检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Li, Yi Wang, Songqi Zhao, Yaping Wang, Shaocong Zhang, Woyu He, Yangu Lin, Ning Cui, Binbin Chen, Xi Zhang, Shiming Jiang, Hao Lin, Peng Zhang, Xumeng Qi, Xiaojuan Wang, Zhongrui Xu, Xiaoxin Shang, Dashan Liu, Qi Cheng, Kwang-Ting Liu, Ming Department of Electrical and Electronic Engineering The University of Hong Kong Hong Kong Laboratory of Microelectronic Devices & Integrated Technology Institute of Microelectronics Chinese Academy of Sciences Beijing100029 China ACCESS – AI Chip Center for Emerging Smart Systems InnoHK Centers Hong Kong Science Park Hong Kong State Key Lab of Fabrication Technologies for Integrated Circuits Institute of Microelectronics Chinese Academy of Sciences Beijing100029 China Frontier Institute of Chip and System Fudan University Shanghai200433 China College of Computer Science and Technology Zhejiang University Zhejiang310027 China Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Hong Kong University of Chinese Academy of Sciences Beijing100049 China Institute of Mind The University of Hong Kong Hong Kong

The rapid advancement of artificial intelligence (AI) has been marked by the large language models exhibiting human-like intelligence. However, these models also present unprecedented challenges to energy consumption and environmental sustainability, which are exacerbated by the increasing number of users and the development of even larger models. One promising solution is to revisit analogue computing, a technique that predates digital computing and exploits emerging analogue electronic devices, such as resistive memory, which features in-memory computing, high scalability, and nonvolatility that addresses the von Neumann bottleneck, slowdown of Moore’s law, and volatile DRAM of conventional digital hardware. However, analogue computing still faces the same challenges as before: programming nonidealities and expensive programming due to the underlying devices physics. Therefore, leveraging the efficiency advantage while mitigating the programming disadvantage of analogue computing with resistive memory is a major open problem in AI hardware and electronics communities. Here, we report a universal solution, software-hardware co-design using structural plasticity-inspired edge pruning to optimize the topology of a randomly weighted analogue resistive memory neural network. Software-wise, the topology of a randomly weighted neural network is optimized by pruning connections rather than precisely tuning resistive memory weights. Hardware-wise, we reveal the physical origin of the programming stochasticity using transmission electron microscopy, which is leveraged for large-scale and low-cost implementation of an overparameterized random neural network containing high-performance sub-networks. We implemented the co-design on a 40nm 256K resistive memory macro, observing 17.3% and 19.9% accuracy improvements in image and audio classification on FashionMNIST and Spoken digits datasets, as well as 9.8% (2%) improvement in PR (ROC) in image segmentation on DRIVE datasets, r

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Optical coherent dot-product chip for sophisticated deep learning regression

arXiv

引用

arXiv 2021年

作者： Xu, Shaofu Wang, Jing Shu, Haowen Zhang, Zhike Yi, Sicheng Bai, Bowen Wang, Xingjun Liu, Jianguo Zou, Weiwen Department of Electronic Engineering Shanghai Jiao Tong University 800 Dongchuan Road Shanghai200240 China State Key Laboratory of Advanced Optical Communications System and Networks Department of Electronics School of Electronics Engineering and Computer Science Peking University Beijing100871 China Institution of Semiconductors Chinese Academy of Sciences Beijing100083 China

Optical implementations of neural networks (ONNs) herald the next-generation high-speed and energy-efficient deep learning computing by harnessing the technical advantages of large bandwidth and high parallelism of optics. However, due to the problems of incomplete numerical domain, limited hardware scale, or inadequate numerical accuracy, the majority of existing ONNs were studied for basic classification tasks. Given that regression is a fundamental form of deep learning and accounts for a large part of current artificial intelligence applications, it is necessary to master deep learning regression for further development and deployment of ONNs. Here, we demonstrate a silicon-based optical coherent dot-product chip (OCDC) capable of completing deep learning regression tasks. The OCDC adopts optical fields to carry out operations in complete real-value domain instead of in only positive domain. Via reusing, a single chip conducts matrix multiplications and convolutions in neural networks of any complexity. Also, hardware deviations are compensated via in-situ backpropagation control provided the simplicity of chip architecture. Therefore, the OCDC meets the requirements for sophisticated regression tasks and we successfully demonstrate a representative neural network, the AUTOMAP (a cutting-edge neural network model for image reconstruction). The quality of reconstructed images by the OCDC and a 32-bit digital computer is comparable. To the best of our knowledge, there is no precedent of performing such state-of-the-art regression tasks on ONN chip. It is anticipated that the OCDC can promote novel accomplishment of ONNs in modern AI applications including autonomous driving, natural language processing, and scientific study. © 2021, CC BY-NC-SA.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

SYNAPSE: An international roadmap to large brain imaging

引用

Physics Reports 2023年第1期999卷 1-60页

作者： Stampfl, Anton P.J. Liu, Zhongdong Hu, Jun Sawada, Kei Takano, H. Kohmura, Yoshiki Ishikawa, Tetsuya Lim, Jae-Hong Je, Jung-Ho Low, Chian-Ming Teo, Alvin Tok, Eng Soon Tan, Tin Wee Ban, Kenneth Libedinsky, Camilo Tan, Francis Chee Kuan Chen, Kuan-Peng Yang, An-Cheng Chuang, Chao-Chun Chen, Nan-Yow Shih, Chi-Tin Lee, Ting-Kuo Yang, De-Nian Lai, Hsu-Chao Shuai, Hong-Han Cheng, Chang-Chieh Ching, Yu-Tai Li, Chia-Wei Charng, Ching-Che Lo, Chung-Chuan Chiang, Ann-Shyn Recur, Benoit Petibois, Cyril Cheng, Chia-Liang Chen, Hsiang-Hsin Yang, Shun-Min Hwu, Yeukuang Rojviriya, Catleya Rugmai, Supagorn Rujirawat, Saroj Margaritondo, Giorgio Australian Nuclear Science and Technology Organization Lucas Heights 2234 NSW Australia Henan university of Technology Zhengzhou China Shanghai Institute of Applied Physics Chinese Academy of Science Shanghai China RIKEN SPring-8 Center Hyogo 679-5148 Japan Pohang Accelerator Laboratory Pohang University of Science and Technology Pohang Gyeongbuk 37673 South Korea Department of Materials Science and Engineering Pohang University of Science and Technology Gyeongbuk Pohang 37673 South Korea Department of Pharmacology Department of Anaesthesia and Healthy Longevity Research Programme Yong Loo Lin School of Medicine National University of Singapore Singapore School of Chemical and Life Sciences Nanyang Polytechnic Singapore eMaGIC-Lab Department of Physics National University of Singapore Singapore National Supercomputing Center Singapore Department of Psychology The N1 Institute for Health National University of Singapore Singapore Institute of Molecular and Cell Biology A*STAR Singapore Department of Anaesthesia National University Hospital National University Health System Singapore National Center for High-Performance Computing Hsinchu Taiwan Department of Physics Tunghai University Taichung Taiwan Department of Physics National Tsing Hua University Hsinchu Taiwan Institute of Information Science Academia Sinica Taipei Taiwan Department of Computer Science National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Electronics and Electrical Engineering National Yang Ming Chiao Tung University Hsinchu Taiwan Information Technology Service Center National Yang Ming Chiao Tung University Hsinchu Taiwan National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Life Science National Tsing Hua University Hsinchu Taiwan Institute of Systems Neuroscience and Brain Research Center National Tsing Hua University Hsinchu Taiwan Institute of Physics Academia Sinica Taipei Taiwan Department of Physics National Dong Hwa University

Since 2020, synchrotron radiation facilities in several Asia-Pacific countries have been collaborating in a major project called “SYNAPSE” (Synchrotrons for Neuroscience: an Asia-Pacific Scientific Enterprise). They use x-ray imaging to attack in a coordinated fashion one of the major issues in modern science: the structure of animal and human brains, including neurons and connections. The objective is to develop Google-like maps also including detailed structural and functional information for selected regions of interest. The sheer mass of data needed for the objective poses huge problems for the acquisition, processing, storage and use of images. In order to complete the task within a reasonable time, the key element of the SYNAPSE strategy is the parallel and coordinated work of several facilities on the same specimens. This article reviews different aspects of the enterprise, including the foundations of synchrotron radiation, coherence and of its role in advanced imaging, electron accelerators, x-ray optics and detectors. This will provide the foundation for an extensive presentation of the different components of SYNAPSE, with an overview of results already obtained within the consortium. © 2022

关键词： Brain mapping Phase contrast Radiology Supercomputing Synchrotron X-rays

来源：评论

学校读者我要写书评

暂无评论

Fast all-optical random number generator

arXiv

引用

arXiv 2022年

作者： Zhao, Jiamin Li, Pu Zhang, Xing Gao, Zhensen Jia, Zhiwei Bogris, Adonis Alan Shore, K. Wang, Yuncai Key Laboratory of Advanced Transducers and Intelligent Control System Ministry of Education Taiyuan University of Technology Taiyuan030024 China School of Information Engineering Guangdong University of Technology Guangzhou510006 China Guangdong Provincial Key Laboratory of Photonics Information Technology Guangzhou510006 China Key Laboratory of Specialty Fiber Optics and Optical Access Networks Shanghai University Shanghai200444 China Changchun Institute of Optics Fine Mechanics and Physics Chinese Academy of Sciences Changchun130033 China Department of Informatics and Computer Engineering University of West Attica Athens12243 Greece School of Electronic Engineering Bangor University WalesLL57 1UT United Kingdom

We propose a simple and all-optical method for fast random number generation based on the laser mode hopping. Through periodically restarting a two-mode laser operating in the bistable state, a random number stream can be generated due to the spontaneous emission noise. To validate the feasibility of this method, we perform a theoretical simulation using the common vertical-cavity surface-emitting laser (VCSEL) with two polarization modes. Numerical results demonstrate that fast 2.5 Gb/s random number streams can be continuously obtained with verified randomness. This scheme provide a fully monolithic solution for random number generator, due to its simple and all-optical structure. Copyright © 2022, The Authors. All rights reserved.

关键词： Random number generation

来源：评论

学校读者我要写书评

暂无评论

Multi-functional metasurface architecture for amplitude, polarization and wavefront control

arXiv

引用

arXiv 2022年

作者： Pitilakis, A. Seckel, M. Tasolamprou, A.C. Liu, F. Deltsidis, A. Manessis, D. Ostmann, A. Kantartzis, N.V. Liaskos, C. Soukoulis, C.M. Tretyakov, S.A. Kafesaki, M. Tsilipakos, O. School of Electrical and Computer Engineering Aristotle University of Thessaloniki ThessalonikiGR-54124 Greece System Integration and Interconnection Technologies Fraunhofer IZM Berlin13355 Germany Institute of Electronic Structure and Laser Foundation for Research and Technology Hellas Crete HeraklionGR-70013 Greece Xi'an Jiaotong University School of Electronic Science and Engineering Xi'An710049 China Department of Electronics and Nanoengineering Aalto University AaltoFI-00076 Finland Department of Computer Science Engineering University of Ioannina Ioannina Greece Ames Laboratory U.S. DOE Department of Physics Astronomy Iowa State University AmesIA50011 United States Department of Materials Science and Technology University of Crete Crete HeraklionGR-70013 Greece

Metasurfaces (MSs) have been utilized to manipulate different properties of electromagnetic waves. By combining local control over the wave amplitude, phase, and polarization into a single tunable structure, a multifunctional and reconfigurable metasurface can be realized, capable of full control over incident radiation. Here, we experimentally validate a multi-functional metasurface architecture for the microwave regime, where in principle variable loads are connected behind the backplane to reconfigurably shape the complex surface impedance. As a proof-of-concept step, we fabricate several metasurface instances with static loads in different configurations (surface mount capacitors and resistors of different values in different connection topologies) to validate the approach and showcase the different achievable functionalities. Specifically, we show perfect absorption for oblique incidence (both polarizations), broadband linear polarization conversion, and beam splitting, demonstrating control over the amplitude, polarization state, and wavefront, respectively. Measurements are performed in the 4-18 GHz range inside an anechoic chamber and show good agreement with theoretically-anticipated results. Our results clearly demonstrate the practical potential of the proposed architecture for reconfigurable electromagnetic wave manipulation. © 2022, CC BY.

关键词： Electromagnetic waves

来源：评论

学校读者我要写书评

暂无评论

Measuring fairness preferences is important for artificial intelligence in health care

引用

The Lancet digital Health 2024年第5期6卷 e302-e304页

作者： Näher, Anatol-Fiete Krumpal, Ivar Antão, Esther-Maria Ong, Erika Rojo, Marina Kaggwa, Fred Balzer, Felix Celi, Leo Anthony Braune, Katarina Wieler, Lothar H Agha-Mir-Salim, Louis Digital Global Public Health Hasso Plattner Institute for Digital Engineering University of Potsdam Potsdam14482 Germany Institute of Medical Informatics Charité–Universitätsmedizin Berlin Berlin Germany Faculty of Social Science and Philosophy University of Leipzig Leipzig Germany College of Medicine University of the Philippines Manila Philippines Department of Public Health School of Medicine University of Buenos Aires Buenos Aires Argentina Department of Computer Science Faculty of Computing and Informatics Mbarara University of Science and Technology Mbarara Uganda Laboratory for Computational Physiology Massachusetts Institute of Technology CambridgeMA United States Division of Pulmonary Critical Care and Sleep Medicine Beth Israel Deaconess Medical Center BostonMA United States Department of Biostatistics Harvard T H Chan School of Public Health BostonMA United States

来源：评论

学校读者我要写书评

暂无评论

Dynamically Controllable Two-Dimensional Microvehicle by Coordinated Optical Pulling-Lateral Force

IEEE Photonics Journal

引用

IEEE Photonics Journal 2023年第6期15卷

作者： Shi, Yuzhi Luo, Hong Xiong, Sha He, Tao Zhu, Tongtong Song, Qinghua Zhou, Lei-Ming Wu, Pin Chieh Wang, Zhanshan Qiu, Cheng-Wei Cheng, Xinbin Moe Key Laboratory of Advanced Micro-Structured Materials Shanghai200092 China Tongji University Shanghai Institute of Intelligent Science and Technology Shanghai200092 China Shanghai Frontiers Science Center of Digital Optics Shanghai200092 China Central South University School of Automation Changsha410083 China Dalian University of Technology School of Physics Dalian116024 China Tsinghua University Tsinghua Shenzhen International Graduate School Shenzhen518055 China Hefei University of Technology Department of Optical Engineering School of Physics Hefei230601 China National University of Singapore Department of Electrical and Computer Engineering 117583 Singapore National Cheng Kung University Department of Photonics Tainan70101 Taiwan

Harnessing exotic optical forces promises a plethora of biophysical applications and novel light-matter interactions. The exotic optical pulling force (OPF) and optical lateral force (OLF) have been studied separately, yet synthesizing both candidates simultaneously remains an unsolved challenge and could offer a more powerful manoeuvre of particles. Here, we report a coordinated scheme to harness these two forces together and present a dynamically controlled two-dimensional (2D) microvehicle. The strategy is to leverage unexplored helicity-dependent features of both forces, while the particle size and incident angle of light can also reverse optical forces. The underlying physics of the pulling-lateral force is beyond the dipole approximation, and can be the combined effect from the linear momentum transfer, spin-orbit interactions, etc. Notably, the ratio of both forces can be dynamically and arbitrarily controlled by the ellipticity of incident light solely. The configured 2D microvehicle provides a nontrivial recipe other than using metastructures which require exquisite designs and subtle fabrication processes. © 2009-2012 IEEE.

关键词： Particle size

来源：评论

学校读者我要写书评

暂无评论

Improved end-to-end dysarthric speech recognition via meta-learning based model re-initialization

arXiv

引用

arXiv 2020年

作者： Wang, Disong Yu, Jianwei Wu, Xixin Sun, Lifa Liu, Xunying Meng, Helen Human-Computer Communications Laboratory Department of System Engineering and Engineering Management Chinese University of Hong Kong Hong Kong Department of Engineering University of Cambridge United Kingdom SpeechX Limited Shenzhen China

Dysarthric speech recognition is a challenging task as dysarthric data is limited and its acoustics deviate significantly from normal speech. Model-based speaker adaptation is a promising method by using the limited dysarthric speech to fine-tune a base model that has been pre-trained from large amounts of normal speech to obtain speaker-dependent models. However, statistic distribution mismatches between the normal and dysarthric speech data limit the adaptation performance of the base model. To address this problem, we propose to reinitialize the base model via meta-learning to obtain a better model initialization. Specifically, we focus on end-to-end models and extend the model-agnostic meta learning (MAML) and Reptile algorithms to meta update the base model by repeatedly simulating adaptation to different dysarthric speakers. As a result, the re-initialized model acquires dysarthric speech knowledge and learns how to perform fast adaptation to unseen dysarthric speakers with improved performance. Experimental results on UASpeech dataset show that the best model with proposed methods achieves 54.2% and 7.6% relative word error rate reduction compared with the base model without finetuning and the model directly fine-tuned from the base model, respectively, and it is comparable with the state-of-the-art hybrid DNN-HMM model. Copyright © 2020, The Authors. All rights reserved.

关键词： Speech

来源：评论

学校读者我要写书评

暂无评论

Constructing Epidermal Rete Ridges Using Composite Hydrogel to Enhance Multiple Signaling Pathways for the Maintenance of Epidermal Stem Cell Niche

SSRN

引用

SSRN 2023年

作者： Shen, Zhizhong Liu, Zixian Sun, Lei Li, Meng Han, Lu Wang, Jianming Wu, Xunwei Sang, Shengbo Shanxi Key Laboratory of Micro Nano Sensors & Artificial Intelligence Perception College of Information and Computer Taiyuan University of Technology Taiyuan030024 China Key Lab of Advanced Transducers and Intelligent Control System the Ministry of Education Taiyuan University of Technology Taiyuan030024 China Shanxi Research Institute of 6D Artificial Intelligence Biomedical Science Taiyuan030031 China General Hospital of TISCO North Street Xinghualing District Taiyuan030809 China Engineering Laboratory for Biomaterials and Tissue Regeneration Ningbo Stomatology Hospital Savaid Stomatology School Hangzhou Medical College Ningbo China Department of Tissue Engineering and Regeneration School and Hospital of Stomatology Cheeloo College of Medicine Shandong University Shandong Key Laboratory of Oral Tissue Regeneration Shandong Engineering Laboratory for Dental Materials and Oral Tissue Regeneration Shandong Jinan China Cutaneous Biology Research Center Massachusetts General Hospital Harvard Medical School BostonMA United States

The undulating microstructure rete ridge (RR) located at the junction between the dermis and epidermis plays a crucial role in improving skin mechanical properties and maintaining skin homeostasis. However, the investigation of RR microstructures are usually neglected in current tissue engineering for skin regeneration. Here, to create an epidermal model with RR microstructures, keratinocytes were inoculated onto a patterned GelMA-PEGDA hydrogel constructed using molding technology. Furthermore, grafting acryloylated Arg-Gly-Asp (RGD) peptides on the hydrogel surface significantly improved cell adhesion, fusion, and development. RT-PCR, Western blot, and immunofluorescence staining confirmed that cells on RR microstructures exhibited higher gene and protein expression associated with epidermal stem cells. RNA sequencing analysis of cells on RR microstructure showed higher gene expression profiles related to stem cell maintenance, basement membrane formation, and epidermal development. Furthermore, RT-PCR analysis of epidermal models of various dimensions demonstrated that smaller microstructures were more conducive to epidermal stem cell marker gene expression, which is analogous to human skin. Overall, we have successfully developed a method for integrating RR microstructures into an epidermal model that mimics natural skin to maintain epidermal stem cell niche, providing a valuable reference for researching skin regeneration within the fields of tissue engineering and regenerative medicine. © 2023, The Authors. All rights reserved.

关键词： Microstructure

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient Cell-Free Massive MIMO Through Sparse Large-Scale Fading Processing

arXiv

引用

arXiv 2022年

作者： Chen, Shuaifei Zhang, Jiayi Björnson, Emil Demir, Özlem Tuğfe Ai, Bo The School of Electronic and Information Engineering The Frontiers Science Center for Smart Highspeed Railway System Beijing Jiaotong University Beijing100044 China Purple Mountain Laboratories Nanjing211111 China The Department of Computer Science KTH Royal Institute of Technology Kista164 40 Sweden The Department of Electrical and Electronics Engineering TOBB University of Economics and Technology Ankara Turkey The State Key Laboratory of Rail Traffic Control and Safety Beijing Jiaotong University Beijing100044 China The Henan Joint International Research Laboratory of Intelligent Networking and Data Analysis Zhengzhou University Zhengzhou450001 China

Cell-free massive multiple-input multiple-output (CF mMIMO) systems serve the user equipments (UEs) by geographically distributed access points (APs) by means of joint transmission and reception. To limit the power consumption due to fronthaul signaling and processing, each UE should only be served by a subset of the APs, but it is hard to identify that subset. Previous works have tackled this combinatorial problem heuristically. In this paper, we propose a sparse distributed processing design for CF mMIMO, where the AP-UE association and long-term signal processing coefficients are jointly optimized. We formulate two sparsity-inducing mean-squared error (MSE) minimization problems and solve them by using efficient proximal approaches with block-coordinate descent. For the downlink, more specifically, we develop a virtually optimized large-scale fading precoding (V-LSFP) scheme using uplink-downlink duality. The numerical results show that the proposed sparse processing schemes work well in both uplink and downlink. In particular, they achieve almost the same spectral efficiency as if all APs would serve all UEs, while the energy efficiency is 2-4 times higher thanks to the reduced processing and signaling. Copyright © 2022, The Authors. All rights reserved.

关键词： Energy efficiency

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：