检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Xu, Yongxin Jiang, Xinke Chu, Xu Qiu, Rihong Feng, Yujie Ding, Hongxin Zhao, Junfeng Wang, Yasha Xie, Bing School of Computer Science School of Software & Microelectronics Peking University Beijing China Key Laboratory of High Confidence Software Technologies Ministry of Education Beijing China National Engineering Research Center For Software Engineering Peking University Beijing China Center on Frontiers of Computing Studies Peking University Beijing China Peking University Information Technology Institute Binhai Tianjin China Nanhu Laboratory Jiaxing China Department of Computing The Hong Kong Polytechnic University Hong Kong

ISBN: (纸本)157735897X

Exploring the correlations between medical features is essential for extracting patient health patterns from electronic health records (EHR) data, and strengthening medical predictions and decision-making. To constrain the hypothesis space of pure data-driven deep learning in the context of limited annotated data, a common trend is to incorporate external knowledge, especially knowledge priors related to personalized health contexts, to optimize model training. However, most existing methods lack flexibility and are constrained by the uncertainties brought about by fixed feature correlation priors. In addition, in utilizing knowledge, these methods overlook the knowledge informative for personalized healthcare. To this end, we propose DearLLM, a novel and effective framework that leverages feature correlations deduced by large language models (LLMs) to enhance personalized healthcare. Concretely, DearLLM captures and learns quantitative correlations between medical features by calculating the conditional perplexity of LLMs’ deduction based on personalized patient backgrounds. Then, DearLLM enhances healthcare predictions by emphasizing knowledge that carries unique patient information through a feature-frequency-aware graph pooling method. Extensive experiments on two real-world benchmark datasets show significant performance gains brought by DearLLM. Furthermore, the discovered findings align well with medical literature, offering meaningful clinical interpretations. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Direct Zernike Coefficient Prediction from Point Spread Functions and Extended Images using Deep Learning

arXiv

引用

arXiv 2024年

作者： Kok, Yong En Bentley, Alexander Parkes, Andrew Wright, Amanda J. Somekh, Michael G. Pound, Michael School of Computer Science University of Nottingham NottinghamNG8 1BB United Kingdom Optics and Photonics Group Department of Electrical and Electronic Engineering University of Nottingham NottinghamNG7 2RD United Kingdom Research Center for Humanoid Sensing Zhejiang Laboratory Hangzhou3111100 China

Optical imaging quality can be severely degraded by system and sample induced aberrations. Existing adaptive optics systems typically rely on iterative search algorithm to correct for aberrations and improve images. This study demonstrates the application of convolutional neural networks to characterise the optical aberration by directly predicting the Zernike coefficients from two to three phase-diverse optical images. We evaluated our network on 600,000 simulated Point Spread Function (PSF) datasets randomly generated within the range of -1 to 1 radians using the first 25 Zernike coefficients. The results show that using only three phase-diverse images captured above, below and at the focal plane with an amplitude of 1 achieves a low RMSE of 0.10 radians on the simulated Point Spread Function (PSF) dataset. Furthermore, this approach directly predicts Zernike modes simulated extended 2D samples, while maintaining a comparable RMSE of 0.15 radians. We demonstrate that this approach is effective using only a single prediction step, or can be iterated a small number of times. This simple and straightforward technique provides rapid and accurate method for predicting the aberration correction using three or less phase-diverse images, paving the way for evaluation on real-world dataset. Copyright © 2024, The Authors. All rights reserved.

关键词： Optical transfer function

来源：评论

学校读者我要写书评

暂无评论

Digital versus Analog Transmissions for Federated Learning over Wireless Networks

Digital versus Analog Transmissions for Federated Learning o...

引用

IEEE International Conference on Communications (ICC)

作者： Jiacheng Yao Wei Xu Zhaohui Yang Xiaohu You Mehdi Bennis H. Vincent Poor National Mobile Communications Research Laboratory Southeast University Nanjing China Purple Mountain Laboratories Nanjing China College of Information Science and Electronic Engineering Zhejiang University Hangzhou China Center for Wireless Communications University of Oulu Oulu Finland Department of Elect. & Computer Eng. Princeton University Princeton NJ USA

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

In this paper, we quantitatively compare these two effective communication schemes, i.e., digital and analog ones, for wireless federated learning (FL) over resource-constrained networks, highlighting their essential differences as well as their respective application scenarios. We first examine both digital and analog transmission methods, together with a unified and fair comparison scheme under practical constraints. A universal convergence analysis under various imperfections is established for FL performance evaluation in wireless networks. These analytical results reveal that the fundamental difference between the two paradigms lies in whether communication and computation are jointly designed or not. The digital schemes decouple the communication design from specific FL tasks, making it difficult to support simultaneous uplink transmission of massive devices with limited bandwidth. In contrast, the analog communication allows over-the-air computation (AirComp), thus achieving efficient spectrum utilization. However, computation-oriented analog transmission reduces power efficiency, and its performance is sensitive to computational errors. Finally, numerical simulations are conducted to verify these theoretical observations.

关键词： Performance evaluation Wireless sensor networks Accuracy Uncertainty Federated learning Wireless networks Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Interaction homotopy and interaction homology

arXiv

引用

arXiv 2023年

作者： Liu, Jian Chen, Dong Wei, Guo-Wei Mathematical Science Research Center Chongqing University of Technology Chongqing400054 China Department of Mathematics Michigan State University MI48824 United States Department of Electrical and Computer Engineering Michigan State University MI48824 United States Department of Biochemistry and Molecular Biology Michigan State University MI48824 United States

Interactions in complex systems are widely observed across various fields, drawing increased attention from researchers. In mathematics, efforts are made to develop various theories and methods for studying the interactions between spaces. In this work, we present an algebraic topology framework to explore interactions between spaces. We introduce the concept of interaction spaces and investigate their homotopy, singular homology, and simplicial homology. Furthermore, we demonstrate that interaction singular homology serves as an invariant under interaction homotopy. We believe that the proposed framework holds potential for practical applications. Copyright © 2023, The Authors. All rights reserved.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Predicting Visual Acuity Recovery in Open Globe Injuries: A Deep Learning Approach Using Orbital Computed Tomography

SSRN

引用

SSRN 2025年

作者： Wang, Kexin Zhang, Hongda Xie, Zhuoting Li, Shihuai Giunchiglia, Fausto Xu, Hao Han, Qinghe Guo, Deming Qin, Yanguo Orthopaedic Medical Center The Second Norman Bethune Hospital of Jilin University Changchun130041 China Joint International Research Laboratory of Ageing Active Strategy and Bionic Health in Northeast Asia Ministry of Education Jilin Provincial Key Laboratory of Orthopaedics Jilin University Changchun130041 China College of Computer Science and Technology Jilin University Changchun130012 China Department of Information Engineering and Computer Science University of Trento The Autonomous Province of Trento Italy Radiology Department The Second Norman Bethune Hospital of Jilin University Changchun China

Purpose: This study aims to develop and validate an artificial intelligence (AI) model based on DenseNet3D to predict visual outcomes in patients with open globe injury (OGI) using orbital computed tomography (CT) scans. The model seeks to guide emergency treatment decisions and improve clinical management. Design: A retrospective study designed to develop and validate a deep learning-based prediction model for visual outcomes in OGI patients. Participants: The study included 391 OGI patients treated at the Second Hospital of Jilin University. Methods: Orbital CT scans from 391 patients were annotated using ITK-SNAP to identify key OGI features, including intraocular foreign bodies, hyphema, traumatic cataract, lens dislocation, vitreous hemorrhage, posterior scleral rupture, retinal tears, retinal detachment, subretinal hemorrhage, retinal prolapse, choroidal injuries, optic nerve damage, and orbital fractures. A modified DenseNet architecture extracted imaging features from the annotated regions. The model's performance in predicting postoperative visual recovery was evaluated using five-fold cross-validation, with metrics including accuracy, F1 score, precision, recall, area under the precision-recall curve (PR-AUC), and area under the receiver operating characteristic curve (ROC-AUC). Gradient-weighted Class Activation Mapping (Grad-CAM) was applied to generate visual heatmaps, providing insights into the model's interpretability. The model’s performance was compared with the Ocular Trauma Score (OTS) to assess its consistency with traditional clinical methods. Main Outcome Measures: Accuracy, F1 score, PR AUC, Precision, Recall, and ROC AUC of the *** model exhibited strong predictive performance, achieving an accuracy of 0.824, an F1 score of 0.870, a PR-AUC of 0.946, a precision of 0.858, a recall of 0.889, and an ROC-AUC of 0.906. These metrics demonstrated performance comparable to the conventional OTS system. Conclusions: The proposed AI model

关键词： computerized tomography

来源：评论

学校读者我要写书评

暂无评论

Boundary-to-Solution Mapping for Groundwater Flows in a Toth Basin

arXiv

引用

arXiv 2023年

作者： Sun, Jingwei Li, Jun Hao, Yonghong Qi, Cuiting Ma, Chunmei Sun, Huazhi Begashaw, Negash Comet, Gurcan Sun, Yi Wang, Qi School of Computer Science Tianjin Normal University Tianjin100084 China School of Mathematical Science Tianjin Normal University Tianjin300387 China Center for Ground Water Research Tianjin Normal University Tianjin100084 China School of Geographical and Environmental Science Tianjin Normal University Tianjin100084 China Mathematics Computer Science Physics and Engineering Department Benedict College ColumbiaSC20204 United States Department of Mathematics University of South Carolina ColumbiaSC29208 United States

In this paper, the authors propose a new approach to solving the groundwater flow equation in the Toth basin of arbitrary top and bottom topographies using deep learning. Instead of using traditional numerical solvers, they use a DeepONet to produce the boundary-to-solution mapping. This mapping takes the geometry of the physical domain along with the boundary conditions as inputs to output the steady state solution of the groundwater flow equation. To implement the DeepONet, the authors approximate the top and bottom boundaries using truncated Fourier series or piecewise linear representations. They present two different implementations of the DeepONet: one where the Toth basin is embedded in a rectangular computational domain, and another where the Toth basin with arbitrary top and bottom boundaries is mapped into a rectangular computational domain via a nonlinear transformation. They implement the DeepONet with respect to the Dirichlet and Robin boundary condition at the top and the Neumann boundary condition at the impervious bottom boundary, respectively. Using this deep-learning enabled tool, the authors investigate the impact of surface topography on the flow pattern by both the top surface and the bottom impervious boundary with arbitrary geometries. They discover that the average slope of the top surface promotes long-distance transport, while the local curvature controls localized circulations. Additionally, they find that the slope of the bottom impervious boundary can seriously impact the long-distance transport of groundwater flows. Overall, this paper presents a new and innovative approach to solving the groundwater flow equation using deep learning, which allows for the investigation of the impact of surface topography on groundwater flow patterns. Copyright © 2023, The Authors. All rights reserved.

关键词： Groundwater flow

来源：评论

学校读者我要写书评

暂无评论

Low-energy electronic structure of perovskite and Ruddlesden-Popper semiconductors in the Ba-Zr-S system probed by bond-selective polarized x-ray absorption spectroscopy, infrared reflectivity, and Raman scattering

引用

Physical Review B 2022年第19期105卷 195203-195203页

作者： Kevin Ye Nathan Z. Koocher Stephen Filippone Shanyuan Niu Boyang Zhao Matthew Yeung Sharon Bone Adam J. Robinson Patrick Vora André Schleife Long Ju Alexey Boubnov James M. Rondinelli Jayakanth Ravichandran R. Jaramillo Department of Materials Science and Engineering Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Department of Materials Science and Engineering Northwestern University Evanston Illinois 60208 USA Mork Family Department of Chemical Engineering and Materials Science University of Southern California Los Angeles California 90089 USA Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Stanford Synchrotron Radiation Lightsource SLAC National Accelerator Laboratory Menlo Park California 94025 USA Department of Physics and Astronomy and Quantum Science and Engineering Center George Mason University Fairfax Virginia 22030 USA Department of Materials Science and Engineering Materials Research Laboratory and National Center for Supercomputing Applications University of Illinois at Urbana-Champaign Urbana Illinois 61801 USA Department of Physics Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Institute of Nanotechnology Karlsruhe Institute of Technology Eggenstein-Leopoldshafen 76344 Germany

Chalcogenides in perovskite and the related layered Ruddlesden-Popper crystal structures (chalcogenide perovskites for brevity) are an exciting family of semiconductors but remain experimentally little studied. Chalcogenide perovskites share crystal structures and some physical properties with ionic compounds such as oxide and halide perovskites, but the metal-chalcogen bonds responsible for semiconducting behavior are substantially more covalent than in these more-studied perovskites. Here, we use complementary experimental and theoretical methods to study how the mixed ionic-covalent Zr-S bonds support the electronic structure and physical properties of perovskite BaZrS3 and Ruddlesden-Popper Ba3Zr2S7. We apply theoretical methods to assign features of experimentally measured x-ray absorption spectroscopy (XAS) to particular orbital transitions, enabling a clear physical interpretation of angle-dependent, polarized XAS data measured on single-crystal samples, and an atomistic view of the covalent bonding network that facilitates charge transport. Polarized Raman measurements identify signatures of crystalline anisotropy in Ba3Zr2S7 and enable the first assignments of mode symmetry in this material. Infrared reflectivity reveals electronic transport properties that augur well for the use of chalcogenide perovskites in optoelectronic and energy-conversion technologies.

关键词： Electrical conductivity Electronic structure of atoms & molecules Chalcogenides Perovskites Raman spectroscopy Reflectivity X-ray absorption near-edge spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Squeezing-enhanced quantum sensing with quadratic optomechanics

Optica Quantum

引用

Optica Quantum 2024年第4期2卷 222-229页

作者： Sheng-Dian Zhang Jie Wang Qian Zhang Ya-Feng Jiao Yun-Lan Zuo Şahin K. Özdemir Cheng-Wei Qiu Franco Nori Hui Jing Key Laboratory of Low-Dimensional Quantum Structures and Quantum Control of Ministry of Education Department of Physics and Synergetic Innovation Center for Quantum Effects and Applications Hunan Normal Universityhttps://***/053w1zy07 Changsha 410081 China Department of Engineering Science and Mechanics and Materials Research Institute Pennsylvania State University University Park State College Pennsylvania 16802 USA Department of Electrical and Computer Engineering National University of Singapore Singapore 117583 Singapore Theoretical Quantum Physics Laboratory RIKEN Cluster for Pioneering Research Wako-shi Saitama 351-0198 Japan Physics Department The University of Michigan Ann Arbor Michigan 48109-1040 USA

Cavity optomechanical (COM) sensors, enhanced by quantum squeezing or entanglement, have become powerful tools for measuring ultra-weak forces with high precision and sensitivity. However, these sensors usually rely on linear COM couplings, a fundamental limitation when measurements of the mechanical energy are desired. Very recently, a giant enhancement of the signal-to-noise ratio was predicted in a quadratic COM system. Here we show that the performance of such a system can be further improved surpassing the standard quantum limit by using quantum squeezed light. Our approach is compatible with available engineering techniques of advanced COM sensors and provides new opportunities for using COM sensors in tests of fundamental laws of physics and quantum metrology applications.

关键词： Optical systems Photonic crystals Photonic entanglement Quantum metrology Quantum noise Squeezed states

来源：评论

学校读者我要写书评

暂无评论

Transport-Hub-Aware Spatial-Temporal Adaptive Graph Transformer for Traffic Flow Prediction

arXiv

引用

arXiv 2023年

作者： Xu, Xiao Zhang, Lei Liu, Bailong Liang, Zhizhen Zhang, Xuefei The School of Computer Science and Technology China University of Mining and Technology Jiangsu Xuzhou221116 China Mine Digitalization Department of Engineering Research Center Jiangsu Xuzhou221116 China Jiangsu Hengwang Digital Technology Co. Ltd. Jiangsu Suzhou215000 China

As a core technology of Intelligent Transportation System (ITS), traffic flow prediction has a wide range of applications. Traffic flow data are spatial-temporal, which are not only correlated to spatial locations in road networks, but also vary with temporal time indices. Existing methods have solved the challenges in traffic flow prediction partly, focusing on modeling spatial-temporal dependencies effectively, while not all intrinsic properties of traffic flow data are utilized fully. Besides, there are very few attempts at incremental learning of spatial-temporal data mining, and few previous works can be easily transferred to the traffic flow prediction task. Motivated by the challenge of incremental learning methods for traffic flow prediction and the underutilization of intrinsic properties of road networks, we propose a Transport-Hub-aware Spatial-Temporal adaptive graph transFormer (H-STFormer) for traffic flow prediction. Specifically, we first design a novel spatial self-attention module to capture the dynamic spatial dependencies. Three graph masking matrices are integrated into spatial self-attentions to highlight both short- and long-term dependences. Additionally, we employ a temporal self-attention module to detect dynamic temporal patterns in the traffic flow data. Finally, we design an extra spatial-temporal knowledge distillation module for incremental learning of traffic flow prediction tasks. Through extensive experiments, we show the effectiveness of H-STFormer in normal and incremental traffic flow prediction tasks. The code is available at https://***/Fantasy-Shaw/H-STFormer. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Design of a Robust µ-synthesis controller for Uncertain and Disturbed Nonlinear Wind turbines

Design of a Robust µ-synthesis controller for Uncertain and...

引用

International Conference on Environment and Electrical engineering (EEEIC)

作者： Soroush Azizi Mohammad Hassan Asemani Navid Vafamand Saleh Mobayen Afef Fekih Department of Electrical and Computer Engineering Shiraz University Shiraz Iran Future Technology Research Center National Yunlin University of Science and Technology Yunlin Taiwan University of Louisiana at Lafayette Lafayette Louisiana USA

ISBN: (数字)9781665485371

ISBN: (纸本)9781665485388

This work proposes a novel robust linear output-feedback controller that aims to ensure the closed-loop stability and disturbance reduction in nonlinear wind turbine systems. The developed approach uses the numerical µ-synthesis and mixed µ-synthesis methods to design the robust controller. Since the designed controller is linear, its implementation has a low online computational burden. Also, only measurements of the wind turbine are used in the control feedback law, thus eliminating the need for any additional sensors or signal processing methods to acquire the state vector information. In this control method, the wind turbine's parameter uncertainties and external disturbances are well treated by means of robust stability and good tracking. Simulation results indicate the high performance of the proposed controller. Further comparison with a $H$ ∞ control approach, confirmed its superior performance, especially in terms of the fast convergence rate of the controlled outputs to their desired values.

关键词： Uncertain systems Uncertainty Simulation Power system stability Signal processing Aerodynamics Stability 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：