检索结果-内蒙古大学图书馆

Vertical Federated Learning: Challenges, Methodologies and Experiments

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wei, Kang Li, Jun Ma, Chuan Ding, Ming Wei, Sha Wu, Fan Chen, Guihai Ranbaduge, Thilina School of Electrical and Optical Engineering Nanjing University of Science and Technology Nanjing210094 China Ministry of Education China Data61 CSIRO SydneyNSW2015 Australia Beijing100191 China The Shanghai Key Laboratory of Scalable Computing and Systems Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai200240 China The State Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China

Recently, federated learning (FL) has emerged as a promising distributed machine learning (ML) technology, owing to the advancing computational and sensing capacities of end-user devices, as well as the increasing concerns on users' privacy. As a special architecture in FL, vertical FL (VFL) is capable of constructing a hyper ML model by embracing sub-models from different types of clients. These sub-models are trained locally by vertically partitioned data with distinct attributes. Therefore, the design of VFL is fundamentally different from that of conventional FL, raising new and unique research issues. In this paper, we aim to discuss key challenges in VFL with effective solutions, and conduct experiments on real-life datasets to shed light on these issues. Specifically, we first propose a general framework on VFL, and highlight the key differences between VFL and conventional FL. Then, we discuss research challenges rooted in VFL systems from four aspects, i.e., security and privacy risks, expensive computation and communication costs, possible structural damage caused by model splitting, and system heterogeneity. Afterwards, we develop solutions to addressing the aforementioned challenges, and conduct extensive experiments to showcase the effectiveness of the proposed solutions. Copyright © 2022, The Authors. All rights reserved.

关键词： Federated learning

AI applications in robotics, diagnostic image analysis and precision medicine: Current limitations, future trends, guidelines on CAD systems for medicine

学校读者我要写书评

暂无评论

Informatics in Medicine Unlocked 2021年 24卷

作者： Habuza, Tetiana Navaz, Alramzana Nujum Hashim, Faiza Alnajjar, Fady Zaki, Nazar Serhani, Mohamed Adel Statsenko, Yauhen Department of Computer Science and Software Engineering College of Information Technology UAE University Al Ain United Arab Emirates Radiology Department College of Medicine and Health Sciences UAE University Al Ain United Arab Emirates

Background: AI in medicine has been recognized by both academia and industry in revolutionizing how healthcare services will be offered by providers and perceived by all stakeholders. Objectives: We aim to review recent tendencies in building AI applications for medicine and foster its further development by outlining obstacles. Sub-objectives: (1) to highlight AI techniques that we have identified as key areas of AI-related research in healthcare;(2) to offer guidelines on building reliable AI-based CAD-systems for medicine;and (3) to reveal open research questions, challenges, and directions for future research. Methods: To address the tasks, we performed a systematic review of the references on the main branches of AI applications for medical purposes. We focused primarily on limitations of the reviewed studies. Conclusions: This study provides a summary of AI-related research in healthcare, it discusses the challenges and proposes open research questions for further research. Robotics has taken huge leaps in improving the healthcare services in a variety of medical sectors, including oncology and surgical interventions. In addition, robots are now replacing human assistants as they learn to become more sociable and reliable. However, there are challenges that must still be addressed to enable the use of medical robots in diagnostics and interventions. AI for medical imaging eliminates subjectivity in a visual diagnostic procedure and allows for the combining of medical imaging with clinical data, lifestyle risks and demographics. Disadvantages of AI solutions for radiology include both a lack of transparency and dedication to narrowed diagnostic questions. Designing an optimal automatic classifier should incorporate both expert knowledge on a disease and state-of-the-art computer vision techniques. AI in precision medicine and oncology allows for risk stratification due to genomics aberrations discovered on molecular testing. To summarize, AI cannot substitute a

关键词： AI in medical imaging Deep learning Neural networks Oncology assistive technologies Precision medicine Robotics and health Survey

A Secure Federated Data-Driven Evolutionary Multi-objective Optimization Algorithm

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Liu, Qiqi Yan, Yuping Ligeti, Péter Jin, Yaochu Trustworthy and General AI Lab School of Engineering Westlake University China Faculty of Technology Bielefeld University Bielefeld33619 Germany Faculty of Informatics Department of Computeralgebra Eötvös Loránd University Hungary Smart Data Group E-Group ICT Software Zrt Hungary Department of Computer Science University of Surrey Surrey GuildfordGU2 7XH United Kingdom

Data-driven evolutionary algorithms usually aim to exploit the information behind a limited amount of data to perform optimization, which have proved to be successful in solving many complex real-world optimization problems. However, most data-driven evolutionary algorithms are centralized, causing privacy and security concerns. Existing federated Bayesian optimization algorithms and data-driven evolutionary algorithms mainly protect the raw data on each client. To address this issue, this paper proposes a secure federated data-driven evolutionary multiobjective optimization algorithm to protect both the raw data and the newly infilled solutions obtained by optimizing the acquisition function conducted on the server. We select the query points on a randomly selected client at each round of surrogate update by calculating the acquisition function values of the unobserved points on this client, thereby reducing the risk of leaking the information about the solution to be sampled. In addition, since the predicted objective values of each client may contain sensitive information, we mask the objective values with Diffie-Hellman-based noise, and then send only the masked objective values of other clients to the selected client via the server. Since the calculation of the acquisition function also requires both the predicted objective value and the uncertainty of the prediction, the predicted mean objective and uncertainty are normalized to reduce the influence of noise. Experimental results on a set of widely used multi-objective optimization benchmarks show that the proposed algorithm can protect privacy and enhance security with only negligible sacrifice in the performance of federated data-driven evolutionary optimization. Copyright © 2022, The Authors. All rights reserved.

关键词： Evolutionary algorithms

Recent development in disease diagnosis by information, communication and technology

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Urooj, Shabana Sharma, Astha Sinha, Chitransh Alrowais, Fadwa Department of Electrical Engineering College of Engineering Princess Nourah Bint Abdulrahman University Riyadh Saudi Arabia Gautam Buddha University Uttar Pradesh India Department of Electronics and Communication Engineering IILM-AHL- College of Engineering and Technology Greater Noida Uttar Pradesh India Department of Software Development Kalinga Institute of Information and Technology Bhubaneshwar Odisha India Department of Computer Sciences College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University Riyadh Saudi Arabia

The usage of Information, Communication and technology (ICT) in health sector has a great potential in improving the health of individuals and communities, disease detection, prevention and overall strengthening the healthcare systems, vital for development and poverty reduction. Large ICT establishments offer a variety of Artificial Intelligence (AI) based solutions;and their tenacities are inclusive of wearable therapeutic devices, healthcare management arrangements, extrapolative healthcare diagnostics, ailment prevention systems, detection and screening of diseases and automated tactics. In the field of healthcare related instrumentation, AI plays a prevalent role with the amalgamation of several technological progressions. This enables machines to sense, comprehend, act and learn to perform organisational and clinical healthcare functions as well as serves the research and training purposes. Additionally, it enables to accomplish the anticipated directorial and medicinal benefits. The major causes of life threats reported in literature are;heart and brain diseases. In this paper, an extensive review is presented exploring the evolving ICT technologies in machine learning and AI to help ICT enthusiasts to be able to catch up with the emerging trends in healthcare. Copyright © 2021, The Authors. All rights reserved.

关键词： Health care

Constructionist attempts at supporting the learning of computer programming: A survey 31

学校读者我要写书评

暂无评论

Constructionist attempts at supporting the learning of compu...

31st International Olympiad in Informatics, IOI 2019

作者： Lodi, Michael Malchiodi, Dario Monga, Mattia Morpurgo, Anna Spieler, Bernadette Department of Computer Science and Engineering Alma Mater Studiorum – Università di Bologna & INRIA Focus Italy Department of Computer Science Università degli Studi di Milano Italy Department for Software Technology Graz University of Technology Austria

Although programming is often seen as a key element of constructionist approaches, the research on learning to program through a constructionist strategy is somewhat limited, mostly focusing on how to bring the abstract and formal nature of programming languages into "concrete", possibly tangible objects, graspable even by children with limited abstraction power. We survey the literature in programming education and analyse some programming languages designed to help novices from a constructionist perspective. © 2019 IOI, Vilnius University.

关键词： Surveys

EER: Enterprise expert ranking using employee reputation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Mahmood, Saba Ghani, Anwar Daud, Ali Saqlain, Syed Muhammad Department of Computer Science and Software Engineering International Islamic University Islamabad Department of Computer Science Bahria University Islamabad Department of Information Systems and Technology College of Computer Science and Engineering University of Jeddah Jeddah Saudi Arabia

The emergence of online enterprises spread across continents, have given rise to the need of expert identification in this domain. Scenarios that includes the intention of the employer to find tacit expertise and knowledge of an employee that is not documented or self-disclosed has been addressed in this article. The existing reputation based approaches towards expertise ranking in enterprises utilize pagerank ,normal distribution and hidden markov model for expertise ranking. These models suffer issue of negative referral, collusion ,reputation inflation and dynamism. The authors have however proposed a Bayesian approach utilizing beta probability distribution based reputation model for employee ranking in enterprises. The experimental results reveal improved performance compared to previous techniques in terms of Precision and Mean Average Error (MAE) with almost 7% improvement in precision on average for the three data sets. The proposed technique is able to differentiate categories of interactions in dynamic context. The results reveal that the technique is independent of the rating pattern and density of data. Copyright © 2020, The Authors. All rights reserved.

关键词： Personnel

An HHL-based algorithm for computing hitting probabilities of quantum walks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Guan, Ji Wang, Qisheng Ying, Mingsheng State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences Beijing100190 China Department of Computer Science and Technology Tsinghua University Beijing100084 China Centre for Quantum Software and Information University of Technology SydneyNSW2007 Australia

We present a novel application of the HHL (Harrow-Hassidim-Lloyd) algorithm - a quantum algorithm solving systems of linear equations - in solving an open problem about quantum walks, namely computing hitting (or absorption) probabilities of a general (not only Hadamard) one-dimensional quantum walks with two absorbing boundaries. This is achieved by a simple observation that the problem of computing hitting probabilities of quantum walks can be reduced to inverting a matrix. Then a quantum algorithm with the HHL algorithm as a subroutine is developed for solving the problem, which is faster than the known classical algorithms by numerical experiments. Copyright © 2020, The Authors. All rights reserved.

关键词： Quantum theory

Ensemble Neural Networks with Random Weights for Classification Problems

学校读者我要写书评

暂无评论

Ensemble Neural Networks with Random Weights for Classificat...

作者： Ye Liu Weipeng Cao Zhong Ming Qiang Wang Jiyong Zhang Zhiwu Xu College of Computer Science and Software Engineering Shenzhen University Department of Computer Science and Engineering Southern University of Science and Technology School of Automation Hangzhou Dianzi University

To improve the prediction accuracy and stability of neural networks with random weights(NNRWs), we propose a novel ensemble NNRWs(E-NNRW) in this paper, which initializes its base learners by different distributions to improve their diversity. The final prediction results of the E-NNRW model are determined by these base learners through a voting mechanism, which minimizes the specific "blind zone" of a single learner, thus achieving higher prediction accuracy and better stability. Taking the random vector functional link network(RVFL), one of the most representative algorithms in NNRWs, as an example, we fully evaluate the performance of the proposed algorithm on nine benchmark classification *** experimental results fully demonstrate the effectiveness of our method.

关键词： Neural networks with random weights random vector functional link network ensemble learning randomized neural networks

A heterogeneous group CNN for image super-resolution

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Tian, Chunwei Zhang, Yanning Zuo, Wangmeng Lin, Chia-Wen Zhang, David Yuan, Yixuan The School of Software Northwestern Polytechnical University Shaanxi Xi’an710129 China The National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Shaanxi Xi’an710129 China The School of Computer Science Northwestern Polytechnical University The National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Shaanxi Xi’an710129 China The School of Computer Science and Technology Harbin Institute of Technology Heilongjiang Harbin150001 China The Department of Electrical Engineering The Institute of Communications Engineering National Tsing Hua University Hsinchu Taiwan Guangdong Shenzhen518172 China The Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China The Department of Electronic Engineering The Chinese University of Hong Kong Hong Kong

Convolutional neural networks (CNNs) have obtained remarkable performance via deep architectures. However, these CNNs often achieve poor robustness for image superresolution (SR) under complex scenes. In this paper, we present a heterogeneous group SR CNN (HGSRCNN) via leveraging structure information of different types to obtain a high-quality image. Specifically, each heterogeneous group block (HGB) of HGSRCNN uses a heterogeneous architecture containing a symmetric group convolutional block and a complementary convolutional block in a parallel way to enhance internal and external relations of different channels for facilitating richer low-frequency structure information of different types. To prevent appearance of obtained redundant features, a refinement block with signal enhancements in a serial way is designed to filter useless information. To prevent loss of original information, a multi-level enhancement mechanism guides a CNN to achieve a symmetric architecture for promoting expressive ability of HGSRCNN. Besides, a parallel up-sampling mechanism is developed to train a blind SR model. Extensive experiments illustrate that the proposed HGSRCNN has obtained excellent SR performance in terms of both quantitative and qualitative analysis. Codes can be accessed at https://***/hellloxiaotian/HGSRCNN. © 2022, CC BY-NC-SA.

关键词： Convolutional neural networks