检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Ganjdanesh, Alireza Gao, Shangqian Alipanah, Hirad Huang, Heng Department of Computer Science University of Maryland College Park College ParkMD20742 United States Department of Electrical and Computer Engineering University of Pittsburgh PittsburghPA15261 United States Department of Mechanical Engineering and Materials Science University of Pittsburgh PittsburghPA15261 United States

Generative Adversarial Networks (GANs) have shown remarkable success in modeling complex data distributions for image-to-image translation. Still, their high computational demands prohibit their deployment in practical scenarios like edge devices. Existing GAN compression methods mainly rely on knowledge distillation or convolutional classifiers’ pruning techniques. Thus, they neglect the critical characteristic of GANs: their local density structure over their learned manifold. Accordingly, we approach GAN compression from a new perspective by explicitly encouraging the pruned model to preserve the density structure of the original parameter-heavy model on its learned manifold. We facilitate this objective for the pruned model by partitioning the learned manifold of the original generator into local neighborhoods around its generated samples. Then, we propose a novel pruning objective to regularize the pruned model to preserve the local density structure over each neighborhood, resembling the kernel density estimation method. Also, we develop a collaborative pruning scheme in which the discriminator and generator are pruned by two pruning agents. We design the agents to capture interactions between the generator and discriminator by exchanging their peer’s feedback when determining corresponding models’ architectures. Thanks to such a design, our pruning method can efficiently find performant sub-networks and can maintain the balance between the generator and discriminator more effectively compared to baselines during pruning, thereby showing more stable pruning dynamics. Our experiments on image translation GAN models, Pix2Pix and CycleGAN, with various benchmark datasets and architectures demonstrate our method’s effectiveness. © 2023, CC BY-NC-SA.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Investigating the Behavior of Diffusion Models for Accelerating Electronic Structure Calculations

arXiv

引用

arXiv 2023年

作者： Rothchild, Daniel Rosen, Andrew S. Taw, Eric Robinson, Connie Gonzalez, Joseph E. Krishnapriyan, Aditi S. Department of Electrical Engineering and Computer Science University of California Berkeley United States Department of Materials Science and Engineering University of California Berkeley United States Materials Science Division Lawrence Berkeley National Laboratory United States Department of Chemical and Biomolecular Engineering University of California Berkeley United States Department of Chemistry University of California Berkeley United States

We present an investigation into diffusion models for molecular generation, with the aim of better understanding how their predictions compare to the results of physics-based calculations. The investigation into these models is driven by their potential to significantly accelerate electronic structure calculations using machine learning, without requiring expensive first-principles datasets for training interatomic potentials. We find that the inference process of a popular diffusion model for de novo molecular generation is divided into an exploration phase, where the model chooses the atomic species, and a relaxation phase, where it adjusts the atomic coordinates to find a low-energy geometry. As training proceeds, we show that the model initially learns about the first-order structure of the potential energy surface, and then later learns about higher-order structure. We also find that the relaxation phase of the diffusion model can be re-purposed to sample the Boltzmann distribution over conformations and to carry out structure relaxations. For structure relaxations, the model finds geometries with ∼ 10× lower energy than those produced by a classical force field for small organic molecules. Initializing a density functional theory (DFT) relaxation at the diffusion-produced structures yields a > 2× speedup to the DFT relaxation when compared to initializing at structures relaxed with a classical force field. Copyright © 2023, The Authors. All rights reserved.

关键词： Density functional theory

来源：评论

学校读者我要写书评

暂无评论

Max-Plus Synchronization in Decentralized Trading Systems

arXiv

引用

arXiv 2023年

作者： Riess, Hans Munger, Michael Zavlanos, Michael M. Department of Electrical and Computer Engineering Duke University United States Department of Political Science Duke University United States Department of Economics Duke University United States Dept. of Mechanical Engineering & Materials Science Duke University United States

We introduce a decentralized mechanism for pricing and exchanging alternatives constrained by transaction costs. We characterize the time-invariant solutions of a heat equation involving a (weighted) Tarski Laplacian operator, defined for max-plus matrix-weighted graphs, as approximate equilibria of the trading system. We study algebraic properties of the solution sets as well as convergence behavior of the dynamical system. We apply these tools to the "economic problem" of allocating scarce resources among competing uses. Our theory suggests differences in competitive equilibrium, bargaining, or cost-benefit analysis, depending on the context, are largely due to differences in the way that transaction costs are incorporated into the decision-making process. We present numerical simulations of the synchronization algorithm (RRAggU), demonstrating our theoretical *** Codes 15A80, 90C24 Copyright © 2023, The Authors. All rights reserved.

关键词： Cost benefit analysis

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Identify the Anomaly Detection in Electricity usage based on Machine Learning Algorithms and Big Data 3

A Novel Approach to Identify the Anomaly Detection in Electr...

引用

3rd International Conference on Smart Electronics and Communication, ICOSEC 2022

作者： Panneerselvam, M. Neela, K. Rajeshwari, R. Vengadapathiraj, M. Sobitha, S. Mohanavel, V. Sona College of Technology Department of Electrical and Electronics Engineering Salem India Sri Sairam Engineering College Department of Information Technology Chennai India S.A Engineering College Department of Master of Computer Application Chennai India Rajalakshmi Instittue of Technology Department of Electronics and Communication Engineering Chennai India Saveetha School of Engineering Simats Department of Computer Science and Engineering Chennai India Centre for Materials Engineering and Regenerative Medicine Bharath Institute of Higher Education and Research Tamil Nadu Chennai600073 India Chandigarh University Department of Mechanical Engineering Punjab Mohali140413 India

ISBN: (数字)9781665497640

ISBN: (纸本)9781665497640

Comprehension mistakes and frauds may both be detected via the analysis of advanced metering information. By means of on-site examinations or even other techniques, the energy suppliers will be made aware of any suspect usage patterns by this study. The analysis of usage patterns using Machine Learning (ML) techniques may reveal errors, measuring interruptions from assaults, or fine motor control with utility companies. Another of the basic instances of an anomaly detection is fraud detection since it is difficult to categories transactional or utilization information. Additionally, scams vary in type, making it difficult to consistently learn from them. This research study presents an unsupervised ML method to identify out-of-the-ordinary qualities in time series, set a threshold for the proportion of out-of-the-ordinary readings towards the overall measurements, and then classify the moving average as questionable or not. Initially, two different methods are provided to find the differences in time-series based data streaming: Spectral Residual-Convolutional Neural Network and anomaly trained model based on martingales. The properly established information is then subjected to Fisher Linear Discriminant analysis and Two-Class Boosted Decision Tree applications. © 2022 IEEE.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Fully-Printed Sensor Based Extended Gate Field Effect Transistors for Wireless Monitoring of Potassium and Ammonium Ions

Fully-Printed Sensor Based Extended Gate Field Effect Transi...

引用

IEEE SENSORS

作者： Munia Ferdoushi Mohammad Shafiqul Islam Wenxin Cai Sandra Lara Galindo Md Farhad Hassan Yasser Khan Ming Hsieh Department of Electrical and Computer Engineering University of Southern California Los Angeles California Department of Chemical Engineering and Materials Science University of Southern California Los Angeles California Alfred E. Mann Department of Biomedical Engineering University of Southern California Los Angeles California

ISBN: (数字)9798350363517

ISBN: (纸本)9798350363524

This paper presents an extended gate field effect transistor based ion sensing platform utilizing flexible printed ion selective electrodes. A custom-printed circuit board with a Bluetooth-enabled microcontroller module enables wireless monitoring of the transistor's transfer characteristics modulated by the ion concentration in the target solution. The utilization of printing techniques such as inkjet printing and direct 3D writing provide the scope for facile and cost-effective fabrication. The system has been utilized for remote monitoring of Potassium and Ammonium ions with ideal Nernstian response, high sensitivity, and selectivity. This work provides a promising path for real-time and continuous ion monitoring which has crucial implications for health, industry, and environmental applications.

关键词： Wireless communication Wireless sensor networks Sensitivity Field effect transistors Logic gates Writing Ions Three-dimensional printing Sensors Potassium

来源：评论

学校读者我要写书评

暂无评论

Evolution-Assisted Deep Reinforcement Learning for Fast Charging Station Coordinated Operation 13

Evolution-Assisted Deep Reinforcement Learning for Fast Char...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Yang, Xiaoying Gu, Yujing Jia, Fuhua Li, Yiran Wang, Hongru Du, Nanjiang Cui, Tianxiang Ye, Yujian Bai, Ruibin State Grid Nantong Power Supply Company Nantong China University of Nottingham Ningbo China Department of Mechanical Materials and Manufacturing Engineering Ningbo China School of Electrical Engineering Southeast University Nanjing China School of Computer Science University of Nottingham Ningbo China Ningbo China University of Nottingham Ningbo China The Digital Port Technologies Lab Ningbo China Nottingham Ningbo China Beacons of Excellence Research and Innovation Institute Ningbo China

ISBN: (纸本)9798350308365

The shift towards transportation electrification, marked by the rising use of electric vehicles (EVs) and the development of fast charging stations (FCS), plays a crucial role in transport decarbonization initiatives. To optimize the rollout of FCS and set appropriate charging service fees (CSF)-a process referred to as the coupled FCS multi-stage bi-level operation problem (FCS-MBOP)-is essential for improving both investment and operational efficiency within the integrated power distribution and transportation network (CPTN). For operators, it's not only necessary to adapt to short-term fluctuations within the environment but also to swiftly respond to changes in the FCS layout resulting from various long-term investment decisions. To address this complexity, we introduce a dual-timescale evolutionary assist deep reinforcement learning framework, which includes two specialized agents with distinct functions: an investment agent (planner) and an operational agent (operator). The planner focuses on annual investments, evolving long-term strategies that weigh social benefits against investment costs through the use of a genetic algorithm (GA). In contrast, the operator acts on an hourly basis, fine-tuning CSF to alleviate traffic congestion and minimize the social costs, while taking into account the planner's feasible investment decisions. Leveraging the integrated capabilities of a graph neural network (GNN), long-short-term memory (LSTM), and attention mechanisms, our framework's agents are adept at extracting both temporal and spatial features and facilitating the transfer of experiences across different investment stages. Empirical evidence underscores the effectiveness of our approach, showcasing its ability to surpass conventional methodologies in delivering high-quality solutions. © 2024 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Study on the Flow, Foaming Characteristics and Structural Strength of Polypropylene Structural Foam Injection Molding by Innovative Nitrogen and Molten Plastic Mixing Mechanism

引用

Polymers 2023年第9期15卷 2116-2116页

作者： Huang, Po-Wei Peng, Hsin-Shu Hwang, Sheng-Jye Huang, Chao-Tsai Program of Mechanical and Aeronautical Engineering Feng Chia University of Engineering and Science Taichung407102 Taiwan Department of Mechanical and Computer Aided Engineering Feng Chia University of Engineering and Science Taichung407102 Taiwan Department of Mechanical Engineering National Cheng Kung University of Engineering Tainan70101 Taiwan Department of Chemical and Materials Engineering Tamkang University of Engineering New Taipei City251301 Taiwan

Plastic foam molding methods include thermoforming, extrusion and injection molding. Injection foam molding is a one-time molding method with high production efficiency and good product quality. It is suitable for foamed plastic products with complex shapes and strict size requirements. It is also the main method for producing structural bubbles. In this investigation, we developed a structural foam injection molding technology using the gas supply equipment connected to the unique plasticizing mechanism of the injection machine and studied its influence on the specimens’ melt rheology quality and foam structures. In the experiment, the forming material was polypropylene (PP), and the gas for mixing/forming foaming characteristics was nitrogen (N2). Additionally, in order to observe the rheological properties of N2/melt mixing, a melt flow specimen mold cavity was designed and the change in the melt viscosity index was observed using a melt pressure sensing element installed at the nozzle position. With the nitrogen supply equipment connected to a unique plasticizing mechanism, the mixing of gas and molten plastic can be achieved at the screw plasticizing stage, where the foaming effect is realized during the melt-filling process due to the thermodynamic instability of the gas. It was also found that an increase in N2 fill content increased melt fluidity, and the trend of melt pressure and melt viscosity index showed that the higher the gas content, the lower the trend. The foaming characteristic depends on the gas thermodynamic instability and the pressure release, so it can be seen from the melt fill path that, the greater the pressure near the gate, the lower the foaming amount and the internal structure (SEM) after molding;the farther from the gate, the greater the relative increase in the foaming growth/amount. This phenomenon will be more obvious when the N2 fill content is increased. © 2023 by the authors.

关键词： Gases

来源：评论

学校读者我要写书评

暂无评论

An Improved Network Segmentation Performance in Lesion Segmentation based on Mask R-CNN 6

An Improved Network Segmentation Performance in Lesion Segme...

引用

6th International Conference on Electronics, Communication and Aerospace Technology, ICECA 2022

作者： Thivya, T. Kalaivanan, C. Juliet, N. Vengadapathiraj, M. Mary Valantina, G. Mohanavel, V. Velammal Engineering College Department of Information Technology Chennai India Sona College of Technology Department of Electrical and Electronics Engineering Salem India S.A Engineering College Department of Master of Computer Application Chennai India Rajalakshmi Instittue of Technology Department of Electronics and Communication Engineering Chennai India Saveetha School of Engineering Simats Department of Computer Science and Engineering Chennai India Centre for Materials Engineering and Regenerative Medicine Bharath Institute of Higher Education and Research Tamil Nadu Chennai600073 India Chandigarh University Department of Mechanical Engineering Punjab Mohali140413 India

ISBN: (纸本)9781665482714

Computational intelligence has grown increasingly common in the analysis of healthcare data in recent years. The deep learning algorithms, in instance, have such a number of fruitful implications in the analysis of medical images. This research study has presented a double awareness deep convolution network-based system for end-to-end healthcare lesion classification. Several fully supervised and poorly supervised extracted features are included in this framework. The issue of the high cost of pixel-level labels with lesions in the medical pictures is alleviated by the weakly supervised segmentation module's use of bounding-box labels of lesion regions, which allows for correct lesion segmentation to be accomplished. In addition to this, a dual attention mechanism has been included for the purpose of improving the network's capacity for visual feature learning. The dual attention mechanism, which consists of both channel and spatial attention, may assist in drawing the network's attention to the extraction of features from significant locations. It is possible for it to considerably minimize the gaps between ground-truth labels and pseudo labels, in comparison to the current standard approach of weakly supervised segmentation using pseudo labels. Final results show that our proposed approach beat the competitors on to an oral disease collection, while our approach was extended to incorporate dermatological disease categorization in the ultimate analysis. © 2022 IEEE.

关键词： Intelligent systems

来源：评论

学校读者我要写书评

暂无评论

Shear strain alters the structure and migration mechanism of self-interstitial atoms in copper

引用

Physical Review materials 2022年第5期6卷 053605-053605页

作者： B. Zhang C. Wheatley P. Chen X. Qian M. J. Demkowicz Department of Materials Science and Engineering Texas A&M University College Station Texas 77845 USA Department of Materials Science and Engineering Penn State University State College Pennsylvania 16801 USA Key Laboratory of Automobile Materials of Ministry of Education School of Materials Science and Engineering Jilin University Changchun People's Republic of China Department of Electrical and Computer Engineering Texas A&M University College Station Texas 77845 USA

We use atomistic modeling to show that externally applied shear strain causes the lowest energy self-interstitial atom (SIA) structure in copper (Cu) to change from a ⟨100⟩-type dumbbell to a ⟨110⟩-type dumbbell. Concurrently, SIA migration switches from the three-dimensional (3D) random walk characteristic of ⟨100⟩-type dumbbells to a 1D mechanism analogous to that of crowdion SIAs. Furthermore, the relative energies of these two dumbbell structures as a function of strain are well predicted using elastic dipole tensors computed at zero strain, indicating that examination of these tensors may be used to assess the likelihood of strain-induced SIA structure transitions in other materials. Changes in lowest energy SIA structures and associated migration mechanisms stand to impact predictions of SIA behavior in irradiated solids.

关键词： Interstitials Face-centered cubic Transition metals Density functional calculations

来源：评论

学校读者我要写书评

暂无评论

Influence of Crystallographic Orientation on the Deformation of Ag Nanoparticles During High-Speed Impact

SSRN

引用

SSRN 2022年

作者： Becker, Michael F. Kovar, Desiderio Chitrakar, T.V. The University of Texas at Austin AustinTX78712 United States Materials Science and Engineering Program Texas Materials Institute United States Department of Electrical and Computer Engineering United States Department of Mechanical Engineering Center for Additive Manufacturing and Design Innovation

Several related aerosol processes utilize high velocity impact of solid nanoparticles to produce nanograined films including the aerosol deposition method, jet molding, vacuum kinetic spraying, and micro cold spray. Although the influence of important variables such as particle size and particle impact velocity, on particle deformation and film formation have been previously explored, other variables have not been systematically explored. Because particles impact with a full range of orientations, an understanding of the effect of particle orientation is required to predict film microstructures. Molecular dynamics simulations were conducted to determine the influence of particle crystallographic orientation on the particle deformation upon impact and the resulting microstructure of the deposit. We show that the orientations that produce the largest overall particle deformation are not correlated to orientations where initiation of plastic deformation is easiest. Rather, the deformation is heterogenous and depends on the mechanisms responsible for deformation. Two deformation mechanisms are identified: 1) dislocation plasticity and 2) disordering followed by viscous flow. The fraction of the atoms in the impacting particle that experience permanent deformation by each mechanism is quantified as a function of particle orientation. The implications of the effects of particle crystallographic orientation on film microstructures are also discussed. © 2022, The Authors. All rights reserved.

关键词： Molecular dynamics

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：