检索结果-内蒙古大学图书馆

Web service location-allocation using discrete NSGA-II with matrix based genetic operations and a repair mechanism

Journal of Ambient Intelligence and Humanized Computing 2023年第10期14卷 14163-14187页

作者： Verma, Shanu Pant, Millie Snasel, Vaclav Department of Applied Mathematics and Scientific Computing Indian Institute of Technology Roorkee Roorkee247667 India Centre for Artificial Intelligence and Data Science Indian Institute of Technology Roorkee Roorkee247667 India Department of Computer Science VSB-Technical University of Ostrava Ostrava70800 Czech Republic

Web Service Location-Allocation Problem (WSLAP) is an important and challenging NP-hard combinatorial optimization problem where multiple objectives affecting the quality of service (QoS) are to be optimized simultaneously. Most of the existing research articles mainly focus on selecting web service locations that minimize two QoS: total cost and total response time. However, another QoS factor throughput also needs attention when selecting sites for allocating web services. Therefore, we have constructed a new WSLAP model that minimizes the total cost and the total negative quality factor. This negative quality factor considers throughput along with latency to strengthen the QoS of the located web services. Further, three NSGA-II variants, NSGA-II.V1, NSGA-II.V2, and NSGA-II.V3, with a novel matrix-based chromosome representation, are developed to deal with the proposed bi-objective WSLAP model. These algorithms involve matrix-based genetic operations for better exploration and a repair mechanism for tackling infeasibility. The proposed model and algorithms are validated on 24 test instances, out of which 14 are existing, and 10 are newly created with increased web services, user centres and service locations. The experimental results claim that the proposed WSLAP model succeeded in improving the throughput with a comparatively lower total cost. Further, the proposed algorithms outperformed the compared algorithms in terms of hypervolume, inverted generational distance and computational time on 89% of instances. © 2023, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Provable adversarial robustness for group equivariant tasks: graphs, point clouds, molecules, and more 23

Provable adversarial robustness for group equivariant tasks:...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Jan Schuchardt Yan Scholten Stephan Günnemann Department of Computer Science & Munich Data Science Institute Technical University of Munich

A machine learning model is traditionally considered robust if its prediction remains (almost) constant under input perturbations with small norm. However, real-world tasks like molecular property prediction or point cloud segmentation have inherent equivariances, such as rotation or permutation equivariance. In such tasks, even perturbations with large norm do not necessarily change an input's semantic content. Furthermore, there are perturbations for which a model's prediction explicitly needs to change. For the first time, we propose a sound notion of adversarial robustness that accounts for task equivariance. We then demonstrate that provable robustness can be achieved by (1) choosing a model that matches the task's equivariances (2) certifying traditional adversarial robustness. Certification methods are, however, unavailable for many models, such as those with continuous equivariances. We close this gap by developing the framework of equivariance-preserving randomized smoothing, which enables architecture-agnostic certification. We additionally derive the first architecture-specific graph edit distance certificates, i.e. sound robustness guarantees for isomorphism equivariant tasks like node classification. Overall, a sound notion of robustness is an important prerequisite for future work at the intersection of robust and geometric machine learning.

关键词：

来源：评论

学校读者我要写书评

暂无评论

All-fiber and self-starting Yb-doped Mamyshev oscillator based on nonlinear polarization evolution

引用

Chinese Optics Letters 2024年第12期22卷 74-79页

作者： Zecong Li Kai Wang Zinan Huang Qianqian Huang YanlüLin Weixi Li Zhichao Luo Chengbo Mou Key Laboratory of Specialty Fiber Optics and Optical Access Networks Shanghai Institute for Advanced Communication and Data ScienceJoint International Research Laboratory of Specialty Fiber Optics and Advanced CommunicationShanghai UniversityShanghai 200444China Key Laboratory of Space Active Opto-Electronics Technology Shanghai Institute of Technical PhysicsChinese Academy of SciencesShanghai 200083China Laboratory of Nanophotonic Functional Materials and Devices School of Information and Optoelectronic Science and EngineeringSouth China Normal UniversityGuangzhou 510006China

We experimentally demonstrate a Yb-doped all-fiber mode-locked laser based on the Mamyshev *** entire experimental setup operates only by injecting pump powers and adjusting polarization controllers(PCs),which realizes *** types of pulse patterns are observed at different pump powers and polarization states,including single pulses and up to eight-pulse bound-state *** operating wavelength of single-pulse mode-locking switching between 1072.3and 1043.1 nm can be realized by increasing the pump power while keeping the PCs in a fixed *** design can provide an attractive experimental model for all-fiber and self-starting Mamyshev oscillators.

关键词： Mamyshev oscillator Yb-doped fiber laser pulse dynamic pattern nonlinear polarization evolution

来源：评论

学校读者我要写书评

暂无评论

Efficient Plant Disease Detection using K-Means clustering and DenseNet-based Classification 3

Efficient Plant Disease Detection using K-Means clustering a...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Yashodha, G. Rajakumar, B. Krishnamoorthy, M. Surendran, R. Kg College of Arts and Science Department of Computer Technology Coimbatore641035 India J.N.N. Institute of Engineering Department of Artificial Intelligence and Data Science Chennai601102 India Panimalar Engineering College Department of Computer Science and Engineering Chennai600 123 India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798331509675

ISBN: (纸本)9798331509675

Early detection is vital in crop health, yet improvement in productivity faces time-consuming and inefficient challenges due to traditional manual techniques of plant disease detection. Thus, we present a deep learning-based novel framework with a method called k-means clustering with DenseNet Classification denoted by Kmc-Densenet. The approach streamlines this plant disease detection process involving the overall pipeline of extensive preprocessing, segmentation, and classification. In the preprocessing stage, we employ contrast enhancement techniques to improve the visibility of disease symptoms, ensuring clearer and more consistent leaf images for further analysis. For the segmentation task, we leverage a k-means clustering which effectively isolates regions of interest, distinguishing between healthy and diseased areas of the plant leaf. In the classification stage, we use a DenseNet variant of CNN, known for its feature reuse capabilities, to classify the plant diseases with high accuracy. The DenseNet model helps overcome the challenges posed by class imbalance, enhancing the overall detection performance. Finally the parameters of the Densenet model is fine-tuned using the Lyrebird Optimization Algorithm. The proposed Kmc-Densenet method is evaluated on the publicly available Plant village dataset, demonstrating superior performance in both segmentation and classification tasks. This framework offers a more efficient and reliable solution for early plant disease detection in precision agriculture, with potential applications in improving crop yield and quality. © 2025 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Hyper-heuristic method for processor allocation in parallel tasks scheduling

Hyper-heuristic method for processor allocation in parallel ...

引用

作者： Yıldız, Gülçin Sevilgen, Fatih Erdoğan Department of Computer Engineering Gebze Technical University Kocaeli Turkey Institute for Data Science and Artificial Intelligence Boğaziçi University İstanbul Turkey

Scheduling the tasks of parallel scientific applications is very important for efficient utilization of resources and reducing the overall execution time (makespan). Parallel applications typically include both data parallelism and task parallelism. It is known that the scheduling problem on multiprocessor systems problem is NP-Hard even for applications involving pure task parallelism. The problem becomes more difficult when data parallelism is also taken into consideration. These problems usually considered in two steps, processor allocation and task scheduling, and various algorithms have been proposed. In this study, we introduce a genetic algorithm based hyper-heuristic approach for the processor allocation problem. Experimental results indicate that the algorithm provides better performance compared to various greedy algorithms. © 2023 John Wiley & Sons, Ltd.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Cocoon morphological Features Based Silk Quality Prediction Using XG Boost Algorithm

Cocoon morphological Features Based Silk Quality Prediction ...

引用

2024 IEEE International Students' Conference on Electrical, Electronics and computer science, SCEECS 2024

作者： Nivetha, M. Sudha, I. Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Artificial Intelligence and Data Science Tamil Nadu Chennai India Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350348460

Silk quality is a critical determinant in the silk industry, significantly influencing the market value of silk products. This study proposes an innovative method for forecasting silk quality by leveraging cocoon morphological features through the application of the XG Boost algorithm. The morphological characteristics of the cocoon, including dimensions, shape, and colour, are identified as key indicators of silk quality. The XG Boost algorithm, known for its effectiveness in handling complex datasets, is employed for building a predictive model. A comprehensive dataset is assembled, encompassing a diverse set of cocoon morphological features obtained from a representative sample of silkworms. These features comprise cocoon size, length, width, shape parameters, and colour attributes. The dataset is then utilized to train and validate the XG Boost model, fine-tuning its parameters to ensure accurate prediction of silk quality. The outcomes of this study showcase the effectiveness of the proposed approach, demonstrating high predictive accuracy in determining silk quality based on cocoon morphological features. Moreover, the XG Boost algorithm facilitates the identification of crucial features that significantly contribute to the prediction, offering valuable insights into the factors influencing silk quality. In conclusion, the integration of cocoon morphological features and the XG Boost algorithm presents a promising avenue for accurate silk quality prediction. This research contributes to the advancement of silk production methodologies, introducing innovative technologies to the silk industry and fostering a pathway for the adoption of cutting-edge practices. © 2024 IEEE.

关键词： data Security Deep Learning Machine Learning Quality Prediction Silk Quality XGBoost Algorithm

来源：评论

学校读者我要写书评

暂无评论

Detection of Indian Regional Sign Language Through Convolutional Neural Network

Detection of Indian Regional Sign Language Through Convoluti...

引用

2023 IEEE Global Conference on Information Technologies and Communications, GCITC 2023

作者： Baroliya, Mohit Srivastava, Himanshu Iyer, Aishwarya Jayanthi, M.G. Kannadaguli, Prashanth Computer Science and Technology Cambridge Institute of Technology Bengaluru India Department of Cse Bengaluru India Senior Data Science Trainer Dhaarini Academy of Technical Education Bengaluru India

ISBN: (纸本)9798350308167

The objective of this endeavor is to create a specialized sign language recognition system created specifically for Kannada, an indigenous language spoken in India. While there are existing recognition systems for widely spoken languages like Hindi and English, there is a lack of resources and options for regional languages like Kannada. By leveraging a dataset of 50,000 Kannada sign language samples, The objective of this initiative is to close the communication divide experienced by individuals who rely on Kannada sign language, effectively addressing the need for improved interaction. The project focuses on exploring and comparing various AI and DL models to identify the most effective approach for Kannada sign language recognition. The ultimate goal is to enhance accessibility, inclusivity, and understanding for individuals communicating in Kannada sign language, thereby promoting cultural diversity and equal opportunities for the Kannada-speaking sign language community in India. The results and consequences of this endeavor have the potential to lay the groundwork for the advancement of location-specific sign language recognition systems. This would significantly enhance communication and promote the self-determination of individuals with hearing impairments residing in the Kannada-speaking areas of India. © 2023 IEEE.

关键词： CNN Kannada Sign language detection

来源：评论

学校读者我要写书评

暂无评论

Neural Pfaffians: Solving Many Many-Electron Schrödinger Equations

arXiv

引用

arXiv 2024年

作者： Gao, Nicholas Günnemann, Stephan Department of Computer Science Munich Data Science Institute Technical University of Munich Germany

Neural wave functions accomplished unprecedented accuracies in approximating the ground state of many-electron systems, though at a high computational cost. Recent works proposed amortizing the cost by learning generalized wave functions across different structures and compounds instead of solving each problem independently. Enforcing the permutation antisymmetry of electrons in such generalized neural wave functions remained challenging as existing methods require discrete orbital selection via non-learnable hand-crafted algorithms. This work tackles the problem by defining overparametrized, fully learnable neural wave functions suitable for generalization across molecules. We achieve this by relying on Pfaffians rather than Slater determinants. The Pfaffian allows us to enforce the antisymmetry on arbitrary electronic systems without any constraint on electronic spin configurations or molecular structure. Our empirical evaluation finds that a single neural Pfaffian calculates the ground state and ionization energies with chemical accuracy across various systems. On the TinyMol dataset, we outperform the ‘gold-standard’ CCSD(T) CBS reference energies by 1.9 mEh and reduce energy errors compared to previous generalized neural wave functions by up to an order of magnitude. © 2024, CC BY.

关键词： Ground state

来源：评论

学校读者我要写书评

暂无评论

EXTRACTING UNLEARNED INFORMATION FROM LLMS WITH ACTIVATION STEERING

arXiv

引用

arXiv 2024年

作者： Seyitoğlu, Atakan Kuvshinov, Aleksei Schwinn, Leo Günnemann, Stephan Department of Computer Science Munich Data Science Institute Technical University of Munich Germany

An unintended consequence of the vast pretraining of Large Language Models (LLMs) is the verbatim memorization of fragments of their training data, which may contain sensitive or copyrighted information. In recent years, unlearning has emerged as a solution to effectively remove sensitive knowledge from models after training. Yet, recent work has shown that supposedly deleted information can still be extracted by malicious actors through various attacks. Still, current attacks retrieve sets of possible candidate generations and are unable to pinpoint the output that contains the actual target information. We propose activation steering as a method for exact information retrieval from unlearned LLMs. We introduce a novel approach to generating steering vectors, named Anonymized Activation Steering. Additionally, we develop a simple word frequency method to pinpoint the correct answer among a set of candidates when retrieving unlearned information. Our evaluation across multiple unlearning techniques and datasets demonstrates that activation steering successfully recovers general knowledge (e.g., widely known fictional characters) while revealing limitations in retrieving specific information (e.g., details about non-public individuals). Overall, our results demonstrate that exact information retrieval from unlearned models is possible, highlighting a severe vulnerability of current unlearning techniques. © 2024, CC BY.

关键词： Information retrieval

来源：评论

学校读者我要写书评

暂无评论

Learning Equivariant Non-Local Electron Density Functionals

arXiv

引用

arXiv 2024年

作者： Gao, Nicholas Eberhard, Eike Gunnemann, Stephan Department of Computer Science Munich Data Science Institute Technical University of Munich Germany

The accuracy of density functional theory hinges on the approximation of nonlocal contributions to the exchange-correlation (XC) functional. To date, machinelearned and human-designed approximations suffer from insufficient accuracy, limited scalability, or dependence on costly reference data. To address these issues, we introduce Equivariant Graph Exchange Correlation (EG-XC), a novel non-local XC functional based on equivariant graph neural networks. EG-XC combines semilocal functionals with a non-local feature density parametrized by an equivariant nuclei-centered point cloud representation of the electron density to capture longrange interactions. By differentiating through a self-consistent field solver, we train EG-XC requiring only energy targets. In our empirical evaluation, we find EG-XC to accurately reconstruct 'gold-standard' CCSD(T) energies on MD17. On out-of-distribution conformations of 3BPA, EG-XC reduces the relative MAE by 35% to 50%. Remarkably, EG-XC excels in data efficiency and molecular size extrapolation on QM9, matching force fields trained on 5 times more and larger molecules. On identical training sets, EG-XC yields on average 51% lower MAEs. © 2024, CC BY.

关键词： Graph neural networks

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：