检索结果-内蒙古大学图书馆

Prediction of ultraslow magnetic solitons via plasmon-induced transparency by artificial neural networks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Cheng, Jiaxi Xu, Siliu Jiang, Shengwang Bo, Zhiqiang Department of Computer Science University of Waikato Hamilton New Zealand School of Electronic and Information Engineering Hubei University of Science and Technology Xianning China University of Electronic Science and Technology of China Chengdu China Software Engineering Institute Xidian University Xi'an China

Plasmon-induced transparency (PIT) in advanced materials has attracted extensive attention for both theoretical and applied physics. Here, we considered a scheme that can produce PIT and studied the characteristics of ultraslow low-power magnetic solitons. The PIT metamaterial is constructed as an array of unit cells that consist of two coupled varactor-loaded split-ring resonators. Simulations verified that ultraslow magnetic solitons can be generated in this type of metamaterial. To solve nonlinear equations, various types of numerical methods can be applied by virtue of exact solutions, which are always difficult to acquire. However, the initial conditions and propagation distance impact the ultimate results. In this article, an artificial neural network (ANN) was used as a supervised learning model to predict the evolution and final mathematical expressions through training based on samples with disparate initial conditions. Specifically, the influences of the number of hidden layers were discussed. Additionally, the learning results obtained by employing several training algorithms were analyzed and compared. Our research opens a route for employing machine learning algorithms to save time in both physical and engineering applications of Schrödinger-type systems. Copyright © 2021, The Authors. All rights reserved.

关键词： Numerical methods

A Brief Review of Explainable Artificial Intelligence in Healthcare

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Sadeghi, Zahra Alizadehsani, Roohallah Cifci, Mehmet Akif Kausar, Samina Rehman, Rizwan Mahanta, Priyakshi Bora, Pranjal Kumar Almasri, Ammar Alkhawaldeh, Rami S. Hussain, Sadiq Alatas, Bilal Shoeibi, Afshin Moosaei, Hossein Hladík, Milan Nahavandi, Saeid Pardalos, Panos M. Institute for Big Data Analytics Faculty of Computer Science Dalhousie University Canada Deakin University Geelong Australia The Institute of Computer Technology Tu Wien University Vienna1040 Austria University of Kotli Azad Jammu and Kashmir Azad Kashmir Kotli Pakistan Centre for Computer Science and Applications Dibrugarh University Assam India Department of Management Information Sys Al-Balqa Applied University Salt19117 Jordan Department of Computer Information Systems The University of Jordan Aqaba77110 Jordan Examination Branch Dibrugarh University Assam Dibrugarh India Department of Software Eng. Firat University Elazig23100 Turkey Data Science and Computational Intelligence Institute University of Granada Spain Department of Informatics Faculty of Science Jan Evangelista Purkyně University in Ústí nad Labem Czech Republic Department of Applied Mathematics School of Computer Science Faculty of Mathematics and Physics Charles University Prague Czech Republic Harvard Paulson School of Engineering and Applied Sciences Harvard University AllstonMA02134 United States Center for Applied Optimization Department of Industrial and Systems Engineering University of Florida Gainesville32611 United States

XAI refers to the techniques and methods for building AI applications which assist end users to interpret output and predictions of AI models. Black box AI applications in high-stakes decision-making situations, such as medical domain have increased the demand for transparency and explainability since wrong predictions may have severe consequences. Model explainability and interpretability are vital successful deployment of AI models in healthcare practices. AI applications' underlying reasoning needs to be transparent to clinicians in order to gain their trust. This paper presents a systematic review of XAI aspects and challenges in the healthcare domain. The primary goals of this study are to review various XAI methods, their challenges, and related machine learning models in healthcare. The methods are discussed under six categories: Features-oriented methods, global methods, concept models, surrogate models, local pixel-based methods, and human-centric methods. Most importantly, the paper explores XAI role in healthcare problems to clarify its necessity in safety-critical applications. The paper intends to establish a comprehensive understanding of XAI-related applications in the healthcare field by reviewing the related experimental results. To facilitate future research for filling research gaps, the importance of XAI models from different viewpoints and their limitations are investigated. © 2023, CC BY-NC-ND.

关键词： Decision making

Work fluctuations due to partial thermalizations in two-level systems

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Quadeer, Maria Korzekwa, Kamil Tomamichel, Marco Centre for Quantum Software and Information School of Computer Science University of Technology Sydney NSW2007 Australia Faculty of Physics Astronomy and Applied Computer Science Jagiellonian University Kraków30-348 Poland Department of Electrical and Computer Engineering Centre for Quantum Technologies National University of Singapore Singapore119077 Singapore

We study work extraction processes mediated by finite-time interactions with an ambient bath—partial thermalizations—as continuous time Markov processes for two-level systems. Such a stochastic process results in fluctuations in the amount of work that can be extracted and is characterized by the rate at which the system parameters are driven in addition to the rate of thermalization with the bath. We analyze the distribution of work for the case where the energy gap of a two-level system is driven at a constant rate. We derive analytic expressions for average work and lower bound for the variance of work showing that such processes cannot be fluctuation-free in general. We also observe that an upper bound for the Monte Carlo estimate of the variance of work can be obtained using Jarzynski’s fluctuation-dissipation relation for systems initially in equilibrium. Finally, we analyse work extraction cycles by modifying the Carnot cycle, incorporating processes involving partial thermalizations and obtain efficiency at maximum power for such finite-time work extraction cycles under different sets of constraints. Copyright © 2021, The Authors. All rights reserved.

关键词： Carnot cycle

Construct Common Latent Representation for Multi-Modal Data Classification Through Domain Adaptation and Consensus Principle

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Nan, Fengtao Yang, Po Yang, Yun Chen, Zhaodong Jia, Qi Zhao, Zhong Xu, Lida School of Information Science and Engineering Yunnan University Kunming650500 China Department of Computer Science University of Sheffield SheffieldS1 4DP United Kingdom National Pilot School of Software Yunnan University Kunming650500 China Yunnan University China Department of Information Technology and Decision Sciences Old Dominion University Norfolk United States

Multi-modal representational learning aims to create models that can process and correlate information from multiple modalities. Different modalities can provide different complementary information, which improves the performance of multi-modal representation learning. However, most existing methods do not fully consider the data distribution of different modalities and require additional matching steps to produce the final feature representation. Second, they also do not fully exploit the consensus principle among different modalities. Based on the above two issues, we propose a novel method for learning a common latent representation through domain adaptation and consensus principle to solve these problems. Specifically, the informative and consistent representation is learned by minimizing the maximum mean discrepancy among the different modalities through domain adaptation, and the consensus principle is to maximize the consistency of the prediction results among the different modalities. Comprehensive experimental results show that the proposed method outperforms 10 traditional methods and 3 state-of-the-art methods on 10 challenging data sets. © 2022, The Authors. All rights reserved.

关键词： Modal analysis

Granular conditional entropy-based attribute reduction for partially labeled data with proxy labels

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Gao, Can Zhou, Jie Miao, Duoqian Yue, Xiaodong Wana, Jun College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518060 China Department of Computer Science and Technology Tongji University Shanghai201804 China School of Computer Engineering and Science Shanghai University Shanghai200444 China

Attribute reduction is one of the most important research topics in the theory of rough sets, and many rough sets-based attribute reduction methods have thus been presented. However, most of them are specifically designed for dealing with either labeled data or unlabeled data, while many real-world applications come in the form of partial supervision. In this paper, we propose a rough sets-based semi-supervised attribute reduction method for partially labeled data. Particularly, with the aid of prior class distribution information about data, we first develop a simple yet effective strategy to produce the proxy labels for unlabeled data. Then the concept of information granularity is integrated into the information-theoretic measure, based on which, a novel granular conditional entropy measure is proposed, and its monotonicity is proved in theory. Furthermore, a fast heuristic algorithm is provided to generate the optimal reduct of partially labeled data, which could accelerate the process of attribute reduction by removing irrelevant examples and excluding redundant attributes simultaneously. Extensive experiments conducted on UCI data sets demonstrate that the proposed semi-supervised attribute reduction method is promising and even compares favourably with the supervised methods on labeled data and unlabeled data with true labels in terms of classification performance. © 2021, CC BY-NC-SA.

关键词： Rough set theory

A High-Performance Hierarchical Snapshot Scheme for Hybrid Storage Systems

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2018年第1期27卷 76-85页

作者： YU Xiao TAN Yu'an ZHANG Changyou LIANG Chen Khaled AOURRA ZHENG Jun ZHANG Quanxin School of Computer Science and Technology Beijing Institute of Technology Department of Computer Science and Technology Shandong University of Technology Institute of Software Chinese Academy of Science

With the increased snapshot operations(creation or deletion), the performance of storage system will degrade severely whether in Copy on write(COW)or Redirect on write(ROW). This paper presents a novel snapshot scheme which could use of storage resources effectively and keep the historical data accurately for SSDHDD-hybrid storage system. SSD(source volume) stores all the active data and executes read/write operations without responding to snapshot operations. In contrast,HDDs(snapshot volume) store the writing data sequentially in the form of logs and are responsible for backup and snapshot operations. In this way, the hybrid storage system can get the high throughput power via SSD even the snapshot functions are active and HDDs can keep pace with the high Input/Output operations per second(IOPS)capability of the SSD. Logs are combined and kept in a separate image volume by the controller. So, fine-grained Recovery point objective(RPO) and good Recovery time objective(RTO) can be achieved. Experiment results show that the prototype can achieve excellent I/O performance while snapshot operations are active and consume a few extra storage space.

关键词： Hybrid disk array Snapshot Copy on write(COW) Redirect on write(ROW) Recovery point objective(RPO) Recovery time objective(RTO)

Super Resolution of Pathology Images with Hierarchical Feature Integration and Local Image Patterns

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Gao, Chaofei Xu, Feng Li, Lei Wang, Shuyang Ling, Ren Zhang, Xie Deng, Xi Zhou, Mengzhe Ling, Jin Department of Gastroenterology The Affiliated Lihuili Hospital of Ningbo University Zhejiang Ningbo315040 China Pathological Diagnosis Department Meikang Medicalsystem Laboratory. Ningbo Zhejiang 315105 China Department of Pathology School of Basic Medical Sciences Fudan University Xuhui District Shanghai200032 China Shanghai Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention Xuhui District Shanghai200032 China Department of Research and Development Shanghai Laizi Software Technology Co. LTD. Fengxian District Shanghai201499 China Ningbo Clinical Pathology Diagnosis Center Zhejiang Ningbo315020 China MOE Key Laboratory of Bioinformatics Bioinformatics Division BNRist Department of Automation Tsinghua University Beijing100084 China

The rapid development of artificial intelligence and digital pathology has demonstrated promise in precision diagnosis in clinical. However, pathology analysis at single cell and sub-cellular level is still limited by the image resolution as the high resolution image acquirement is expensive and time-consuming. Additionally, most super resolution models ignored the intrinsic characteristics of pathology images without reaching satisfactory performance. In this paper, we developed a hierarchical super resolution deep learning framework based on local image patterns (HLIP) to reconstruct super resolution images from the low resolution ones with arbitrary input size and magnification. HLIP integrates hierarchical feature maps and conducts reconstruction by considering the local image patterns of neighboring pixels. Benchmark analysis shows HLIP achieves the best performance on both the internal and external validation datasets and the generated super resolution images approximate high resolution ones, which have reached clinical quality. It is revealed that HLIP can be applied in multiple clinical scenes as a versatile tool to improve the performance of other models on the tasks including the gland segmentation, cell segmentation and Helicobacter pylori detection. Moreover, HLIP can be integrated directly into the existing computer-aided pathology diagnostic systems for real-time super resolution. © 2024, The Authors. All rights reserved.

关键词： Deep learning

A QoE Evaluation of an Augmented Reality Procedure Assistance Application

学校读者我要写书评

暂无评论

A QoE Evaluation of an Augmented Reality Procedure Assistanc...

International Workshop on Quality of Multimedia Experience, QoMEx

作者： Eoghan Hynes Ronan Flynn Brian Lee Niall Murray Department of Computer and Software Engineering Athlone Institute of Technology Athlone Co. Westmeath Ireland Software Research Institute Athlone Institute of Technology Athlone Co. Westmeath Ireland

ISBN: (数字)9781728159652

ISBN: (纸本)9781728159669

Augmented reality (AR) is a key technology to enhance worker effectiveness during increasing automation of repetitive jobs in the workplace. AR will achieve this by assisting the user to successfully perform complex and frequently changing procedures. The design of AR applications for these roles is critical to their acceptability and utility. User quality of experience (QoE) will inform these design decisions. A user arrives at a quality judgment upon post-experience reflection of their degree of delight or annoyance relating to the degree of fulfilment of pragmatic and hedonic needs and expectations of the medium under consideration. QoE researchers have largely depended upon post-experience subjective reports to determine the user's QoE. Subjective reports have been shown to be biased by primacy, recency and maxima of experience stimuli. Recent research involves the identification of implicit metrics that can be used to determine user QoE continuously during a multimedia experience. This work evaluates head rotation frequency as an objective metric of user QoE. The literature shows that emotion is expressed in the frequency of a person's head rotation around three axes of movement (pitch, yaw and roll). Low frequency head rotation has been shown to include expression of happy emotion while high frequency exclusively expresses anger emotion. These emotions are analogous to those reflected on by a user during the quality formation process. This demo paper analyses the amount of high frequency head rotation exhibited by the user upon task completion using an AR procedure assistance application or a paper-based control. An optimal Rubik's Cube solving AR application was used as a proof of concept for AR-based procedure assistance. Preliminary results showed that the AR environment yielded higher task success rates and significantly shorter task completion durations. The AR users exhibited significantly lower amplitudes of anger frequencies in their head rotations tha

关键词：

Achieving Privacy-Preserving Iris Identification Via El Gamal

学校读者我要写书评

暂无评论

computers, Materials & Continua 2019年第8期61卷 727-738页

作者： Yong Ding Lei Tian Bo Han Huiyong Wang Yujue Wang James Xi Zheng Guangxi Key Laboratory of Cryptography and Information Security School of Computer and Information SecurityGuilin University of Electronic TechnologyGuilin541004China School of Mathematics and Computer Science Guilin University of Electronic TechnologyGuilin541004China Software Engineering Department of Computing Macquarie UniversitySydneyNSW 2109 Australia

Currently,many biometric systems maintain the user’s biometrics and templates in plaintext format,which brings great privacy risk to uses’biometric *** are unique and almost unchangeable,which means it is a great concern for users on whether their biometric information would be *** address this issue,this paper proposes a confidential comparison algorithm for iris feature vectors with masks,and develops a privacy-preserving iris verification scheme based on the El Gamal encryption *** our scheme,the multiplicative homomorphism of encrypted features is used to compare of iris features and their mask ***,this paper improves the Hamming distance of iris features,which makes the similarity matching work better than existing *** results confirm the practicality of our proposed schemes in real world applications,that is,for the iris feature vectors and masks of 2048 bits,nearly 12 comparisons can be performed per second.

关键词： Homomorphic encryption template matching Hamming distance