Search

By utilizing hybrid quantum–classical neural networks (HNNs), this research aims to enhance the efficiency of image classification tasks. HNNs allow us to utilize quantum computing to solve machine learning problems, which can be highly power-efficient and provide significant computation speedup compared to classical operations. This is particularly relevant in sustainable applications where reducing computational resources and energy consumption is crucial. This study explores the feasibility of a novel architecture by leveraging quantum devices as the first layer of the neural network, which proved to be useful for scaling HNNs' training process. Understanding the role of quanvolutional operations and how they interact with classical neural networks can lead to optimized model architectures that are more efficient and effective for image classification tasks. This research investigates the performance of HNNs across different datasets, including CIFAR100 and Satellite Images of Hurricane Damage by evaluating the performance of HNNs on these datasets in comparison with the performance of reference classical models. By evaluating the scalability of HNNs on diverse datasets, the study provides insights into their applicability across various real-world scenarios, which is essential for building sustainable machine learning solutions that can adapt to different environments. Leveraging transfer learning techniques with pre-trained models such as ResNet, EfficientNet, and VGG16 demonstrates the potential for HNNs to benefit from existing knowledge in classical neural networks. This approach can significantly reduce the computational cost of training HNNs from scratch while still achieving competitive performance. The feasibility study conducted in this research assesses the practicality and viability of deploying HNNs for real-world image classification tasks. By comparing the performance of HNNs with classical reference models like ResNet, EfficientNet, and VGG-16, this study provides evidence of the potential advantages of HNNs in certain scenarios. Overall, the findings of this research contribute to advancing sustainable applications of machine learning by proposing novel techniques, optimizing model architectures, and demonstrating the feasibility of adopting HNNs for real-world image classification problems. These insights can inform the development of more efficient and environmentally friendly machine learning solutions. [ABSTRACT FROM AUTHOR]

'Science gateway' (SG) ideology means a user-friendly intuitive interface between scientists (or scientific communities) and different software components + various distributed computing infrastructures (DCIs), where researchers can focus on their scientific goals and less on the peculiarities of software/DCI. G.V.Kurdyumov Institute for Metal Physics 'IMP Science Gateway Portal' () is presented for complex workflow management and integration of distributed computing resources (like clusters, service grids, desktop grids, and clouds). It is created on the basis of Web Service - Parallel Grid Run-time and Application Development Environment (WS-PGRADE) and gUSE (grid and cloud User Support Environment) technologies, where WS-PGRADE is designed for science workflow operation and gUSE - for smooth integration of available resources for parallel and distributed computing in various heterogeneous DCIs. Some use cases (scientific workflows) are considered for molecular dynamics simulations of complex behavior of various nanostructures. The modular approach allows scientists to use SG portals as research hubs of various virtual experimental labs in the context of practical applications in material science, physics, and nanotechnologies. In addition, workflows and their components are proposed to be used as Lego-style construction units for learning modules of various scale by duration, complexity, targeted audience, and so on. These workflows can be used also in e-Learning infrastructures as constituent elements of learning hubs for the management of learning content, tools, resources, and users in the regular, vocational, lifelong, and informal learning. Copyright © 2015 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

The rate equation for exchange-driven aggregation of monomers between clusters of size n by power-law exchange rate (~ nα), where detaching and attaching processes were considered separately, is reduced to Fokker-Planck equation. Its exact solution was found for unbiased aggregation and agreed with asymptotic conclusions of other models. Asymptotic transitions were found from exact solution to Weibull/normal/exponential distribution, and then to power law distribution. Intermediate asymptotic size distributions were found to be functions of exponent α and vary from normal (α = 0) through Weibull (0 < α < 1) to exponential (α =1) ones, that gives the new system for linking these basic statistical distributions. Simulations were performed for the unbiased aggregation model on the basis of the initial rate equation without simplifications used for reduction to Fokker-Planck equation. The exact solution was confirmed, shape and scale parameters of Weibull distribution (for 0 < α < 1) were determined by analysis of cumulative distribution functions and mean cluster sizes, which are of great interest, because they can be measured in experiments and allow to identify details of aggregation kinetics (like α). In practical sense, scaling analysis of evolving series of aggregating cluster distributions can give much more reliable estimations of their parameters than analysis of solitary distributions. It is assumed that some apparent power and fractal laws observed experimentally may be manifestations of such simple migration-driven aggregation kinetics even. [ABSTRACT FROM AUTHOR]