检索结果-内蒙古大学图书馆

SSRN 2022年

作者： Mousavi, Seyedhamidreza Hamid Loni, Mohammad Alibeigi, Mina Daneshtalab, Masoud School of Innovation Design and Engineering Malardalen University Vasteras72218 Sweden Zenseact AB Lindholmspiren2 Goteborg41756 Sweden The Department of Computer Science and Technology The University of Cambridge 15 JJ Thomson Ave CambridgeCB3 0FD United Kingdom School of Innovation Design and Engineering Malardalen Vasteras72218 Sweden Department of Computer Systems TalTech University Tallinn19086 Estonia

The deployment of Convolutional Neural Networks (CNNs) on edge devices is hindered by the substantial gap between performance requirements and available processing power. While recent research has made large strides in developing network pruning methods for reducing the computing overhead of CNNs, there remains considerable accuracy loss, especially at high pruning ratios. We find that the architectures designed for non-pruned networks are not effective for pruned networks since there is no support for characteristics of pruned operations in the search space. Plus, conventional NAS methods use a search objective specialized for non-pruned networks. We propose an algorithm to search architectures for pruning methods by defining a new search space and a novel search objective. To improve the generalization of the pruned networks, we propose two novel PrunedConv and PrunedLinear operations. Specifically, these operations mitigate the problem of unstable gradients by regularizing the objective function of the pruned networks. The proposed search objective enables us to train architecture parameters regarding the pruned weight elements. Quantitative analyses demonstrate that our searched architectures outperform those used in the state-of-the-art pruning networks on CIFAR-10 and ImageNet datasets. In terms of hardware effectiveness, PR-DARTS increases the accuracy of the pruned version of MobileNet-v2 from 73.44% to 81.35% (+7.91% improvement) with 3.87× faster inference time. © 2022, The Authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Lattice Hamiltonians and Stray Interactions within Quantum Processors

arXiv

引用

arXiv 2024年

作者： Xu, Xuexin Manabputra Vignes, Chloé Ansari, Mohammad H. Martinis, John M. Peter Grünberg Institute PGI-2 Forschungszentrum Jülich Jülich52428 Germany Department of Physics Syracuse University SyracuseNY13244 United States Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States Department of Physics University of California Santa Barbara Santa BarbaraCA93106 United States Qolab Inc 9854 National Blvd 1436 Los AngelesCA90034 United States

Developing Hamiltonian models for quantum processors with many qubits on the same chip is crucial for advancing quantum computing technologies. Stray couplings between qubits lead to errors in gate operations. This study underscores the importance of incorporating lattice Hamiltonians into quantum circuit design. By comparing many-body effects with two-body stray couplings, we show how adjusting circuit parameters can enhance two-qubit gate fidelity. We find that loosely decoupled qubits result in weaker stray interactions and higher gate fidelity, challenging conventional assumptions. We investigate the scenario where three-body ZZZ interaction surpasses two-body ZZ interactions, highlighting the transformative potential of lattice Hamiltonians for novel multi-qubit gates. Moreover, we investigate the cross-resonance gate within the lattice Hamiltonian framework and examine the impact of microwave pulses on stray coupling. This emphasizes the necessity of developing a comprehensive theoretical framework that includes lattice interactions, which are now critical given the sophistication of contemporary quantum hardware. These insights are vital for developing fault-tolerant quantum computing and next-generation quantum processors. © 2024, CC BY.

关键词： Hamiltonians

来源：评论

学校读者我要写书评

暂无评论

Energy Mapping of Existing Building Stock in Cambridge using Energy Performance Certificates and Thermal Infrared Imagery

arXiv

引用

arXiv 2024年

作者： He, Yinglong Pan, Jiayu Debnath, Ramit Bardhan, Ronita Cullen, Luke Jenkins, Marco Gomez Mackie, Erik Hawker, George Parry, Ian Institute of Astronomy University of Cambridge CambridgeCB3 0HA United Kingdom Department of Architecture University of Cambridge CambridgeCB2 1PX United Kingdom EPRG Cambridge Judge Business School University of Cambridge CambridgeCB2 1AG United Kingdom Cambridge Zero Department of Computer Science and Technology Department of Architecture CambridgeCB3 OFD United Kingdom Division of Humanities and Social Science California Institute of Technology Pasadena91125 United States School of Mechanical Engineering Sciences University of Surrey SurreyGU2 7XH United Kingdom

Energy performance certificate (EPC) and thermal infrared (TIR) images both play a key role in the energy performance mapping of the urban building stock. In this paper, we developed parametric building archetypes using an EPC database and conducted temperature clustering on TIR images acquired through drones and satellite datasets. We evaluated 1725 EPCs of existing building stock in Cambridge, UK to generate energy consumption profiles. Drone-based TIR images of individual buildings in two Cambridge University colleges were processed using a machine learning pipeline for thermal anomaly detection and investigated the influence of two specific factors that affect the reliability of TIR for energy management applications: ground sample distance (GSD) and angle of view (AOV). The EPC-level results suggest that the construction year of the buildings influences their energy consumption. For example, modern buildings were over 30% more energy-efficient than older ones. Parallelly, older buildings were found to show almost double the energy savings potential through retrofitting than newly constructed buildings. TIR imaging results showed that thermal anomalies can only be properly identified in images with a GSD of 1m/pixel or less. A GSD of 1-6m/pixel can detect hot areas of building surfaces. We found that a GSD > 6m/pixel cannot characterise individual buildings but does help identify urban heat island effects. Additional sensitivity analysis showed that building thermal anomaly detection is more sensitive to AOV than to GSD. Our study informs newer approaches to building energy diagnostics using thermography and supports decision-making for retrofitting at a large scale. Copyright © 2024, The Authors. All rights reserved.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Enabling Smart Manufacturing by Empowering Data Integration with Industrial IoT Support

Enabling Smart Manufacturing by Empowering Data Integration ...

引用

2020 International Conference on Technology and Entrepreneurship, ICTE 2020

作者： Bosi, Filippo Corradi, Antonio Di Modica, Giuseppe Foschini, Luca Montanari, Rebecca Patera, Lorenzo Solimando, Michele University of Bologna Department of Computer Science and Engineering Viale Risorgimento 2 Bologna40136 Italy Imola Informatica SpA Via Selice 66/A ImolaBO 40026 Italy

ISBN: (纸本)9781728149028

Industry 4.0 (I4.0) denotes the start of a new era where another industrial revolution is urged to explode. Simply put, I4.0 aims at enhancing the competitiveness in the manufacturing industry, while at the same time improving the safety and security of work centers. The achievement of that goal is bound to the adoption of enabling information technologies (IoT, Cloud, Big Data) to build a unified and integrated view of the information generated at all stages of the industrial processes, both operational and business oriented ones. This paper focuses on the need for manufacturing companies of tools that concretely implement this data integration as a requirement to enforce the improvements fostered and needed by the I4.0 revolution. This paper follows a former work that proposed the design and implementation of a platform realizing the integration of the operational and the business views at the data level. We enhanced the existing platform features by introducing new component that implements the interface with IoT devices via well-known communication protocols, namely MQTT and AMQP. Experiments run to test the newly introduced feature show the viability of the proposed approach with respect to the time constraints that are typical of a manufacturing production environment. © 2020 IEEE.

关键词： Cyber Physical System

来源：评论

学校读者我要写书评

暂无评论

Predicting Bitcoin Market Trends with Enhanced Technical Indicator Integration and Classification Models

arXiv

引用

arXiv 2024年

作者： Hafid, Abdelatif Rahouti, Mohamed Kong, Linglong Ebrahim, Maad Serhani, Mohamed Adel ESISA Analytica Laboratory Higher School of Engineering in Applied Sciences Fez24000 Morocco The Department of Computer & Information Science Fordham University New YorkNY10023 United States The Department of Mathematical and Statistical Sciences Alberta Machine Intelligence Institute University of Alberta EdmontonABT6G 2G1 Canada GAIA Ericsson Montreal Canada The Department of Information Systems University of Sharjah United Arab Emirates

Thanks to the high potential for profit, trading has become increasingly attractive to investors as the cryptocurrency and stock markets rapidly expand. However, because financial markets are intricate and dynamic, accurately predicting prices remains a significant challenge. The volatile nature of the cryptocurrency market makes it even harder for traders and investors to make decisions. This study presents a machine learning model based on classification to forecast the direction of the cryptocurrency market, i.e., whether prices will increase or decrease. The model is trained using historical data and important technical indicators such as the Moving Average Convergence Divergence, the Relative Strength Index, and Bollinger Bands. We illustrate our approach with an empirical study of the closing price of Bitcoin. Several simulations, including a confusion matrix and Receiver Operating Characteristic curve, are used to assess the model’s performance, and the results show a buy/sell signal accuracy of over 92%. These findings demonstrate how machine learning models can assist investors and traders of cryptocurrencies in making wise/informed decisions in a very volatile market. Copyright © 2024, The Authors. All rights reserved.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Applying NOX Error Mitigation Protocols to Calculate Real-time Quantum Field Theory Scattering Phase Shifts

arXiv

引用

arXiv 2022年

作者： Parks, Zachary Carignan-Dugas, Arnaud Gustafson, Erik Meurice, Yannick Dreher, Patrick Department of Computer Science North Carolina State University RaleighNC27695 United States Keysight Technologies Canada KanataONK2K 2W5 Canada Fermi National Accelerator Laboratory BataviaIL60510 United States Department of Physics and Astronomy The University of Iowa Iowa CityIA52242 United States Department of Electrical and Computer Engineering North Carolina State University RaleighNC27695 United States

Real-time scattering calculations on a Noisy Intermediate Scale Quantum (NISQ) quantum computer are disrupted by errors that accumulate throughout the circuits. To improve the accuracy of such physics simulations, one can supplement the application circuits with a recent error mitigation strategy known as Noiseless Output eXtrapolation (NOX). We tested these error mitigation protocols on a Transverse Field Ising model and improved upon previous phase shift calculations. Our proof-of-concept 4-qubit application circuits were run on several IBM quantum computing hardware architectures. We introduce metrics that show between 22% and 73% error reduction for circuit depths ranging from 13 to 37 hard cycles, confirming that the NOX technique applies to circuits with a broad range of failure rates. We also observed an approximate 28% improvement in the accuracy of the time delay calculation of the scattering phase shift. These observations on different cloud-accessible devices confirm that NOX improves performance even when circuits are executed in substantially time-separated batches. Finally, we provide a heuristic method to obtain systematic error bars on the mitigated results, compare them with empirical errors and discuss their effects on phase shift estimates. © 2022, CC BY-SA.

关键词： Ising model

来源：评论

学校读者我要写书评

暂无评论

Indoor Pm2.5 Forecasting and the Association with Outdoor Air Pollution: A Modelling Study Based on Sensor Data in Australia

SSRN

引用

SSRN 2025年

作者： Yu, Wenhua Nakisa, Bahareh Loke, Seng W. Stevanovic, Svetlana Rastgoo, Mohammad Naim Guo, Yuming Climate Air Quality Research Unit School of Public Health and Preventive Medicine Monash University Level 2 553 St Kilda Road MelbourneVIC3004 Australia School of IT Faculty of Science and Built Environment Deakin University Australia School of Engineering Deakin University Waurn Ponds 3216 Australia School of Computer and Mathematical Sciences University of Adelaide SA Australia Department of Epidemiology and Preventive Medicine

Exposure to poor indoor air quality poses significant health risks, necessitating thorough assessment to mitigate associated dangers. This study aims to predict hourly indoor fine particulate matter (PM2.5) concentrations and investigate their correlation with outdoor PM2.5 levels across 24 distinct buildings in Australia. Indoor air quality data were gathered from 91 monitoring sensors in eight Australian cities spanning 2019 to 2022. Employing an innovative three-stage deep ensemble machine learning framework (DEML), comprising three base models (Support Vector Machine, Random Forest, and eXtreme Gradient Boosting) and two meta-models (Random Forest and Generalized Linear Model), hourly indoor PM2.5 concentrations were predicted. The model's accuracy was evaluated using a rolling windows approach, comparing its performance against three benchmark algorithms (SVM, RF, and XGBoost). Additionally, a SHAP (SHapley Additive exPlanations) analysis was used to assess the relationship between indoor and outdoor PM2.5 concentrations. Results indicate that the DEML model consistently outperformed benchmark models, achieving an R2 ranging from 0.63 to 0.99 and RMSE from 0.01 to 0.66 µg/m3 for most sensors. Notably, outdoor PM2.5 concentrations significantly impacted indoor air quality, particularly at high levels above 15 µg/m3. This study underscores the importance of accurate indoor air quality prediction, crucial for developing location-specific early warning systems and informing effective interventions. © 2025, The Authors. All rights reserved.

关键词： Indoor air pollution

来源：评论

学校读者我要写书评

暂无评论

Progress towards machine learning methodologies for laser-induced breakdown spectroscopy with an emphasis on soil analysis

arXiv

引用

arXiv 2022年

作者： Huang, Yingchao Harilal, Sivanandan S. Bais, Abdul Hussein, Amina E. Faculty of Engineering and Applied Science University of Regina SKS4S 0A2 Canada Pacific Northwest National Laboratory RichlandWA99354 United States Department of Electrical and Computer Engineering University of Alberta EdmontonABT6G 2H5 Canada

Optical emission spectroscopy of laser-produced plasmas, commonly known as laser-induced breakdown spectroscopy (LIBS), is an emerging analytical tool for rapid soil analysis. However, specific challenges with LIBS exist, such as matrix effects and quantification issues, that require further study in the application of LIBS, particularly for analysis of heterogeneous samples such as soils. Advancements in the applications of Machine Learning (ML) methods can address some of these issues, advancing the potential for LIBS in soil analysis. This article aims to review the progress of LIBS application combined with ML methods, focusing on methodological approaches used in reducing matrix effect, feature selection, quantification analysis, soil classification, and self-absorption. The performance of various adopted ML approaches is discussed, including their shortcomings and advantages, to provide researchers with a clear picture of the current status of ML applications in LIBS for improving its analytical capability. The challenges and prospects of LIBS development in soil analysis are proposed, offering a path toward future research. This review article emphasize ML tools for LIBS soil analysis that are broadly relevant for other LIBS applications. © 2022, CC BY-NC-SA.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Refashioning Emotion Recognition Modelling: The Advent of Generalised Large Models

arXiv

引用

arXiv 2023年

作者： Zhang, Zixing Peng, Liyizhe Pang, Tao Han, Jing Zhao, Huan Schuller, Björn W. The College of Computer Science and Electronic Engineering Hunan University Changsha410082 China The Department of Computer Science and Technology University of Cambridge 15 JJ Thomson Ave CambridgeCB3 0FD United Kingdom GLAM - the Group on Language Audio & Music Imperial College London 180 Queen's Gate Huxley Bldg. LondonSW7 2AZ United Kingdom Embedded Intelligence for Health Care and Wellbeing University of Augsburg Eichleitnerstr. 30 Augsburg86159 Germany

After the inception of emotion recognition or affective computing, it has increasingly become an active research topic due to its broad applications. Over the past couple of decades, emotion recognition models have gradually migrated from statistically shallow models to neural network-based deep models, which can significantly boost the performance of emotion recognition models and consistently achieve the best results on different benchmarks. Therefore, in recent years, deep models have always been considered the first option for emotion recognition. However, the debut of large language models (LLMs), such as ChatGPT, has remarkably astonished the world due to their emerged capabilities of zero/few-shot learning, in-context learning, chain-of-thought, and others that are never shown in previous deep models. In the present paper, we comprehensively investigate how the LLMs perform in emotion recognition in terms of diverse aspects, including in-context learning, few-short learning, accuracy, generalisation, and explanation. Moreover, we offer some insights and pose other potential challenges, hoping to ignite broader discussions about enhancing emotion recognition in the new era of advanced and generalised large models. Copyright © 2023, The Authors. All rights reserved.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Multi-user cooperative computation framework based on bertrand game

arXiv

引用

arXiv 2021年

作者： Zhang, Nan Zhang, Guopeng Wang, Kezhi Yang, Kun School of Computer Science and Technology China University of Mining and Technology Xuzhou221116 China Department of Computer and Information Science Northumbria University NewcastleNE2 1XE United Kingdom School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China School of Computer Technology and Engineering Changchun Institute of Technology Changchun China

In this paper, a multi-user cooperative computing framework is applied to enable mobile users to utilize available computing resources from other neighboring users via direct communication links. An incentive scheme based on Bertrand game is proposed for the user to determine who and how to cooperate. We model the resource demand users as buyers who aim to use minimal payments to maximize energy savings, whereas resource supply users as sellers who aim to earn payments for their computing resource provision. A Bertrand game against buyer’s market is formulated. When the users have complete information of their opponents, the Nash equilibrium (NE) of the game is obtained in closed form, while in the case of incomplete information, a distributed iterative algorithm is proposed to find the NE. The simulation results verify the effectiveness of the proposed scheme. Copyright © 2021, The Authors. All rights reserved.

关键词： Game theory

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：