检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Guo, Wenzhe Fouda, Mohammed E. Eltawil, Ahmed M. Salama, Khaled N. Department of Electrical and Computer Engineering King Abdullah University of Science and Technology Thuwal23955 Saudi Arabia Center for Embedded & Cyber-physical Systems University of California IrvineCA92612 United States

Empowered by the backpropagation (BP) algorithm, deep neural networks have dominated the race in solving various cognitive tasks. The restricted training pattern in the standard BP requires end-to-end error propagation, causing large memory cost and prohibiting model parallelization. Existing local training methods aim to resolve the training obstacle by completely cutting off the backward path between modules and isolating their gradients to reduce memory cost and accelerate the training process. These methods prevent errors from flowing between modules and hence information exchange, resulting in inferior performance. This work proposes a novel local training algorithm, BackLink, which introduces inter-module backward dependency and allows errors to flow between modules. The algorithm facilitates information to flow backward along with the network. To preserve the computational advantage of local training, BackLink restricts the error propagation length within the module. Extensive experiments performed in various deep convolutional neural networks demonstrate that our method consistently improves the classification performance of local training algorithms over other methods. For example, in ResNet32 with 16 local modules, our method surpasses the conventional greedy local training method by 4.00% and a recent work by 1.83% in accuracy on CIFAR10, respectively. Analysis of computational costs reveals that small overheads are incurred in GPU memory costs and runtime on multiple GPUs. Our method can lead up to a 79% reduction in memory cost and 52% in simulation runtime in ResNet110 compared to the standard BP. Therefore, our method could create new opportunities for improving training algorithms towards better efficiency and biological plausibility. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Ransomware Attacks Detection Methodology to Protect IoT-Enabled Critical Infrastructures

Ransomware Attacks Detection Methodology to Protect IoT-Enab...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Amit Kumar Mishra Mohammad S. Obaidat Harshit Bhajpai Pranjal Trivedi Mohammad Wazid D. P. Singh Joel J. P. C. Rodrigues Balqies Sadoun Department of Computer Science and Engineering Graphic Era Hill University Dehradun India Department of Computer Science and Engineering Graphic Era Deemed to be University Dehradun India King Abdullah II School of Information Technology The University of Jordan Amman Jordan University of Science and Technology Beijing Beijing China College of Computer Science and Technology China University of Petroleum (East China) Qingdao China Instituto de Telecomunições Covilhã Portugal College of Engineering Al- Balqa' Applied University Al-Salt Jordan

Critical infrastructure is a collection of physical and cyber systems, which are essentially required to support the day-to-day operations of our daily life. In the critical infrastructure, the computing systems (i.e., Internet of Things (loT) devices) communicate through the Internet. Therefore, most of the time, critical infrastructures are targeted by hackers by launching some cyber-attack, i.e., ransomware. Hence we require some security mechanisms to protect the data and systems of critical infrastructures. This paper proposes a scheme for the detection, analysis and mitigation of ransomware attacks to protect Internet of Things (loT)-enabled critical infrastructure (in short, RADM-ICI). We also practically demonstrated RADM-ICI and computed essential performance parameters, i.e., accuracy and F1-score under different machine learning models. The conducted security analysis of RADM-ICI proved its excellent security for the ransom ware attacks. During the performance comparison of the proposed RADM-ICI and other similar competing existing schemes, it has been observed that the proposed RADM-ICI achieved better accuracy than the other existing competing schemes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

From Non-Expert to Expert: Recurrent Refined Learning for Medical Image Segmentation

From Non-Expert to Expert: Recurrent Refined Learning for Me...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Ruohui Jiang Changtai Li Xiaojuan Ban Shihua Yin Chao Yao Yu Guo Mohammad S. Obaidat Beijing Advanced Innovation Center for Materials Genome Engineering University of Science and Technology Beijing Beijing China School of Intelligence Science and Technology University of Science and Technology Beijing Beijing China Institute of Materials Intelligent Technology Liaoning Academy of Materials Shenyang China Second Affiliated Hospital of Guangxi Medical University Nanning China School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China King Abdullah II School of Information Technology University of Jordan Amman Jordan Amity University Noida UP India

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Precise lesion segmentation is essential in computer-aided diagnosis and treatment. Prevalent deep-learning-based approaches need high-quality annotations to achieve satisfying performance. However, blurring effects and infiltration hamper the accurate delineation of lesions. A common solution is to divide the annotation process into initial annotation by non-specialized personnel and subsequent modification by expert physicians, where the latent correction patterns and non-expert labels are seldom utilized effectively. To explore their helpfulness, we propose ReReNet, a recurrent refined network for lesion segmentation that learns from non-expert to expert. It achieves progressively refined segmentation results through multiple iterations with tailored discrepancy-aware supervision. During training, as the iterative process perceives discrepancies between refining and expert labels, the model gradually grasps the knowledge of turning the barely correct into the clinically accurate. We validate ReReNet’s capability on three medical image segmentation (MIS) datasets, including magnetic resonance (MR) and computed tomography (CT) modalities. Comparison results indicate that the proposed approach achieves superior performance by introducing the designed recurrent mechanism and outperforms mainstream methods, demonstrating the effectiveness of mining hidden correction patterns by utilizing non-expert information.

关键词： Training Image segmentation Accuracy Annotations Computed tomography Refining computer architecture Turning Lesions Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

FALCON: Scalable Reasoning over Inconsistent ALC Ontologies

arXiv

引用

arXiv 2022年

作者： Hinnerichs, Tilman Tang, Zhenwei Peng, Xi Zhang, Xiangliang Hoehndorf, Robert Computational Bioscience Research Center Computer Electrical and Mathematical Sciences & Engineering Division King Abdullah University of Science and Technology Saudi Arabia University of Toronto Canada Delft University of Technology Netherlands University of Notre Dame United States

Ontologies are one of the richest sources of knowledge. Real-world ontologies often contain thousands of axioms and are often human-made. Hence, they may contain inconsistency and incomplete information which may impair classical reasoners to compute entailments that are considered as useful. To overcome these two challenges, we propose FALCON, a Fuzzy Ontology Neural reasoner to approximate reasoning over ALC ontologies. We provide an approximate technique for the model generation step in classical ALC reasoners. Our approximation is not guaranteed to construct exact logical models, but can approximate arbitrary models, which is notably faster for some large ontologies. Moreover, by sampling multiple approximate logical models, our technique supports approximate entailment also over inconsistent ontologies. Theoretical results show that more models generated lead to closer, i.e., faithful approximation of entailment over ALC entailments. Experimental results show that FALCON enables approximate reasoning and reasoning in the presence of inconsistency. Our experiments further demonstrate how ontologies can improve knowledge base completion in biomedicine by incorporating knowledge expressed in ALC. Copyright © 2022, The Authors. All rights reserved.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Multi-index Sequential Monte Carlo ratio estimators for Bayesian Inverse problems

arXiv

引用

arXiv 2022年

作者： Law, Kody J.H. Walton, Neil Yang, Shangda Jasra, Ajay School of Mathematics University of Manchester ManchesterM13 9PL United Kingdom Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia

We consider the problem of estimating expectations with respect to a target distribution with an unknown normalizing constant, and where even the unnormalized target needs to be approximated at finite resolution. This setting is ubiquitous across science and engineering applications, for example in the context of Bayesian inference where a physics-based model governed by an intractable partial differential equation (PDE) appears in the likelihood. A multi-index Sequential Monte Carlo (MISMC) method is used to construct ratio estimators which provably enjoy the complexity improvements of multi-index Monte Carlo (MIMC) as well as the efficiency of Sequential Monte Carlo (SMC) for inference. In particular, the proposed method provably achieves the canonical complexity of MSE−1, while single level methods require MSE−ξ for ξ > 1. This is illustrated on examples of Bayesian inverse problems with an elliptic PDE forward model in 1 and 2 spatial dimensions, where ξ = 5/4 and ξ = 3/2, respectively. It is also illustrated on more challenging log Gaussian process models, where single level complexity is approximately ξ = 9/4 and multilevel Monte Carlo (or MIMC with an inappropriate index set) gives ξ = 5/4 + ω, for any ω > 0, whereas our method is again canonical. We also provide novel theoretical verification of the product-form convergence results which MIMC requires for Gaussian processes built in spaces of mixed regularity defined in the spectral domain, which facilitates acceleration with fast Fourier transform methods via a cumulant embedding strategy, and may be of independent interest in the context of spatial statistics and machine learning. Copyright © 2022, The Authors. All rights reserved.

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

Novel Dopamine-Derived Carbon Modified Fe-Based Metal-Organic Framework Enhanced Aptasensor Rendering Ultrasensitive Electrochemical Detection of Oxytetracycline

SSRN

引用

SSRN 2024年

作者： Song, Jialing Fan, Xiaoxuan Ee, Liang Ying Lin, Xuanhao Li, Sam Fong Yau Huang, Manhong Singapore Donghua University China Shanghai University College of Science Department of Chemistry International Joint Laboratory Catalyst Chemistry Shanghai200444 China College of Environmental Science and Engineering State Environmental Protection Engineering Center for Pollution Treatment and Control in Textile Industry Donghua University Shanghai201620 China National University of Singapore Department of Chemistry 3 Science Drive 3 Singapore117543 Singapore Advanced Membranes and Porous Materials Center Physical Science and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia Chemical Engineering Program Physical Science and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia National University of Singapore Environmental Research Institute T Lab Building 5A Engineering Drive 1 Singapore117411 Singapore

In the present work, an aptasensor with dual binding sites based on Au NPs modified with self-polymerized polydopamine (PDA) and Fe-MOF is successfully constructed. In terms of standard electron transfer, the cyclic voltammetry (CV) analysis showcases a rate constant (ks) of 96.0 s-1, signifying an exceptional electron transfer rate of oxytetracycline at the sensor interface of Apt/Au@PDA@NH2-MIL-101(Fe)/GCE. Further deepening the investigation into electrochemical kinetics using chronoamperometry (CA), we investigated the electrocatalytic parameters (Kcat) of Au@PDA@NH2-MIL-101(Fe) in relation to oxytetracycline. The results divulged an impressive average Kcat value of 3.06×105 M-1 S-1. This evidence robustly suggests the superior electrocatalytic activity of Au@PDA@NH2-MIL-101(Fe) when it comes to oxytetracycline oxidation. A custom-designed portable rapid detection platform was developed, achieving detection limits of 6.88 nM under static water flow conditions and 5.56 nM under dynamic conditions. Remarkably, the self-polymerization of PDA on Au NPs not only refines their size but also boosts their interaction with biomolecules, thereby promising excellent biocompatibility. Moreover, the abundance of -COOH groups and unsaturated Fe3+ sites on the NH2-MIL-101(Fe) surface furnishes a considerable number of grafting sites for the aptamer. More importantly, the redox interplay between Au@PDA and NH2-MIL-101(Fe) accelerates electron movement, thus amplifying the electrochemical signal and the detection sensitivity for oxytetracycline. © 2024, The Authors. All rights reserved.

关键词： Rate constants

来源：评论

学校读者我要写书评

暂无评论

ON STABILITY AND GENERALIZATION OF BILEVEL OPTIMIZATION PROBLEM

arXiv

引用

arXiv 2022年

作者： Ding, Meng Lei, Mingxi Lei, Yunwen Wang, Di Xu, Jinhui Department of Computer Science and Engineering The State University of New York at Buffalo United States School of Computer Science University of Birmingham United Kingdom Division of CEMSE King Abdullah University of Science and Technology Saudi Arabia

(Stochastic) bilevel optimization is a frequently encountered problem in machine learning with a wide range of applications such as meta-learning, hyper-parameter optimization, and reinforcement learning. Most of the existing studies on this problem only focused on analyzing the convergence or improving the convergence rate, while little effort has been devoted to understanding its generalization behaviors. In this paper, we conduct a thorough analysis on the generalization of first-order (gradient-based) methods for the bilevel optimization problem. We first establish a fundamental connection between algorithmic stability and generalization error in different forms and give a high probability generalization bound which improves the previous best one from O(√n) to O(log n), where n is the sample size. We then provide the first stability bounds for the general case where both inner and outer level parameters are subject to continuous update, while existing work allows only the outer level parameter to be updated. Our analysis can be applied in various standard settings such as strongly-convex-strongly-convex (SC-SC), convex-convex (C-C), and nonconvex-nonconvex (NC-NC). Our analysis for the NC-NC setting can also be extended to a particular nonconvex-strongly-convex (NC-SC) setting that is commonly encountered in practice. Finally, we corroborate our theoretical analysis and demonstrate how iterations can affect the generalization error by experiments on meta-learning and hyper-parameter optimization. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

CONVERGENCE SPEED AND APPROXIMATION ACCURACY OF NUMERICAL MCMC

arXiv

引用

arXiv 2022年

作者： Cui, Tiangang Dong, Jing Jasra, Ajay Tong, Xin T. School of Mathematics Monash University Australia Graduate School of Business Columbia University United States Applied Mathematics and Computational Science Program Computer Electrical Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Saudi Arabia Department of Mathematics National University of Singapore Singapore

When implementing Markov Chain Monte Carlo (MCMC) algorithms, perturbation caused by numerical errors is sometimes inevitable. This paper studies how perturbation of MCMC affects the convergence speed and Monte Carlo estimation accuracy. Our results show that when the original Markov chain converges to stationarity fast enough and the perturbed transition kernel is a good approximation to the original transition kernel, the corresponding perturbed sampler has similar convergence speed and high approximation accuracy as well. We discuss two different analysis frameworks: ergodicity and spectral gap, both are widely used in the literature. Our results can be easily extended to obtain non-asymptotic error bounds for MCMC estimators. We also demonstrate how to apply our convergence and approximation results to the analysis of specific sampling algorithms, including Random walk Metropolis and Metropolis adjusted Langevin algorithm with perturbed target densities, and parallel tempering Monte Carlo with perturbed densities. Finally we present some simple numerical examples to verify our theoretical claims. Copyright © 2022, The Authors. All rights reserved.

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

Assessing the mutagenic potential of methyl phenlactonoate 3 and Nijmegen-1 in bacterial reverse mutation assays

引用

Heliyon 2024年第23期10卷 e40526页

作者： Jamil, Muhammad Wang, Jian You Patel, Kinjal A. Nagane, Rajendra M. Patel, Manish V. Totre, Jalindar Bhoge, Satish E. Al-Babili, Salim The BioActives Lab BESE King Abdullah University of Science and Technology Thuwal 23955-6900 Saudi Arabia Jai Research Foundation Department of Toxicology Valvada dist Gujarat Valsad 396 105 India UPL House Express Highway Bandra-East Maharashtra Mumbai 400 051 India Plant Science Program Biological and Environmental Science and Engineering Division King Abdullah University of Science and Technology (KAUST) Thuwal 23955-6900 Saudi Arabia

The use of strigolactone (SL) analogs as suicidal germination agents to control the seed banks of the parasitic weed Striga hermonthica has gained interest for field applications in recent years. However, concerns about the environmental safety of these SL analogs remain. In this study, we evaluated the mutagenic potential of two selected SL analogs, Methyl Phenlactonoate 3 (MP3) and Nijmegen-1, across concentrations ranging from 1.5 to 5000 μg per plate. We conducted this assessment using five histidine-deficient mutant tester strains of Salmonella typhimurium. After incubating the SL analogs with the tester strains, we observed no significant increase in the number of revertants, with and without the S9 mix, compared to both negative and positive laboratory controls. These results suggest that MP3 and Nijmegen-1 are non-mutagenic according to the bacterial reverse mutation test, supporting their potential as environmentally safe agents for managing Striga populations. © 2024 The Authors

关键词： Methyl phenlactonoates Mutagenicity Nijmegen-1 Reverse mutation Striga hermonthica Strigolactone analogs

来源：评论

学校读者我要写书评

暂无评论

A Time-domain Carrier Generation Rate Model for Optoelectronic Device Simulations

A Time-domain Carrier Generation Rate Model for Optoelectron...

引用

IEEE International Symposium on Antennas and Propagation

作者： Liang Chen Hakan Bagci Division of Computer Electrical and Mathematical Science and Engineering (CEMSE) King Abdullah University of Science and Technology (KAUST) Thuwal Saudi Arabia

ISBN: (纸本)9781728146713

The nonlinear interactions between electromagnetic fields and carriers in optoelectronic devices call for a multiphysics simulation approach to be carried out in the time domain. The widely used Poynting vector-based carrier generation rate model leads to an unphysical divergent solution when strong low-frequency fields are generated by moving carriers. This work proposes a material absorption-based carrier generation rate model for time-domain simulation of optoelectronic devices. The optoelectronic material is described by a Lorentz dispersion model with poles in the optical frequency range and negligible absorption at lower frequencies. The instantaneous optical absorption, which is computed from the polarization current density, is used to obtain the generation rate. Numerical examples show that the proposed method is more accurate than the Poynting vector-based model and is stable even when the generated lower-frequency fields are strong.

关键词： Optical polarization Absorption Computational modeling Meetings Optoelectronic devices Optical materials Numerical models

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：