase transformation process produces mobile screw dislocations that enable collective polarity control of 3R MoS2 via an electric field. Polarization–electric-field measurements reveal a switching field of 0.036 V nm−1 for shear-transformed 3R MoS2. Our sliding ferroelectric transistors are non-volatile memory units with thicknesses of only two atomic layers and exhibit an average memory window of 7 V with an applied voltage of 10 V, retention times greater than 104 seconds and endurance greater than 104 cycles. © 2023, The Author(s), under exclusive licence to Springer Nature Limited.

关键词： Molybdenum disulfide

来源：评论

学校读者我要写书评

暂无评论

Transformation between elastic dipoles, quadrupoles, octupoles and hexadecapoles driven by surfactant self-assembly in nematic emulsion

arXiv

引用

arXiv 2022年

作者： Senyuk, Bohdan Mozaffari, Ali Crust, Kevin Zhang, Rui de Pablo, Juan J. Smalyukh, Ivan I. Department of Physics and Soft Materials Research Center University of Colorado Boulder CO United States Pritzker School of Molecular Engineering The University of Chicago ChicagoIL United States Department of Physics The Hong Kong University of Science and Technology Clear Water Bay Kowloon Hong Kong Center for Molecular Engineering Argonne National Laboratory LemontIL United States Department of Electrical Computer and Energy Engineering Materials Science and Engineering Program University of Colorado Boulder CO United States Renewable and Sustainable Energy Institute National Renewable Energy Laboratory University of Colorado Boulder CO United States

Emulsions comprising isotropic fluid drops within a nematic host are of interest for applications ranging from biodetection to smart windows, which rely on changes of molecular alignment structures around the drops in response to chemical, thermal, electric and other stimuli. We show that absorption or desorption of trace amounts of common surfactants can drive continuous transformations of elastic multipoles induced by the droplets within the uniformly aligned nematic host. Out-of-equilibrium dynamics of director structures emerge from a controlled self-assembly or desorption of different surfactants at the drop-nematic interfaces, with ensuing forward and reverse transformations between elastic dipoles, quadrupoles, octupoles and hexadecapoles. We characterize inter-transformations of droplet-induced surface and bulk defects, probe elastic pair interactions and discuss emergent prospects for fundamental science and applications of the reconfigurable nematic emulsions. © 2022, CC BY-NC-ND.

关键词： Surface active agents

来源：评论

学校读者我要写书评

暂无评论

Structural, optical, and thermal properties of BN thin films grown on diamond via pulsed laser deposition

arXiv

引用

arXiv 2023年

作者： Biswas, Abhijit Alvarez, Gustavo A. Li, Tao Christiansen-Salameh, Joyce Jeong, Eugene Puthirath, Anand B. Iyengar, Sathvik Ajay Li, Chenxi Gray, Tia Zhang, Xiang Pieshkov, Tymofii S. Kannan, Harikishan Elkins, Jacob Vajtai, Robert Birdwell, A. Glen Neupane, Mahesh R. Garratt, Elias J. Pate, Bradford B. Ivanov, Tony G. Zhao, Yuji Tian, Zhiting Ajayan, Pulickel M. Department of Materials Science and Nanoengineering Rice University HoustonTX77005 United States Sibley School of Mechanical and Aerospace Engineering Cornell University IthacaNY14853 United States Department of Electrical and Computer Engineering Rice University HoustonTX77005 United States Applied Physics Graduate Program Smalley-Curl Institute Rice University HoustonTX77005 United States DEVCOM Army Research Laboratory RF Devices and Circuits AdelphiMD20783 United States Naval Research Laboratory Chemistry Division WashingtonDC20375 United States

Heterostructures based on ultrawide-bandgap (UWBG) semiconductors (bandgap >4.0 eV), such as boron nitride (BN) and diamond, hold significant importance for the development of high-power electronics in the next generation. However, achieving in-situ hetero-epitaxy of BN/diamond or vice versa remains exceptionally challenging due to the complex growth kinetics involved. In this study, we have grown BN thin film on (100) single crystal diamonds using pulsed laser deposition and investigated its structural, magnetic, optical, and thermal properties. The structural analyses confirmed the growth of BN films, which exhibited diamagnetic behavior at room temperature. Notably, the film demonstrated anisotropic refractive index characteristics within the visible-to-near-infrared wavelength range. The room temperature cross-plane thermal conductivity of BN is 1.53 (± 0.77) W/mK, while the thermal conductance of the BN/diamond interface is 20 (± 2) MW/m2K. These findings have significant implications for a range of device applications based on UWBG BN/diamond heterostructures. © 2023, CC BY.

关键词： Diamonds

来源：评论

学校读者我要写书评

暂无评论

A parameter-free differential evolution algorithm for the analytic continuation of imaginary time correlation functions

arXiv

引用

arXiv 2022年

作者： Nichols, Nathan S. Sokol, Paul Del Maestro, Adrian Data Science and Learning Division Argonne National Laboratory ArgonneIL60439 United States Department of Physics University of Vermont BurlingtonVT05405 United States Materials Science Program University of Vermont BurlingtonVT05404 United States Department of Physics Indiana University BloomingtonIN47408 United States Department of Physics and Astronomy University of Tennessee KnoxvilleTN37996 United States Min H. Kao Department of Electrical Engineering and Computer Science University of Tennessee KnoxvilleTN37996 United States

We report on Differential Evolution for Analytic Continuation (DEAC): a parameter-free evolutionary algorithm to generate the dynamic structure factor from imaginary time correlation functions. Our approach to this long-standing problem in quantum many-body physics achieves enhanced spectral fidelity while using fewer compute (CPU) hours. The need for fine-tuning of algorithmic control parameters is eliminated by embedding them within the genome to be optimized for this evolutionary computation based algorithm. Benchmarks are presented for models where the dynamic structure factor is known exactly, and experimentally relevant results are included for quantum Monte Carlo simulations of bulk 4He below the superuid transition temperature. Copyright © 2022, The Authors. All rights reserved.

关键词： Functional analysis

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：