检索结果-内蒙古大学图书馆

Machine-Learning Enabled Multidimensional Data Utilization Through Multi-Resonance Architecture: A Pathway to Enhanced Accuracy in Biosensing

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Aalizadeh, Majid Afshar, Morteza Azmoudeh Fan, Xudong Department of Biomedical Engineering University of Michigan Ann ArborMI48109 United States Department of Electrical Engineering and Computer Science University of Michigan Ann ArborMI48109 United States University of Michigan Ann ArborMI48109 United States Max Harry Weil Institute for Critical Care Research and Innovation University of Michigan Ann ArborMI48109 United States Informatics Institute Istanbul Technical University Istanbul34485 Turkey

A novel framework is proposed that combines multi-resonance biosensors with machine learning (ML) to significantly enhance the accuracy of parameter prediction in biosensing. Unlike traditional single-resonance systems, which are limited to one-dimensional datasets, this approach leverages multi-dimensional data generated by a custom-designed nanostructure—a periodic array of silicon nanorods with a triangular cross-section over an aluminum reflector. High bulk sensitivity values are achieved for this multi-resonant structure, with certain resonant peaks reaching up to 1706 nm/RIU. The field analysis reveals Mie resonances as the physical reason behind the peaks. The predictive power of multiple resonant peaks from transverse magnetic (TM) and transverse electric (TE) polarizations is evaluated using Ridge Regression modeling. Systematic analysis reveals that incorporating multiple resonances yields up to three orders of magnitude improvement in refractive index detection precision compared to single-peak analyses. This precision enhancement is achieved without modifications to the biosensor hardware, highlighting the potential of data-centric strategies in biosensing. The findings establish a new paradigm in biosensing, demonstrating that the synergy between multi-resonance data acquisition and ML-based analysis can significantly enhance detection accuracy. This study provides a scalable pathway for advancing high-precision biosensing technologies. Copyright © 2024, The Authors. All rights reserved.

关键词： Nanorods

A Novel Air Traffic Complexity Assessment Method Based on Deep Metric Learning

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Zhang, Linghui Chen, Haiyan Wu, Lingxiao Liu, Li Xue, Dabin College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China State Key Laboratory of Air Traffic Management System Nanjing China Department of Aeronautical and Aviation Engineering The Hong Kong Polytechnic University Hong Kong Department of Land Surveying and Geo-Informatics The Hong Kong Polytechnic University Hong Kong

Air traffic complexity is related to the workload of air traffic control officers and pilots, subsequently leading to potential effects on flight safety and efficiency. However, how to assess air traffic complexity accurately is still a question in the concept of Air Traffic Management (ATM). In this study, a model for air traffic complexity assessment is proposed based on deep metric learning. Specifically, the air traffic data collected by surveillance radars are adopted to generate the air traffic image set, from which air traffic features are extracted based on the Convolutional Neural Networks (CNN) model. After that, the Deep Metric learning method based on Asymmetric distance, Aggregation loss, and Edge hard loss, called AAEDM, is applied to address the problem of class imbalance in air traffic images. Finally, the traffic complexity assessment model is proposed based on AAEDM. The proposed model is validated through comprehensive experimentation using two established standard datasets. The results of these experiments indicate the outstanding proficiency of AAEDM, particularly in scenarios involving unbalanced data. The proposed model can extract deeper features of air traffic than traditional machine learning methods, outperforming other models for air traffic complexity assessment. This study can help assess air traffic complexity and improve the robustness of the ATM system. © 2023, The Authors. All rights reserved.

关键词： Air traffic control

A Framework for Tele-rehabilitation Gaming System

学校读者我要写书评

暂无评论

A Framework for Tele-rehabilitation Gaming System

Applied Computational Intelligence in Information Systems (ACIIS), International Conference on

作者： Ahmed M. S. Elaklouk Ratna Zuarni Ramli Ibrahim Edris Nena Padilla-Valdez Adel Al Jumaily School of Computing and Informatics Universiti Teknologi Brunei Bandar Seri Begawan Brunei Darussalam College of Computing Informatics and Mathematics Universiti Teknologi MARA Negeri Seremban Malaysia Centre for Communication Universiti Teknologi Brunei Bandar Seri Begawan Brunei Darussalam School of Computing University of Technology Brunei Department of Computer Science & Software Engineering University of Western Australia and ENSTA Bretagne University France

Brain damage resulting from stroke and traumatic brain injury is the primary cause of complicated disability and one of the major causes of mortality in most countries, burdening healthcare systems owing to the growth in the number of brain injury victims. Rehabilitation is required to deal with the brain damage. With fewer therapists available, it is very challenging to manage patients and even motivate them to participate actively in their rehabilitation. Patients have frequently complained that many traditional rehabilitation systems are monotonous and uninteresting. The study’s objective is to propose a telerehabilitation gaming system framework that can serve as guidance for developing an affordable rehabilitation gaming system that can motivate and engage patients and increase rehabilitation effectiveness. The research methodology is based on synthesizing related research and currently available technologies. The proposed framework includes the therapist and patient attached with vital constructs: tailoring tools, instructional content, game characteristics, tailored games, and performance. It is an internet-based communication structure that connects the Rehabilitation Gaming System to the hospital and other care provider networks. The proposed framework is evaluated by a panel of five experts. The results reveal that all experts believe that the proposed framework can serve as a useful guide for creating gaming interventions that can boost the patient’s motivation and engagement and increase the effectiveness of rehabilitation.

关键词：

Generative Adversarial Networks for Heterogeneous Unsupervised Domain Adaptation Detection

学校读者我要写书评

暂无评论

Generative Adversarial Networks for Heterogeneous Unsupervis...

IEEE International Conference on Intelligent engineering Systems (INES)

作者： Homayoun Safarpour Motealegh Mahalegi Amirfarhad Farhadi György Molnár Enikő Nagy Institute of Mechatronics and Vehicle Engineering Óbuda University Budapest Hungary Department of Computer Engineering Science and Research Branch Islamic Azad University Tehran Iran Kandó Kálmán Faculty of Electrical Engineering Óbuda University Budapest Hungary János Csere Apáczai Faculty Széchenyi University Győr Hungary Antal Bejczy Center for Intelligent Robotics John von Neumann Faculty of Informatics Óbuda University Budapest Hungary

ISBN: (数字)9798350367591

ISBN: (纸本)9798350367607

Domain adaptation, a subset of transfer learning, involves generating examples from two related but differently distributed source and target domains. This paper introduces an incremental adversarial learning method for unsupervised domain adaptation, where the source data is labeled, and the target data remains unlabeled. We employ a discriminative adversarial strategy as the primary technique for unsupervised domain adaptation to minimize the significant distributional variance between the source and target domains. Traditional domain adaptation methods typically train the source domain to align with a fixed target domain distribution. However, in many real-world scenarios, the target domain's distribution continuously evolves, leading to a lack of generalizability in the trained model for subsequent domains. To tackle this issue, we suggest using a continuous sequential domain adaptation method. This involves including instances from several related target domains in the training process in a sequential and continuous manner. This process ensures that all receiving domains are correctly aligned with their respective sources, and to maintain sequence continuity, the source domain is incrementally updated by incorporating the most reliable data from the current target domain. This enhancement aims to improve the network's training for forthcoming domain data. Leveraging the effectiveness of deep neural networks, our method utilizes deep GANs for domain adaptation. The approach enhances domain matching by augmenting the source with selected target data and introduces a novel strategy for continuous sequential domain matching. This involves a progressive update of source domain samples with reliable data from the current target domain, facilitating seamless adaptation to successive domain shifts. The proposed method demonstrated a high classification accuracy, achieving an impressive 99% and 95% in different settings. Furthermore, it outperforms previous studies rega

关键词： Training Adaptation models Machine learning algorithms Accuracy Transfer learning Clustering algorithms Sensitivity and specificity

Evaluation of augmentation methods in classifying autism spectrum disorders from fMRI data with 3D convolutional neural networks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Jönemo, Johan Abramian, David Eklund, Anders Division of Medical Informatics Department of Biomedical Engineering Division of Statistics & Machine Learning Department of Computer and Information Science Linköping University Sweden

Classifying subjects as healthy or diseased using neuroimaging data has gained a lot of attention during the last 10 years. Here we apply deep learning to derivatives from resting state fMRI data, and investigate how different 3D augmentation techniques affect the test accuracy. Specifically, we use resting state derivatives from 1,112 subjects in ABIDE preprocessed to train a 3D convolutional neural network (CNN) to perform the classification. Our results show that augmentation only provide minor improvements to the test accuracy. © 2021, CC BY.

关键词： Convolutional neural networks

Efficiently Tackling Million-Dimensional Multiobjective Problems: A Direction Sampling and Fine-Tuning Approach

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hong, Haokai Jiang, Min Lin, Qiuzhen Tan, Kay Chen Department of Artificial Intelligence Key Laboratory of Digital Protection and Intelligent Processing of Intangible Cultural Heritage of Fujian and Taiwan Ministry of Culture and Tourism School of Informatics Xiamen University Fujian361005 China College of Computer Science and Software Engineering Shenzhen University China Department of Computing Hong Kong Polytechnic University Hong Kong

We define very large-scale multiobjective optimization problems as optimizing multiple objectives (VLSMOPs) with more than 100,000 decision variables. These problems hold substantial significance, given the ubiquity of real-world scenarios necessitating the optimization of hundreds of thousands, if not millions, of variables. However, the larger dimension in VLSMOPs intensifies the curse of dimensionality and poses significant challenges for existing large-scale evolutionary multiobjective algorithms, rendering them more difficult to solve within the constraints of practical computing resources. To overcome this issue, we propose a novel approach called the very large-scale multiobjective optimization framework (VMOF). The method efficiently samples general yet suitable evolutionary directions in the very large-scale space and subsequently fine-tunes these directions to locate the Pareto-optimal solutions. To sample the most suitable evolutionary directions for different solutions, Thompson sampling is adopted for its effectiveness in recommending from a very large number of items within limited historical evaluations. Furthermore, a technique is designed for fine-tuning directions specific to tracking Pareto-optimal solutions. To understand the designed framework, we present our analysis of the framework and then evaluate VMOF using widely recognized benchmarks and real-world problems spanning dimensions from 100 to 1,000,000. Experimental results demonstrate that our method exhibits superior performance not only on LSMOPs but also on VLSMOPs when compared to existing algorithms. Copyright © 2023, The Authors. All rights reserved.

关键词： Multiobjective optimization

Quantum Generative Diffusion Model: A Fully Quantum-Mechanical Model for Generating Quantum State Ensemble

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Chuangtao Zhao, Qinglin Zhou, MengChu He, Zhimin Sun, Zhili Situ, Haozhen Faculty of Innovation Engineering Macau University of Science and Technology Macao999078 China Department of Electrical and Computer Engineering New Jersey Institute of Technology NewarkNJ07102 United States School of Electronic and Information Engineering Foshan University Foshan528000 China Institute of Communication Systems University of Surrey GuildfordGU2 7XH United Kingdom College of Mathematics and Informatics South China Agricultural University Guangzhou510642 China

Classical diffusion models have shown superior generative results. Exploring them in the quantum domain can advance the field of quantum generative learning. This work introduces Quantum Generative Diffusion Model (QGDM) as their simple and elegant quantum counterpart. Through a non-unitary forward process, any target quantum state can be transformed into a completely mixed state that has the highest entropy and maximum uncertainty about the system. A trainable backward process is used to recover the former from the latter. The design requirements for its backward process includes non-unitarity and small parameter count. We introduce partial trace operations to enforce non-unitary and reduce the number of trainable parameters by using a parameter-sharing strategy and incorporating temporal information as an input in the backward process. We present QGDM’s resource-efficient version to reduce auxiliary qubits while preserving generative capabilities. QGDM exhibits faster convergence than Quantum Generative Adversarial Network (QGAN) because its adopted convex-based optimization can result in better convergence. The results of comparing it with QGAN demonstrate its effectiveness in generating both pure and mixed quantum states. It can achieve 53.02% higher fidelity in mixed-state generation than QGAN. The results highlight its great potential to tackle challenging quantum generation tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

The OPS-SAT benchmark for detecting anomalies in satellite telemetry

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ruszczak, Bogdan Kotowski, Krzysztof Evans, David Nalepa, Jakub Faculty of Electrical Engineering Automatic Control and Informatics Department of Informatics Opole University of Technology Prószkowska Str. 76 Opole45-758 Poland Faculty of Automatic Control Electronics and Computer Science Department of Algorithmics and Software Silesian University of Technology Akademicka Str. 16 Gliwice44-100 Poland KP Labs Bojkowska Str. 37J Gliwice44-100 Poland European Space Agency ESOC Robert-Bosch-Str. 5 Darmstadt64293 Germany

Detecting anomalous events in satellite telemetry is a critical task in space operations. This task, however, is extremely time-consuming, error-prone and human dependent, thus automated data-driven anomaly detection algorithms have been emerging at a steady pace. However, there are no publicly available datasets of real satellite telemetry accompanied with the ground-truth annotations that could be used to train and verify anomaly detection supervised models. In this article, we address this research gap and introduce the AI-ready benchmark dataset (OPSSAT-AD) containing the telemetry data acquired on board OPS-SAT—a CubeSat mission which has been operated by the European Space Agency which has come to an end during the night of 22–23 May 2024 (CEST). The dataset is accompanied with the baseline results obtained using 30 supervised and unsupervised classic and deep machine learning algorithms for anomaly detection. They were trained and validated using the training-test dataset split introduced in this work, and we present a suggested set of quality metrics which should be always calculated to confront the new algorithms for anomaly detection while exploiting OPSSAT-AD. We believe that this work may become an important step toward building a fair, reproducible and objective validation procedure that can be used to quantify the capabilities of the emerging anomaly detection techniques in an unbiased and fully transparent way. © 2024, CC BY.

关键词： Anomaly detection

The equivalence between correctability of deletions and insertions of separable states in quantum codes

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Shibayama, Taro Ouyang, Yingkai Department of Mathematics and Informatics Graduate School of Science Chiba University Japan Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore

In this paper, we prove the equivalence of inserting separable quan-tum states and deletions. Hence any quantum code that corrects dele-tions automatically corrects separable insertions. First, we describe the quantum insertion/deletion error using the Kraus operators. Next, we develop an algebra for commuting Kraus operators corresponding to insertions and deletions. Using this algebra, we prove the equiv-alence between quantum insertion codes and quantum deletion codes using the Knill-Laflamme conditions. Copyright © 2021, The Authors. All rights reserved.

关键词： Algebra