检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Tavakkoli, Fatemeh Sarhadi, Pouria Clement, Benoit Naeem, Wasif The Faculty of Electrical and Computer Engineering Department Babol Noshirvani University of Technology Babol Iran The School of Physics Engineering and Computer Science University of Hertfordshire Hertfordshire HatfieldAL10 9AB United Kingdom ENSTA Bretagne CROSSING ILR CNRS 2010 Flinders University Australia The School of Electronics Electrical Engineering and Computer Science Queen’s University Belfast United Kingdom

Deep Reinforcement Learning (DRL) techniques have received significant attention in control and decision-making algorithms. Most applications involve complex decision-making systems, justified by the algorithms' computational power and cost. While model-based versions are emerging, model-free DRL approaches are intriguing for their independence from models, yet they remain relatively less explored in terms of performance, particularly in applied control. This study conducts a thorough performance analysis comparing the data-driven DRL paradigm with a classical state feedback controller, both designed based on the same cost (reward) function of the linear quadratic regulator (LQR) problem. Twelve additional performance criteria are introduced to assess the controllers' performance, independent of the LQR problem for which they are designed. Two Deep Deterministic Policy Gradient (DDPG)-based controllers are developed, leveraging DDPG's widespread reputation. These controllers are aimed at addressing a challenging setpoint tracking problem in a Non-Minimum Phase (NMP) system. The performance and robustness of the controllers are assessed in the presence of operational challenges, including disturbance, noise, initial conditions, and model uncertainties. The findings suggest that the DDPG controller demonstrates promising behavior under rigorous test conditions. Nevertheless, further improvements are necessary for the DDPG controller to outperform classical methods in all criteria. While DRL algorithms may excel in complex environments owing to the flexibility in the reward function definition, this paper offers practical insights and a comparison framework specifically designed to evaluate these algorithms within the context of control engineering. © 2024, CC BY.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

A Multi-Task Multiple Instance Learning Algorithm to Analyze Large Whole Slide Images from Bright Challenge 2022

A Multi-Task Multiple Instance Learning Algorithm to Analyze...

引用

Biomedical Imaging Challenges (ISBIC), IEEE International Symposium on

作者： Niccolò Marini Marek Wodzinski Manfredo Atzori Henning Mü ller University of Applied Sciences Western Switzerland Information Systems Institute Sierre Switzerland Department of Computer Science University of Geneva Switzerland Department of Measurement and Electronics AGH University of Science and Technology Krakow Poland Department of Neurosciences University of Padua Italy Medical Faculty University of Geneva Switzerland

ISBN: (数字)9781665451727

ISBN: (纸本)9781665451734

Malignant lesions in breast tissue specimen whole slide images (WSIs), may lead to a dangerous diagnosis, such as cancer. However, WSIs analysis is time-consuming and expensive, requiring the work of expert pathologists. This paper aims to present a method for the 2022 BRIGHT Challenge, that involves the analysis of breast WSIs. The organizers provided over 550 breast WSIs and over 3900 regions of interest (ROIs) to develop and validate methods for breast cancer images. The method presented in this work is based on a Multiple Instance Learning instance-based Convolutional Neural Network (CNN), allowing the combination of strongly-annotated data (from ROIs) and weakly-annotated data (from WSIs) via the optimization of a multi-task loss function. Furthermore, during the CNN training, the input patches are clustered and filtered according to their entropy, to reduce the non-informative content used to train the model. The CNN reaches an averaged $\text{F1-score}=0.63\pm 0.02$ on the 3-class classification task and averaged $\text{F1-score}=0.39 \pm 0.08$ on the 6-class classification task, considering the val-idation partition; an averaged $\text{F1-score}=0.65$ on the cancer risk classification task and averaged $\text{F1-score}=0.45$ on the sub-typing cancer risk classification task, considering the best result achieved on the test partition. These results show that Multiple Instance Learning instance-based CNNs may represent a good resource to tackle this kind of problem.

关键词： Training Breast tissue Multitasking Entropy Partitioning algorithms Classification algorithms Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Oracle problems as communication tasks and optimization of quantum algorithms

arXiv

引用

arXiv 2024年

作者： Te’eni, Amit Schwartzman-Nowik, Zohar Nowakowski, Marcin Horodecki, Pawel Cohen, Eliahu Faculty of Engineering The Institute of Nanotechnology and Advanced Materials Bar Ilan University Ramat Gan Israel Department of Computer Science and Engineering Hebrew University of Jerusalem Jerusalem Israel Faculty of Applied Physics and Mathematics Gdansk University of Technology Gdansk80-952 Poland National Quantum Information Center Gdansk80-309 Poland ObserveQ Gdansk80-890 Poland

Quantum query complexity mainly studies the number of queries needed to learn some property of a black box with high probability. A closely related question is how well an algorithm can succeed with this learning task using only a fixed number of queries. In this work, we propose measuring an algorithm’s performance using the mutual information between the output and the actual value. A key observation is that if an algorithm is only allowed to make a single query and the goal is to optimize this mutual information, then we obtain a task which is similar to a basic task of quantum communication, where one attempts to maximize the mutual information of the sender and receiver. We make this analogy precise by formally considering the oracle as a separate subsystem, whose state records the unknown oracle identity. The oracle query prepares a state, which is then measured;and the target property of the oracle plays the role of a message that should be deduced from the measurement outcome. Thus we obtain a link between the optimal single-query algorithm and minimization of the extent of quantum correlations between the oracle and the computer subsystems. We also find a lower bound on this mutual information, which is related to quantum coherence. These results extend to multiple-query non-adaptive algorithms. As a result, we gain insight into the task of finding the optimal non-adaptive algorithm that uses at most a fixed number of queries, for any oracle problem. © 2024, CC BY.

关键词： Quantum communication

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Machine Learning Models with Different Types of Data Representations for Detecting Malicious Files 4

Comparative Analysis of Machine Learning Models with Differe...

引用

4th International Scientific Conference "Information technology and Implementation", IT and I 2022

作者： Nafiiev, Alan Kholodulkin, Hlib Rodionov, Andrii Lande, Dmytro National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" Institute of Physics and Technology Kyiv Ukraine Taras Shevchenko National University of Kyiv Faculty of Computer Science and Cybernetics Kyiv Ukraine Institute for Information Recording of National Academy of Sciences of Ukraine Kyiv Ukraine

Nowadays, malicious software authors are creating more and more advanced, sophisticated malware. To detect such programs, machine learning methods are increasingly used. However, these solutions can consume a lot of computing resources to perform their operations. Therefore, the problem arises of creating an optimal machine learning model regarding learning rate and the accuracy of malware detection. Also, usually one method is not enough for high-quality file detection, so it is more efficient to use several types of data representation. The purpose of this work is to conduct research on several machine learning models based on the support vector machine. Compare them with each other and identify the best model for two different types of data representation. The set of data used was collected from various Internet sources. It consists of 12824 executable files in .exe format, 11844 of which are malicious and 980 are benign. This article presents recommended methods for feature selection and input data generation for a machine learning model. These methods allow you to find the best option for preparing features that describe a malicious file, which will be used in the process of training and determining the model with the best parameters. © 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Onedata4Sci: Life science data management solution based on Onedata

arXiv

引用

arXiv 2023年

作者： Svoboda, Tomáš Raček, Tomáš Handl, Josef Sabo, Jozef Rošinec, Adrián Opiola, Lukasz Jesionek, Wojciech Ešner, Milan Pernisová, Markéta Valasevich, Natallia Madzia Křenek, Aleš Svobodová, Radka CEITEC Central European Institute of Technology Kamenice 753/5 Brno625 00 Czech Republic National Centre for Biomolecular Research Faculty of Science Masaryk University Kamenice 753/5 Brno625 00 Czech Republic Institute of Computer Science Masaryk University Botanická 68a Brno602 00 Czech Republic ACC Cyfronet AGH AGH University of Science and Technology Nawojki 11 Kraków30-950 Poland

Life-science experimental methods generate vast and ever-increasing volumes of data, which provide highly valuable research resources. However, management of these data is nontrivial and applicable software solutions are currently subject to intensive development. The solutions mainly fall into one of the two groups: general data management systems (e.g. Onedata, iRODS, B2SHARE, CERNBox) or very specialised data management solutions (e.g. solutions for biomolecular simulation data, biological imaging data, genomic data). To bridge this gap between them, we provide Onedata4Sci, a prototype data management solution, which is focused on the management of life science data and covers four key steps of the data life cycle, i.e. data acquisition, user access, computational processing and archiving. Onedata4Sci is based on the Onedata data management system. It is written in Python, fully containerised, with the support for processing the stored data in Kubernetes. The applicability of Onedata4Sci is shown in three distinct use cases – plant imaging data, cellular imaging data, and cryo-electron microscopy data. Despite the use cases covering very different types of data and user patterns, Onedata4Sci demonstrated an ability to successfully handle all these conditions. Complete source codes of Onedata4Sci are available on GitHub (https://***/CERIT-SC/onedata4sci), and its documentation and manual for installation are also provided. © 2023, CC BY-SA.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

NOMA-Based Ze-RIS Empowered Backscatter Communication with Energy-Efficient Resource Management

引用

IEEE Transactions on Communications 2025年

作者： Asif, Muhammad Bao, Xu Ihsan, Asim Khan, Wali Ullah Li, Xingwang Chatzinotas, Symeon Dobre, Octavia A. The School of Computer Science and Communication Engineering Jiangsu University Zhenjiang China The Department of Electrical Engineering University of Cambridge Cambridge United Kingdom University of Luxembourg Luxembourg City1855 Luxembourg The School of Physics and Electronic Information Engineering Henan Polytechnic University the Jiaozuo Key Laboratory of Crow-Sensing Network Jiaozuo454003 China The Department of Electrical and Computer Engineering Faculty of Engineering and Applied Science Memorial University St. John’sNLA1B 3X9 Canada

This manuscript introduces a novel energy-efficient optimization strategy for a zero-energy reconfigurable intelligent reflecting surface (Ze-RIS) supported backscatter communication system employing non-orthogonal multiple access (NOMA). The central objective is to maximize the energy efficiency of the system by optimizing the several key parameters, including the amplitude reflection coefficient of Ze-RIS, the reflection coefficients of the backscatter tags, transmit beamforming at the base station, and passive beamforming at the Ze-RIS node, while incorporating a practical non-linear energy harvesting model both for the Ze-RIS and backscatter nodes. The proposed algorithm addresses the complex non-convex problem through three stages. Firstly, the transmit beamforming vectors are determined by leveraging the semi-definite programming and successive-convex approximation, while handling the rank-1 constraint with the semi-definite relaxation. Secondly, we determine the amplitude reflection coefficient of Ze-RIS by leveraging the monotonicity property of the objective function. Simultaneously, we compute the reflection coefficients of backscatter tags using the Dinkelbach algorithm, Lagrange duality, and the sub-gradient method. Thirdly, we compute passive beamforming using successive-convex approximation and semi-definite programming techniques, achieving a rank-1 solution through the penalty-based method. Finally, the numerical simulations confirm the effectiveness of the proposed approach, demonstrating its superiority over the benchmark competitors with rapid convergence within a few iterations. © 1972-2012 IEEE. All rights reserved.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

Applying oculography and biosignals measurements to gameplay modes analysis 1st

Applying oculography and biosignals measurements to gameplay...

引用

1st International Conference on Human Interaction and Emerging Technologies, IHIET 2019

作者： Grabska-Gradzińska, Iwona Nowak, Leszek A. Argasiński, Jan K. Faculty of Physics Astronomy and Applied Computer Science Jagiellonian University Kraków Poland

ISBN: (纸本)9783030256289

In this paper we present results of an experiment designed to measure correlation between Galvanic Skin Response – GSR (also: Electrodermal Activity – EDA) and Oculographic (Eye Tracking) data. In the experiment we analyze the participants reactions on changing gameplay modes (changing dynamic and pace of user interaction with the game mechanics). The goal of the analysis is to demonstrate what influence of changing gameplay modes have on physiology (especially spontaneous body response). The gameplay modes used in experiment were chase mode (dynamic) and sneak mode (satisfying). In the experiment we have used Pupil Lab eye-tracker and EDA sensor based on BITalino platform. The game Serial Cleaner by iFun4all was used to conduct the experiment. The experiment resulted in multiple observations. Substantial correlation between participant failure and change of EDA pattern were observed as well as over 10% change in fixations durations. © 2020, Springer Nature Switzerland AG.

关键词： Eye tracking

来源：评论

学校读者我要写书评

暂无评论

Novel Size-Reduced Coaxial Cavity Filters For 5G Networks

Novel Size-Reduced Coaxial Cavity Filters For 5G Networks

引用

IEEE Asia-Pacific Conference on Antennas and Propagation (APCAP)

作者： Tran Thi Thu Huong An Tri Bao Trong-Hieu Le Vu Van Yem Faculty of Electronics and Computer Engineering University of Economics - Technology for Industries Hanoi Vietnam 11 Physics 2 Hanoi - Amsterdam High School for The Gifted Hanoi Vietnam Faculty of Electronics and Telecommunications Electric Power University Hanoi Vietnam School of Electrical and Electronic Engineering Hanoi University of Science and Technology Hanoi Vietnam

ISBN: (数字)9798350351019

ISBN: (纸本)9798350351026

This paper presents a novel coaxial cavity filter using in 5G NR Base Stations. The resonant cylinder utilizes stepped impedance technology combined with folding edge bending and rectangular zigzag arm structure on the top to improve the electrical coupling mainline. Therefore, dimensions of cavity filter are reduced while maintaining the filter performance parameters. This proposed resonator is verified in the design of one filter integrated in C-Band 5G station with ten orders and four zero points. The dimensions of the filter are only 56.81 mm x 15.56 mm x 9.5 mm (length x width x height). The simulation results show the passband of 304MHz, the return loss of 15.6dB, the insertion loss of 0.1dB. These specifications are suitable for applying in Band C 5G network.

关键词： Couplings 5G mobile communication C-band Simulation Resonator filters Resonant frequency Insertion loss Cavity resonators Passband Impedance

来源：评论

学校读者我要写书评

暂无评论

Probing UHECR and cosmic ray ensemble scenarios with a global CREDO network 37

Probing UHECR and cosmic ray ensemble scenarios with a globa...

引用

37th International Cosmic Ray Conference, ICRC 2021

作者： Tursunov, Arman Homola, Piotr Alvarez Castillo, David E. Budnev, Nikolai Gupta, Alok C. Hnatyk, Bohdan Kasztelan, Marcin Kovacs, Peter Lozowski, Bartosz Medvedev, Mikhail V. Mozgova, Alona Niedźwiecki, Michal Pawlik, Maciej Rosas, Matías Rzecki, Krzysztof Smelcerz, Katarzyna Smolek, Karel Stasielak, Jaroslaw Stuglik, Slawomir Manana, Svanidze Sushchov, Oleksandr Verbetsky, Yuri Wibig, Tadeusz Zamora-Saa, Jilberto Research Centre for Theoretical Physics and Astrophysics Institute of Physics Silesian University in Opava Bezručovo nám. 13 OpavaCZ-74601 Czech Republic Bogoliubov Laboratory for Theoretical Physics Joint Institute for Nuclear Research Joliot-Curie street 6 Dubna141980 Russia Institute of Nuclear Physics Polish Academy of Sciences Radzikowskiego 152 Kraków31-342 Poland Irkutsk State University Russia Manora Peak Nainital263001 India Astronomical Observatory of Taras Shevchenko National University of Kyiv Kyiv04053 Ukraine National Centre for Nuclear Research Andrzeja Soltana 7 Otwock-Świerk05-400 Poland Institute for Particle and Nuclear Physics Wigner Research Centre for Physics Konkoly-Thege Miklós út 29-33 Budapest1121 Hungary Faculty of Natural Sciences University of Silesia in Katowice Bankowa 9 Katowice40-007 Poland Department of Physics and Astronomy University of Kansas LawrenceKS66045 United States Laboratory for Nuclear Science Massachusetts Institute of Technology CambridgeMA02139 United States Department of Computer Science Faculty of Computer Science and Telecommunications Cracow University of Technology Warszawska 24 Kraków31-155 Poland ACC Cyfronet AGH-UST Kraków30-950 Poland AGH University of Science and Technology Mickiewicz Ave. Kraków30-059 Poland Liceo 6 Francisco Bauzá Montevideo Uruguay Institute of Experimental and Applied Physics Czech Technical University in Prague E. Andronikashvili Institute of Physics under Tbilisi State University Georgia University of Lódź Faculty of Physics and Applied Informatics 90-236 Pomorska Lódź149/153 Poland Universidad Andres Bello Departamento de Ciencias Fisicas Facultad de Ciencias Exactas Avenida Republica 498 Santiago Chile

Among theoretical approaches in unveiling the physics of ultra-high-energy cosmic rays (UHECR) one can distinguish the models assuming interactions of exotic super-heavy matter (including extra dimensions, Lorentz invariance violation, cosmic strings, dark matter particles or particles beyond the standard model etc.) and acceleration scenarios describing processes, in which the particles are accelerated by a particular astrophysical object (shocks in relativistic plasma jets, unipolar induction mechanisms, second-order Fermi acceleration, energy transfer from black holes or compact stars etc.). Primary UHECR particles can produce cascades already above the Earth atmosphere, which may be detected as the so-called cosmic ray ensembles (CRE) - the phenomena composed of at least two cosmic ray particles, including photons, with a common primary interaction vertex or the same parent particle with correlated arrival directions and arrival times. In this contribution, we give a brief description of a novel approach to the probing of cosmic ray scenarios with the global Cosmic Ray Extremely Distributed Observatory (CREDO) network. © Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0)

关键词： Cosmic rays

来源：评论

学校读者我要写书评

暂无评论

Strong-coupling superconductivity of SrIr2 and SrRh2: Phonon engineering of metallic Ir and Rh

arXiv

引用

arXiv 2021年

作者： Gutowska, Sylwia Górnicka, Karolina Wójcik, Pawel Klimczuk, Tomasz Wiendlocha, Bartlomiej Faculty of Physics and Applied Computer Science AGH University of Science and Technology Aleja Mickiewicza 30 Krakow30-059 Poland Faculty of Applied Physics and Mathematics and Advanced Materials Centre Gdansk University of Technology ul. Narutowicza 11/12 Gdá nsk80-233 Poland

Experimental and theoretical studies on superconductivity in SrIr2 and SrRh2 Laves phases are presented. The measured resistivity, heat capacity, and magnetic susceptibility confirm the superconductivity of these compounds with Tc = 6.07 K and 5.41 K, respectively. Electronic structure calculations show that the Fermi surface is mostly contributed by 5d (4d) electrons of Ir (Rh), with Sr atoms playing the role of electron donors. The effect of the spin-orbit coupling is analyzed and found to be important in both materials. Lattice dynamics and electron-phonon coupling (EPC) are studied and the strong electron-phonon interaction is found, contributed mostly by the low-frequency Ir and Rh vibrations. The enhancement of EPC when compared to weakly-coupled metallic Ir and Rh is explained by the strong modifications in the propagation of phonons in the network of Ir (Rh) tetrahedrons, which are the building blocks of the Laves phase, and originate from the metallic fcc structures of elemental iridium and rhodium. Copyright © 2021, The Authors. All rights reserved.

关键词： Electron-phonon interactions

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：