检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Lin, Yi Liu, Yanfei Chen, Hao Yang, Xin Ma, Kai Zheng, Yefeng Cheng, Kwang-Ting The Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong The Department of Chemical and Biological Engineering The Hong Kong University of Science and Technology Hong Kong Shenzhen United Imaging Research Institute of Innovative Medical Equipment Real-Time Laboratory Shenzhen518048 China The Department of Electronic Information and Communications Huazhong University of Science and Technology Wuhan430074 China Jarvis Research Center Tencent YouTu Lab Shenzhen518057 China

Radiation therapy treatment planning requires balancing the delivery of the target dose while sparing normal tissues, making it a complex process. To streamline the planning process and enhance its quality, there is a growing demand for knowledge-based planning (KBP). Ensemble learning has shown impressive power in various deep learning tasks, and it has great potential to improve the performance of KBP. However, the effectiveness of ensemble learning heavily depends on the diversity and individual accuracy of the base learners. Moreover, the complexity of model ensembles is a major concern, as it requires maintaining multiple models during inference, leading to increased computational cost and storage overhead. In this study, we propose a novel learning-based ensemble approach named LENAS, which integrates neural architecture search with knowledge distillation for 3D radiotherapy dose prediction. Our approach starts by exhaustively searching each block from an enormous architecture space to identify multiple architectures that exhibit promising performance and significant diversity. To mitigate the complexity introduced by the model ensemble, we adopt the teacher-student paradigm, leveraging the diverse outputs from multiple learned networks as supervisory signals to guide the training of the student network. Furthermore, to preserve high-level semantic information, we design a hybrid-loss to optimize the student network, enabling it to recover the knowledge embedded within the teacher networks. The proposed method has been evaluated on two public datasets, OpenKBP and AIMIS. Extensive experimental results demonstrate the effectiveness of our method and its superior performance to the state-of-the-art methods. Code: ***/hust-linyi/LENAS. © 2021, CC BY.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Quantum generative training using rényi divergences

arXiv

引用

arXiv 2021年

作者： Kieferová, Mária Marrero, Carlos Ortiz Wiebe, Nathan Centre for Quantum Computation and Communication Technology Centre for Quantum Software and Information University of Technology Sydney Data Sciences & Analytics Group Department of Electrical & Computer Engineering Pacific Northwest National Laboratory North Carolina State University Department of Computer Science Department of Physics High Performance Computing Group University of Toronto University of Washington Pacific Northwest National Laboratory

Quantum neural networks (QNNs) are a framework for creating quantum algorithms that promises to combine the speedups of quantum computation with the widespread successes of machine learning. A major challenge in QNN development is a concentration of measure phenomenon known as a barren plateau that leads to exponentially small gradients for a range of QNNs models. In this work, we examine the assumptions that give rise to barren plateaus and show that an unbounded loss function can circumvent the existing no-go results. We propose a training algorithm that minimizes the maximal Rényi divergence of order two and present techniques for gradient computation. We compute the closed form of the gradients for Unitary QNNs and Quantum Boltzmann Machines and provide sufficient conditions for the absence of barren plateaus in these models. We demonstrate our approach in two use cases: thermal state learning and Hamiltonian learning. In our numerical experiments, we observed rapid convergence of our training loss function and frequently archived a 99% average fidelity in fewer than 100 epochs. © 2021, CC BY.

关键词： Quantum computers

来源：评论

学校读者我要写书评

暂无评论

Scalable flux controllers using adiabatic superconductor logic for quantum processors

引用

Physical Review Research 2023年第1期5卷 013145-013145页

作者： Naoki Takeuchi Taiki Yamae Wenhui Luo Fuminori Hirayama Tsuyoshi Yamamoto Nobuyuki Yoshikawa Research Center for Emerging Computing Technologies National Institute of Advanced Industrial Science and Technology (AIST) Tsukuba Ibaraki 305-8568 Japan NEC-AIST Quantum Technology Cooperative Research Laboratory National Institute of Advanced Industrial Science and Technology (AIST) Tsukuba Ibaraki 305-8568 Japan Department of Electrical and Computer Engineering Yokohama National University Yokohama Kanagawa 240-8501 Japan Research Fellow of Japan Society for the Promotion of Science Chiyoda Tokyo 102-0083 Japan Secure System Platform Research Laboratories NEC Corporation Kawasaki Kanagawa 211-0011 Japan Institute of Advanced Sciences Yokohama National University Yokohama Kanagawa 240-8501 Japan

Quantum processors have the potential to accelerate specific computing tasks but are difficult to scale up due to engineering limitations, such as the number of available cables and cooling power for dilution refrigerators for quantum bits (qubits). Hence, it is very important to develop scalable, energy-efficient interface circuits that can control many qubits via a few control lines inside a dilution refrigerator. One of the most important interface circuits is the flux controller (FC), which generates arbitrary dc flux bias to adjust the characteristics of component devices such as qubits. In this paper, we propose and demonstrate FCs using an energy-efficient superconductor logic family, adiabatic quantum-flux-parametron (AQFP) logic. We develop two types of FCs: the AQFP FC and the AQFP/single-flux-quantum (AQFP/SFQ) FC. Both FCs require only a few control lines and have extremely small power dissipation, thus exhibiting high scalability. Furthermore, the AQFP/SFQ FC can control flux bias using ballistic SFQ transmission, which is crucial for integration with qubits. As a proof of concept, we demonstrate AQFP and AQFP/SFQ FCs at 4.2 K, fabricated by the AIST high-speed standard process. Our results indicate that AQFP logic is highly suitable for use as qubit interface circuits for very large-scale quantum processors, especially from the viewpoint of the control line count, power dissipation, and amount of supply currents.

关键词： engineering Devices for digital logic, storage & processing Josephson junctions SQUID Superconducting devices

来源：评论

学校读者我要写书评

暂无评论

Finite-Size Analysis of Prepare-and-Measure and Decoy-State Quantum Key Distribution via Entropy Accumulation

引用

PRX Quantum 2025年第2期6卷 020342-020342页

作者： Lars Kamin Amir Arqand Ian George Norbert Lütkenhaus Ernest Y.-Z. Tan Institute for Quantum Computing and Department of Physics and Astronomy University of Waterloo Waterloo Ontario N2L 3G1 Canada Department of Electrical and Computer Engineering Coordinated Science Laboratory University of Illinois at Urbana-Champaign Urbana Illinois 61801 USA

An important goal in quantum key distribution (QKD) is the task of providing a finite-size security proof without the assumption of collective attacks. For prepare-and-measure QKD, one approach for obtaining such proofs is the generalized entropy-accumulation theorem (GEAT) but thus far it has only been applied to study a small selection of protocols. In this work, we present techniques for applying the GEAT in finite-size analysis of generic prepare-and-measure protocols, with a focus on decoy-state protocols. In particular, we present an improved approach for computing entropy bounds for decoy-state protocols, which has the dual benefits of providing tighter bounds than previous approaches (even asymptotically) and being compatible with methods for computing min-trade-off functions in the GEAT. Furthermore, we develop methods to incorporate some improvements to the finite-size terms in the GEAT and implement techniques to automatically optimize the min-trade-off function. Our approach also addresses some numerical-stability challenges specific to prepare-and-measure protocols, which have not been addressed in previous works.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Intrusion detection using machine learning techniques: An experimental comparison

arXiv

引用

arXiv 2021年

作者： Tait, Kathryn-Ann Khan, Jan Sher Alqahtani, Fehaid Shah, Awais Aziz Khan, Fadia Ali Rehman, Mujeeb Ur Boulila, Wadii Ahmad, Jawad School of Computing Edinburgh Napier University United Kingdom Department of Electrical and Electronics University of Gaziantep Gaziantep27310 Turkey Department of Computer Science King Fahad Naval Academy Al Jubail Saudi Arabia Politecnico di Bari Italy Department of Electrical Engineering Riphah International University Islamabad Pakistan RIADI Laboratory University of Manouba Manouba2010 Tunisia College of Computer Science and Engineering Taibah University Medina42353 Saudi Arabia

Due to an exponential increase in the number of cyber-attacks, the need for improved Intrusion Detection Systems (IDS) is apparent than ever. In this regard, Machine Learning (ML) techniques are playing a pivotal role in the early classification of the attacks in case of intrusion detection within the system. However, due to the large number of algorithms available, the selection of the right method is a challenging task. To resolve this issue, this paper analyses some of the current state of the art intrusion detection methods and discusses their pros and cons. Further, a review of different ML methods is carried out with four methods showing to be the most suitable one for classifying attacks. Several algorithms are selected and investigated to evaluate the performance of IDS. These IDS classifies binary and multiclass attacks in terms of detecting whether or not the traffic has been considered as benign or an attack. The experimental results demonstrate that binary classification has greater consistency in their accuracy results which ranged from 0.9938 to 0.9977, while multiclass ranges from 0.9294 to 0.9983. However, it has been also observed that multiclass provides the best results with the algorithm k-Nearest neighbor giving an accuracy score of 0.9983 while the binary classification highest score is 0.9977 from Random Forest. The experimental results demonstrate that multiclass classification produces better performance in terms of intrusion detection by specifically differentiating between the attacks and allowing a more targeted response to an attack. © 2021, CC BY.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Outage Performance of Uplink Rate Splitting Multiple Access with Randomly Deployed Users

arXiv

引用

arXiv 2022年

作者： Lu, Huabing Xie, Xianzhong Shi, Zhaoyuan Lei, Hongjiang Zhao, Nan Cai, Jun The Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education Dalian University of Technology Dalian116024 China The School of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing400065 China The Key Laboratory of Intelligent Perception and Computing of Anhui Province Anqing Normal University Anqing246011 China The School of Communication and Information Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China Department of Electrical and Computer Engineering Concordia University MontrealQCH3G 1M8 Canada

With the rapid proliferation of smart devices in wireless networks, more powerful technologies are expected to fulfill the network requirements of high throughput, massive connectivity, and diversify quality of service. To this end, rate splitting multiple access (RSMA) is proposed as a promising solution to improve spectral efficiency and provide better fairness for the next-generation mobile networks. In this paper, the outage performance of uplink RSMA transmission with randomly deployed users is investigated, taking both user scheduling schemes and power allocation strategies into consideration. Specifically, the greedy user scheduling (GUS) and cumulative distribution function (CDF) based user scheduling (CUS) schemes are considered, which could maximize the rate performance and guarantee scheduling fairness, respectively. Meanwhile, we re-investigate cognitive power allocation (CPA) strategy, and propose a new rate fairness-oriented power allocation (FPA) strategy to enhance the scheduled users' rate fairness. By employing order statistics and stochastic geometry, an analytical expression of the outage probability for each scheduling scheme combining power allocation is derived to characterize the performance. To get more insights, the achieved diversity order of each scheme is also derived. Theoretical results demonstrate that both GUS and CUS schemes applying CPA or FPA strategy can achieve full diversity orders, and the application of CPA strategy in RSMA can effectively eliminate the secondary user's diversity order constraint from the primary user. Simulation results corroborate the accuracy of the analytical expressions, and show that the proposed FPA strategy can achieve excellent rate fairness performance in high signal-to-noise ratio region. Copyright © 2022, The Authors. All rights reserved.

关键词： Wireless networks

来源：评论

学校读者我要写书评

暂无评论

How to Build a Quantum Supercomputer: Scaling Challenges and Opportunities

arXiv

引用

arXiv 2024年

作者： Mohseni, Masoud Scherer, Artur Johnson, K. Grace Wertheim, Oded Otten, Matthew Aadit, Navid Anjum Bresniker, Kirk M. Camsari, Kerem Y. Chapman, Barbara Chatterjee, Soumitra Dagnew, Gebremedhin A. Esposito, Aniello Fahim, Farah Fiorentino, Marco Khalid, Abdullah Kong, Xiangzhou Kulchytskyy, Bohdan Li, Ruoyu Lott, P. Aaron Markov, Igor L. McDermott, Robert F. Pedretti, Giacomo Gajjar, Archit Silva, Allyson Sorebo, John Spentzouris, Panagiotis Steiner, Ziv Torosov, Boyan Venturelli, Davide Visser, Robert J. Webb, Zak Zhan, Xin Cohen, Yonatan Ronagh, Pooya Ho, Alan Beausoleil, Raymond G. Martinis, John M. Hewlett Packard Labs CA United States BC Canada Quantum Machines Israel Department of Physics University of Wisconsin–Madison WI United States Department of Electrical and Computer Engineering University of California Santa BarbaraCA United States Hewlett Packard Enterprise TX United States Fermi National Accelerator Laboratory IL United States Applied Materials CA United States USRA Research Institute for Advanced Computer Science CA United States NASA Ames Research Center CA United States Synopsys CA United States Qolab CA United States Institute for Quantum Computing University of Waterloo ON Canada Department of Physics & Astronomy University of Waterloo ON Canada Perimeter Institute for Theoretical Physics ON Canada

In the span of four decades, quantum computation has evolved from an intellectual curiosity to a potentially realizable technology. Today, small-scale demonstrations have become possible for quantum algorithmic primitives on hundreds of physical qubits and proof-of-principle error-correction procedures on a single logical qubit. Nevertheless, despite significant progress and excitement, the detailed path toward a full-stack scalable quantum computing technology is largely unknown. There are significant outstanding quantum hardware, fabrication, software architecture, and algorithmic challenges that are either unresolved or overlooked. These issues could seriously undermine the arrival of utility-scale quantum computers for the foreseeable future. Here, we provide a comprehensive review of these scaling challenges. We show how the road to scaling could be paved by adopting existing semiconductor technology to build much higher-quality qubits, employing system engineering approaches, and performing distributed quantum computation within heterogeneous high-performance computing infrastructures. These opportunities for research and development could unlock certain promising applications, in particular, efficient quantum simulation and learning/modeling quantum data generated by natural or engineered quantum systems. In order to estimate the true cost of such promises, we provide a detailed resource and sensitivity analysis for classically hard quantum chemistry calculations on surface-code error-corrected quantum computers given current, target, and desired hardware specifications based on superconducting qubits, accounting for a realistic distribution of errors. We show orders of magnitude enhancement in performance could be obtained by a combination of quantum hardware and algorithmic improvements. Furthermore, we argue that, to tackle today’s industry-scale classical optimization and machine learning problems in a cost-effective manner, distributed quantum-assisted p

关键词： Qubits

来源：评论

学校读者我要写书评

暂无评论

Data-driven Industrial Robot Arm Calibration: A Machine Learning Perspective 18

Data-driven Industrial Robot Arm Calibration: A Machine Lear...

引用

18th IEEE International Conference on Networking, Sensing and Control, ICNSC 2021

作者： Li, Zhibin Li, Shuai Luo, Xin University of Chinese Academy of Sciences School of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences Chongqing School Chongqing China Swansea University Department of Electronics and Electrical Engineering Swansea United Kingdom University of Chinese Academy of Sciences Chongqing Key Laboratory of Big Data and Intelligent Computing Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences Chongqing School Chongqing China

ISBN: (纸本)9781665440486

Robot arms have been widely used in industry. The absolute positioning error of robots without calibration can reach several millimeters, which cannot meet the application requirements of accurate operation. Therefore, it is almost a mandatory procedure for industrial robots to take on-site calibration before being used. Generally, most researchers on robot calibration have mechanical and instrumentation background as the collection of calibration data is tedious and it is usually difficult to access to industrial robots for researchers in other fields. This research explores the calibration problem from a machine learning perspective and provides the first open-access dataset called "RobotCali"in this area so that machine learning scientists can step into this field and verify their algorithms on this problem. In the meanwhile, a new calibration method based on the Levenberg-Marquardt (LM) algorithm and extended Kalman filter (EKF) algorithm is proposed, which can significantly improve the absolute positioning accuracy of the robot after calibration. Firstly, the error model of robot is established, and kinematic parameters are initially identified by LM algorithm. Then the EKF algorithm is used to further calibrate these parameters, which has been verified the effectiveness of the proposed method by experimental results. Lastly, the future research work is discussed. © 2021 IEEE.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Cramér-rao bound optimization for joint radar-communication design

arXiv

引用

arXiv 2021年

作者： Liu, Fan Liu, Ya-Feng Li, Ang Masouros, Christos Eldar, Yonina C. The Department of Electrical and Electronic Engineering Southern University of Science and Technology Shenzhen518055 China The State Key Laboratory of Scientific and Engineering Computing Institute of Computational Mathematicsand Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China The School of Information and Communications Engineering Faculty of Electronic and Information Engineering Xi’an Jiaotong University Xi’anShaanxi710049 China The Department of Electronic and Electrical Engineering University College London LondonWC1E 7JE United Kingdom The Faculty of Mathematics and Computer Science Weizmann Institute of Science Rehovot Israel

In this paper, we propose multi-input multi-output (MIMO) beamforming designs towards joint radar sensing and multi-user communications. We employ the Cramér-Rao bound (CRB) as a performance metric of target estimation, under both point and extended target scenarios. We then propose minimizing the CRB of radar sensing while guaranteeing a pre-defined level of signal-to-interference-plus-noise ratio (SINR) for each communication user. For the single-user scenario, we derive a closed form for the optimal solution for both cases of point and extended targets. For the multi-user scenario, we show that both problems can be relaxed into semidefinite programming by using the semidefinite relaxation approach, and prove that the global optimum can always be obtained. Finally, we demonstrate numerically that the globally optimal solutions are reachable via the proposed methods, which provide significant gains in target estimation performance over state-of-the-art benchmarks. Copyright © 2021, The Authors. All rights reserved.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

A systematic view of leakage risks in deep neural network systems

TechRxiv

引用

TechRxiv 2021年

作者： Hu, Xing Liang, Ling Chen, Xiaobing Deng, Lei Ji, Yu Ding, Yufei Du, Zidong Guo, Qi Sherwood, Timothy Xie, Yuan State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China Department of Electrical and Computer Engineering University of California Santa Barbara United States Department of Computer Science University of California Santa Barbara United States Tsinghua University China University of California Santa Barbara United States

As deep neural networks (DNNs) continue their reach into a wide range of application domains, the neural network architecture of DNN models becomes an increasingly sensitive subject, due to either intellectual property protection or risks of adversarial attacks. In observing the large gap between the architectural surfaces exploration and the model integrity study, this paper first presents the formulated schema of the model leakage risks. Then, we propose DeepSniffer, a learning-based model extraction framework, to obtain the complete model architecture information without any prior knowledge of the victim model. It is robust to architectural and system noises introduced by the complex memory hierarchy and diverse run-time system optimizations. Taking GPU platforms as a showcase, DeepSniffer performs model extraction by learning both the architecture-level execution features of kernels and the inter-layer temporal association information introduced by the common practice of DNN design. We demonstrate that DeepSniffer works experimentally in the context of an off-the-shelf Nvidia GPU platform running a variety of DNN models. The extracted models are directly helpful to the attempting of crafting adversarial inputs. The DeepSniffer project has been released in https://***/xinghu7788/DeepSniffer. © 2021, CC0.

关键词： Deep 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：