检索结果-内蒙古大学图书馆

Efficient and Equivariant Graph Networks for Predicting Quantum Hamiltonian

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yu, Haiyang Xu, Zhao Qian, Xiaofeng Qian, Xiaoning Ji, Shuiwang Department of Computer Science & Engineering Texas A&M University TX United States Department of Materials Science & Engineering Texas A&M University TX United States Department of Electrical & Computer Engineering Texas A&M University TX United States Department of Physics & Astronomy Texas A&M University TX United States

We consider the prediction of the Hamiltonian matrix, which finds use in quantum chemistry and condensed matter physics. Efficiency and equivariance are two important, but conflicting factors. In this work, we propose a SE(3)-equivariant network, named QHNet, that achieves efficiency and equivariance. Our key advance lies at the innovative design of QHNet architecture, which not only obeys the underlying symmetries, but also enables the reduction of number of tensor products by 92%. In addition, QHNet prevents the exponential growth of channel dimension when more atom types are involved. We perform experiments on MD17 datasets, including four molecular systems. Experimental results show that our QHNet can achieve comparable performance to the state of the art methods at a significantly faster speed. Besides, our QHNet consumes 50% less memory due to its streamlined architecture. Our code is publicly available as part of the AIRS library (https://***/divelab/AIRS). Copyright © 2023, The Authors. All rights reserved.

关键词： Quantum chemistry

Compressing Image-to-Image Translation GANs Using Local Density Structures on Their Learned Manifold

学校读者我要写书评

暂无评论

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

High-performance polarization management devices based on thin-film lithium niobate

学校读者我要写书评

暂无评论

Light(science & Applications) 2022年第5期11卷 868-876页

作者： Zhongjin Lin Yanmei Lin Hao Li Mengyue Xu Mingbo He Wei Ke Heyun Tan Ya Han Zhaohui Li Dawei Wang X.Steve Yao Songnian Fu Siyuan Yu Xinlun Cai State Key Laboratory of Optoelectronic Materials and Technologies School of Electronics and Information TechnologySun Yat-sen University510275 GuangzhouChina Department of Electrical and Computer Engineering The University of British ColumbiaVancouverBC V6T 1Z4Canada Photonics Information Innovation Center and Hebei Provincial Center for Optical Sensing Innovations College of Physics Science and TechnologyHebei University071002 BaodingChina Institute of Advanced Photonics Technology School of Information EngineeringGuangdong University of Technology510006 GuangzhouChina

High-speed polarization management is highly desirable for many applications,such as remote sensing,telecommunication,and medical ***,most of the approaches for polarization management rely on bulky optical components that are slow to respond,cumbersome to use,and sometimes with high drive ***,we overcome these limitations by harnessing photonic integrated circuits based on thin-film lithium niobate *** successfully realize a portfolio of thin-film lithium niobate devices for essential polarization management functionalities,including arbitrary polarization generation,fast polarization measurement,polarization scrambling,and automatic polarization *** present devices feature ultra-fast control speeds,low drive voltages,low optical losses and compact *** these devices,we achieve high fidelity polarization generation with a polarization extinction ratio up to 41.9 dB and fast polarization scrambling with a scrambling rate up to 65 Mrad s−1,both of which are best results in integrated *** also demonstrate the endless polarization state tracking operation in our *** demonstrated devices unlock a drastically new level of performance and scales in polarization management devices,leading to a paradigm shift in polarization management.

关键词： polarization lithium film

Sensing property of Ga2O3-based extended-gate field-effect transistors for a living cell viability sensor

学校读者我要写书评

暂无评论

Sensors and Actuators A: Physical 2023年 349卷

作者： Chou, Hsin-Yu Chiang, Jung-Lung Yu, Chang-Tze Ricky Chen, Jo-Mei Maureen Wuu, Dong-Sing Department of Materials Science and Engineering National Chung Hsing University Taichung40227 Taiwan Ph.D. Program Prospective Technology of Electrical Engineering and Computer Science National Chin-Yi University of Technology Taichung411030 Taiwan Department of Applied Chemistry National Chi-Nan University Nantou54561 Taiwan Department of Applied Materials and Optoelectronic Engineering National Chi Nan University Nantou54561 Taiwan

In this study, gallium oxide (Ga2O3) nanorods were deposited onto an indium tin oxide (ITO) glass substrate to develop a real-time living cell viability sensor. Ga2O3 nanorods had characteristics of cell population sensing based on cellular metabolism to detect cell viability. The pH sensing characteristics of Ga2O3-based extended-gate field-effect transistors (EGFETs) with and without the annealing process were measured from pH 1–11. Three different sensor heads, ITO/glass, GaOOH/ITO/glass, and Ga2O3/ITO/glass, were produced as EGFETs and immersed into a pH standard buffer and high-glucose Dulbecco's modified Eagle's medium (DMEM) to investigate hydrogen ion sensing. Results showed that the Ga2O3/ITO/glass probe had the best hydrogen ion sensing sensitivity and stability in different sample buffers. In contrast, GaOOH/ITO/glass had instability and lower sensitivity due to interference with other components in the medium. The drift characteristic of sensors was analyzed to detect long-term stability. The Ga2O3/ITO/glass probe showed lower drift properties, which can potentially monitor the DMEM situation in real-time for a long-term period. During in vitro detection, the Ga2O3/ITO/glass probe revealed a 70 μA/cells change in IDS-VDS analysis and a 10 μV/cells change in IDS-VGS analysis, which was superior linearly dependent within 10,000 cells. The biocompatibility of the Ga2O3/ITO/glass probe was also detected by 3-(4,5-dimethylthiazol-2-yl)− 2,5-diphenyltetrazolium bromide assay and showed no cytotoxicity. Overall, the Ga2O3/ITO/glass EGFET sensor has potential applications as a living cell viability device in microenvironmental studies in the future. © 2022 Elsevier B.V.

关键词： Nanorods

Utilization of Artifact and Noise Affected Electrocardiogram for Simultaneous Heart Rate Computation and Motion Type Inference

学校读者我要写书评

暂无评论

Utilization of Artifact and Noise Affected Electrocardiogram...

IEEE SENSORS

作者： Jihwan Kim Hyun Bin Kim Dengyang Lu Byeong Woon Lee Wooseok Kim Sang Uk Park Hee Kyu Lee Jaejun Lim Yida Wang Seunghwan Seo Sang Min Won Department of Electrical and Computer Engineering Sungkyunkwan University Suwon South Korea Department of Materials Science and Engineering University of Pennsylvania Philadelphia PA USA Research Laboratory of Electronics Massachusetts Institute of Technology Cambridge MA USA

The specific objective of developing neural interfaces is to analyze the electrophysiological signals associated with neural activities within the body, with the aim of providing early diagnosis for diseases related to the neural system. One example is the electrocardiogram (ECG) that assesses the functionality of the heart, identifying arrhythmias and potentially pinpointing the location of any malfunction. Conventional clinical protocols require cardiac electrical signal measurements to be conducted during a stationary phase, with an emphasis on minimizing noise interference as much as possible. While this holds true in clinical scenarios, such motion artifacts can be valuable sources of information for motion tracking. In this study, we present a methodology where we demonstrate the simultaneous monitoring and computation of both heart rate and motion using a wireless ECG sensing system with a three-electrode configuration.

关键词：

Superconductivity from On-Chip Metallization on 2D Topological Chalcogenides

学校读者我要写书评

暂无评论

Physical Review X 2024年第2期14卷 021051-021051页

作者： Yanyu Jia Guo Yu Tiancheng Song Fang Yuan Ayelet J. Uzan Yue Tang Pengjie Wang Ratnadwip Singha Michael Onyszczak Zhaoyi Joy Zheng Kenji Watanabe Takashi Taniguchi Leslie M. Schoop Sanfeng Wu Department of Physics Princeton University Princeton New Jersey 08544 USA Department of Electrical and Computer Engineering Princeton University Princeton New Jersey 08544 USA Department of Chemistry Princeton University Princeton New Jersey 08544 USA Research Center for Electronic and Optical Materials National Institute for Materials Science 1-1 Namiki Tsukuba 305-0044 Japan Research Center for Materials Nanoarchitectonics National Institute for Materials Science 1-1 Namiki Tsukuba 305-0044 Japan

Two-dimensional (2D) transition metal dichalcogenides (TMDs) is a versatile class of quantum materials of interest to various fields including, e.g., nanoelectronics, optical devices, and topological and correlated quantum matter. Tailoring the electronic properties of TMDs is essential to their applications in many directions. Here, we report that a highly controllable and uniform on-chip 2D metallization process converts a class of atomically thin TMDs into robust superconductors, a property belonging to none of the starting materials. As examples, we demonstrate the introduction of superconductivity into a class of 2D air-sensitive topological TMDs, including monolayers of Td−WTe2, 1T′−MoTe2, and 2H−MoTe2, as well as their natural and twisted bilayers, metallized with an ultrathin layer of palladium. This class of TMDs is known to exhibit intriguing topological phases ranging from topological insulator, Weyl semimetal to fractional Chern insulator. The unique, high-quality two-dimensional metallization process is based on our recent findings of the long-distance, non-Fickian in-plane mass transport and chemistry in 2D that occur at relatively low temperatures and in devices fully encapsulated with inert insulating layers. Highly compatible with existing nanofabrication techniques for van der Waals stacks, our results offer a route to designing and engineering superconductivity and topological phases in a class of correlated 2D materials.

关键词： Anyons Topological superconductors Twistronics Superconducting devices Superconductors Topological materials Transition metal dichalcogenides Twisted heterostructures Crystal growth Dilution refrigerator Resistivity measurements

Electronic structure of InSb (001), (110), and (111)B surfaces

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Dong, Jason T. Inbar, Hadass S. Pendharkar, Mihir van Schijndel, Teun A.J. Young, Elliot C. Dempsey, Connor P. Palmstrøm, Christopher J. Materials Department University of California Santa BarbaraCA93106 United States Department of Electrical and Computer Engineering University of California Santa BarbaraCA93106 United States Department of Materials Science and Engineering Stanford University StanfordCA94305 United States

The electronic structure of various (001), (110), and (111)B surfaces of n-type InSb were studied with scanning tunneling microscopy and spectroscopy. The InSb(111)B (3x1) surface reconstruction is determined to be a disordered (111)B (3x3) surface reconstruction. The surface Fermi-level of the In rich and the equal In:Sb (001), (110), and (111)B surface reconstructions was observed to be pinned near the valence band edge. This observed pinning is consistent with a charge neutrality level lying near the valence band maximum. Sb termination was observed to shift the surface Fermi-level position by up to 254 ± 35 meV towards the conduction band on the InSb (001) surface and 60 ± 35 meV towards the conduction band on the InSb(111)B surface. The surface Sb on the (001) can shift the surface from electron depletion to electron accumulation. We propose the shift in the Fermi-level pinning is due to charge transfer from Sb clusters on the Sb terminated surfaces. Additionally, many sub-gap states were observed for the (111)B (3x1) surface, which are attributed to the disordered nature of this surface. This work demonstrates the tuning of the Fermi-level pinning position of InSb surfaces with Sb termination. Copyright © 2023, The Authors. All rights reserved.

关键词： Electronic structure

Localization of Crop Damages Utilizing a Wake-Up Gas Sensor Network

学校读者我要写书评

暂无评论

Localization of Crop Damages Utilizing a Wake-Up Gas Sensor ...

International Conference on Solid State Sensors and Actuators (TRANSDUCERS)

作者： Seungbeom Noh Sayali Tope Farhan Sadik Sium Shakir-Ul Haque Khan Mohit Karkhanis Leo Wang Adwait Deshpande Rana Dalapati Ravi V. Mural Carlos H. Mastrangelo Mingyu Ji Ling Zang James C. Schnable Hanseup Kim Electrical and Computer Engineering University of Utah Salt Lake City UT Materials Science and Engineering University of Utah Salt Lake City UT Department of Agronomy and Horticulture University of Nebraska Lincoln NE

This paper reports the initial demonstration of on-field crop damage prediction through a network of low power gas sensors that were deployed in an actual sorghum field and detected crop stress molecules, specifically hexanal. Such detection could replace current farm-scouting methods that are mostly manual and help reduce production loss of biofuel crops during a farming period. The initially-deployed gas sensor network consisted of a 2×3 array with a 7-meter interspace. When the leaves of sorghums were cut at a random location within the network, three sensors in the gas sensor network responded at 45, 62, and 72 minutes, respectively, allowing us to estimate an origination zone within an area of 21.7 m 2 out of the total monitoring farm area of 294 m 2 . This demonstration proves the potential of precise localization of plant damages through the use of the gas sensor network.

关键词：

Analyzing Errors in Controlled Turret System Given Target Location Input from Artificial Intelligence Methods in Automatic Target Recognition

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Karlson, Matthew Ban, Heng Cole, Daniel G. Abdelhakim, Mai Forsythe, Jennifer Department of Mechanical Engineering and Materials Science University of Pittsburgh PittsburghPA15260 United States Department of Electrical and Computer Engineering University of Pittsburgh PittsburghPA15260 United States DEVCOM Analysis Center Aberdeen Proving GroundMD21005 United States

In this paper, we assess the movement error of a targeting system given target location data from artificial intelligence (AI) methods in automatic target recognition (ATR) systems. Few studies evaluate the impacts on the accuracy in moving a targeting system to an aimpoint provided in this manner. To address this knowledge gap, we assess the performance of a controlled gun turret system given target location from an object detector developed from AI methods. In our assessment, we define a measure of object detector error and examine the correlations with several standard metrics in object detection. We then statistically analyze the object detector error data and turret movement error data acquired from controlled targeting simulations, as well as their aggregate error, to examine the impact on turret movement accuracy. Finally, we study the correlations between additional metrics and the probability of a hit. The results indicate that AI technologies are a significant source of error to targeting systems. Moreover, the results suggest that metrics such as the confidence score, intersection-over-union, average precision and average recall are predictors of accuracy against stationary targets with our system parameters. © 2024, CC BY-NC-ND.

关键词： Object detection