检索结果-内蒙古大学图书馆

Two-Dimensional Magnetic Exciton Polariton with Strongly Coupled Atomic and Photonic Anisotropies

学校读者我要写书评

暂无评论

Physical Review Letters 2024年第26期133卷 266901-266901页

作者： Qiuyang Li Xin Xie Adam Alfrey Christiano W. Beach Nicholas McLellan Yang Lu Jiaqi Hu Wenhao Liu Nikhil Dhale Department of Physics University of Michigan Ann Arbor Michigan 48109 USA Applied Physics Program University of Michigan Ann Arbor Michigan 48109 USA Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor Michigan 48109 USA Department of Physics The University of Texas at Dallas Richardson Texas 75080 USA

Anisotropy is a fundamental property of both material and photonic systems. The interplay between material and photonic anisotropies, however, has hardly been explored due to the vastly different length scales. Here we demonstrate exciton polaritons in a 2D antiferromagnet, CrSBr, coupled with an anisotropic photonic crystal cavity, where the spin, atomic, and photonic anisotropies are strongly correlated. Atomic anisotropy led to exceptionally strong coupling between anisotropic exciton and optical modes, which are stable against excitation densities a few orders of magnitude higher than polaritons in isotropic materials. The resulting polaritons feature anisotropic polarizations determined by the interplay of not only the anisotropies but also the dissipations and coupling of both exciton and photon modes, tunable by tens of degrees via many parameters. The work provides insights of excitons in CrSBr and demonstrates a prototype where atomic- and photonic-scale orders strongly couple, giving rise to unconventional properties in quantum materials and photonic devices.

关键词： Atoms

Development of Support Vector Machine Method to Forecast Dengue-Endemic Areas

学校读者我要写书评

暂无评论

Development of Support Vector Machine Method to Forecast Den...

International Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS)

作者： Hetty Meileni Ermatita Abdiansah Nyayu Latifah Husni Desmawati Doctoral Program in Engineering Science Universitas Sriwijaya (Politeknik Negeri Sriwijaya) Palembang Indonesia Faculty of Computer Science Universitas Sriwijaya Palembang Indonesia Artificial Intelligence Research Development (AIRD) Universitas Sriwijaya Palembang Indonesia Department of Electrical Engineering Politeknik Negeri Sriwijaya Palembang Palembang Indonesia Faculty of Health Science Universitas Pembangunan Nasional Veteran Jakarta Jakarta Indonesia

The cases of dengue hemorrhagic fever (DHF) in Indonesia have increased significantly since 2020. Data shown by the Central Statistics Agency, for example, in South Sumatra Province, there were 6,348 cases of DHF during that period. To suppress the case number of DHF, forecasting areas that are susceptible to this disease is urgently needed. This research aims to develop a machine learning method Support Vector Machine (SVM). The data used include area data, population density data, weather data, and dengue fever case data. The development of the SVM model is carried out using a machine learning method that combines regional data, population density data, weather data, and dengue case data. In the end, using the SVM method can be used to forecast dengue-endemic areas, so that the government and society can take further action to reduce the negative impact of DHF.

关键词：

Reconfigurable Intelligent Surface-Aided Joint Radar and Covert Communications: Fundamentals, Optimization, and Challenges

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Du, Hongyang Kang, Jiawen Niyato, Dusit Zhang, Jiayi Kim, Dong In The School of Computer Science and Engineering The Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore The School of Computer Science and Engineering Nanyang Technological University Singapore The School of Electronic and Information Engineering Beijing Jiaotong University Beijing100044 China The Department of Electrical and Computer Engineering Sungkyunkwan University Suwon16419 Korea Republic of

Future wireless communication systems will evolve toward multi-functional integrated systems to improve spectrum utilization and reduce equipment sizes. A joint radar and communication (JRC) system, which can support simultaneous information transmission and target detection, has been regarded as a promising solution for emerging applications such as autonomous vehicles. In JRC, data security and privacy protection are critical issues. Thus, we first apply covert communication into JRC and propose a joint radar and covert communication (JRCC) system to achieve high spectrum utilization and secure data transmission simultaneously. In the JRCC system, an existence of sensitive data transmission is hidden from a maliciously observant warden. However, the performance of JRCC is restricted by severe signal propagation environment and hardware devices. Fortunately, reconfigurable intelligent surfaces (RISs) can change the signal propagation smartly to improve the networks performance with low cost. We first overview fundamental concepts of JRCC and RIS and then propose the RIS-aided JRCC system design. Furthermore, both covert communication and radar performance metrics are investigated and a game theory-based covert rate optimization scheme is designed to achieve secure communication. Finally, we present several promising applications and future directions of RIS-aided JRCC systems. © 2022, CC BY.

关键词： Data transfer

Deep-learning triage of 3D pathology data for improved disease detection while reducing pathologist workloads

学校读者我要写书评

暂无评论

Deep-learning triage of 3D pathology data for improved disea...

Microscopy Histopathology and Analytics, Microscopy 2024 - Part of Optica Biophotonics Congress: Biomedical Optics

作者： Gao, Gan Wang, Fiona Brenes, David Song, Andrew H. Chow, Sarah S.L. Mahmood, Faisal Liu, Jonathan T.C. Department of Mechanical Engineering University of Washington SeattleWA98195 United States Department of Computer Science University of Washington SeattleWA98195 United States Department of Pathology Brigham and Women's Hospital Harvard Medical School BostonMA02115 United States Department of Pathology Massachusetts General Hospital Harvard Medical School BostonMA02114 United States Cancer Program Broad Institute of Harvard and MIT CambridgeMA02142 United States Data Science Program Dana-Farber Cancer Institute BostonMA02215 United States Department of Bioengineering University of Washington SeattleWA98195 United States Department of Laboratory Medicine & Pathology University of Washington School of Medicine SeattleWA98195 United States

3D pathology can potentially improve disease detection, but the datasets are too large to review. We're developing a deep-learning-based triage method to identify the highest-risk 2D sections within 3D pathology d... 详细信息

关键词： Adversarial machine learning

A Fingerprint Based Indoor Visible Light Positioning System for Tilted Receivers Equipped with a Single Photo Detector

学校读者我要写书评

暂无评论

A Fingerprint Based Indoor Visible Light Positioning System ...

IEEE Conference on Wireless Communications and Networking

作者： Ibrahim M. Abou-Shehada Ali H. Muqaibel Wessam Mesbah Ki-Hong Park Mohamed-Slim Alouini Electrical Engineering Department King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Electrical Engineering Department Center for Communication Systems and Sensing King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Electrical and Computer Engineering Program King Abdullah University of Science and Technology Thuwal Saudi Arabia

ISBN: (数字)9798350303582

ISBN: (纸本)9798350303599

Indoor positioning systems (IPS) are gaining higher attention recently due to the increased demand for indoor location aware services. Visible light communication (VLC) is a promising technology to use for IPS. In particular, received signal strength (RSS) based visible light positioning (VLP) systems are gaining high attention due to their low complexity and cost, in addition to higher positioning accuracy compared to their radio frequency (RF) counterparts. One of the main challenges in RSS based VLP systems is encountered when the receiver (the target) is tilted and not placed in parallel with the transmitters (the anchors). RSS based trilateration techniques require a computationally expensive and time-consuming process to solve the nonlinear problem of tilted receivers. Fingerprint based systems generally provide high positioning accuracy with short positioning time, and maybe used to circumvent the need to deal with the high complexity associated with tilted receivers. However, the design of a fingerprinting VLP system for tilted receiver has not been explored yet as far as receivers with a single photodetector (PD) are concerned. In this work, a fingerprint based VLP system for tilted receivers using artificial neural networks (ANN) is proposed, where different types of input features for training the positioning algorithm are studied. We show that using the components of the normal vector to the PD's surface in addition to RSS values provides an excellent positioning accuracy with an average positioning error of 25.41 cm and a remarkably low average positioning time less than $\mathbf{5} \boldsymbol{\mu} \boldsymbol{s}$ . In addition, important research directions for future work are discussed.

关键词： Training Accuracy Receivers Artificial neural networks Fingerprint recognition Vectors Real-time systems

Multi-turn Dialogue Comprehension from a Topic-aware Perspective

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ma, Xinbei Xu, Yi Zhao, Hai Zhang, Zhuosheng The Department of Computer Science and Engineering Shanghai Jiao Tong University China Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University China

Dialogue related Machine Reading Comprehension requires language models to effectively decoupling and modeling on multi-turn dialogue passages. As a dialogue development goes after the intentions of participants, its topic may not keep constant through the whole passage. Hence, it is non-trivial to detect and leverage the topic shift in dialogue modeling. Topic modeling, although has been widely studied in plain text, deserves far more utilization in dialogue reading comprehension. This paper proposes to model multi-turn dialogues from a topic-aware perspective. We start with a dialogue segmentation algorithm to split a dialogue passage into topic-concentrated fragments in an unsupervised way. Then we use these fragments as topic-aware language processing units in further dialogue comprehension. On one hand, the split segments indict specific topic rather than mixed intention, thus show convenient on in-domain topic detection and location. For this task, we design a clustering system with a self-training auto-encoder, and we build two constructed datasets for evaluation. On the other hand, the split segments are appropriate element of multi-turn dialogue response selection. For this purpose, we further present a novel model, Topic-Aware Dual-Attention Matching (TADAM) Network, which takes topic segments as processing element and matches response candidates with a dual cross-attention. Empirical studies on three public benchmarks show great improvements over baselines. Our work continues the previous studies on document topic, and brings the dialogue modeling to a novel topic-aware perspective with exhaustive experiments and analyses. Copyright © 2023, The Authors. All rights reserved.

关键词： Modeling languages

Deep learning framework for action prediction reveals multi-timescale locomotor control

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Wei-Chen De Comite, Antoine Voloshina, Alexandra Daley, Monica Seethapathi, Nidhi Department of Electrical Engineering and Computer Science School of Engineering MIT CambridgeMA02139 United States McGovern Institute for Brain Research MIT CambridgeMA02139 United States Brain and Cognitive Sciences School of Science MIT CambridgeMA02139 United States Mechanical and Aerospace Engineering Henry Samueli School of Engineering UC Irvine IrvineCA92697 United States Ecology and Evolutionary Biology School of Biological Sciences UC Irvine IrvineCA92697 United States Biomedical Engineering Henry Samueli School of Engineering UC Irvine IrvineCA92697 United States

Modeling human movement in real-world tasks is a fundamental goal for motor control, biomechanics, and rehabilitation engineering. However, existing models of essential tasks like locomotion are not applicable across varying terrain, mechanical conditions, and sensory contexts. This is at least in part due to simplifying assumptions like linear and fixed timescales mappings between inputs and future actions, which may not be broadly applicable. Here, we develop a deep learning-based framework for action prediction, outperforming traditional models across multiple contexts (walking and running, treadmill and overground, varying terrains) and input modalities (multiple body states, visual gaze). We find that neural network architectures with flexible input history-dependence, like GRU and Transformer, and with architecture-dependent trial embeddings perform best overall. By quantifying the model’s predictions relative to an autoregressive baseline, we identify context- and modality-dependent timescales. These analyses reveal that there is greater reliance on fast-timescale predictions in complex terrain, gaze predicts future foot placement before body states, and the full-body state predictions precede those by center-of-mass states. This deep learning framework for human action prediction provides quantifiable insights into the control of real-world locomotion and can be extended to other actions, contexts, and populations. © 2025, CC BY-NC-SA.

关键词： Deep learning

RDA: Robust domain adaptation via Fourier adversarial attacking

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Huang, Jiaxing Guan, Dayan Xiao, Aoran Lu, Shijian Singtel Cognitive and Artificial Intelligence Lab for Enterprises School of Computer Science and Engineering Nanyang Technological University

Unsupervised domain adaptation (UDA) involves a supervised loss in a labeled source domain and an unsupervised loss in an unlabeled target domain, which often faces more severe overfitting (than classical supervised learning) as the supervised source loss has clear domain gap and the unsupervised target loss is often noisy due to the lack of annotations. This paper presents RDA, a robust domain adaptation technique that introduces adversarial attacking to mitigate overfitting in UDA. We achieve robust domain adaptation by a novel Fourier adversarial attacking (FAA) method that allows large magnitude of perturbation noises but has minimal modification of image semantics, the former is critical to the effectiveness of its generated adversarial samples due to the existence of 'domain gaps'. Specifically, FAA decomposes images into multiple frequency components (FCs) and generates adversarial samples by just perturbating certain FCs that capture little semantic information. With FAA-generated samples, the training can continue the 'random walk' and drift into an area with a flat loss landscape, leading to more robust domain adaptation. Extensive experiments over multiple domain adaptation tasks show that RDA can work with different computer vision tasks with superior performance. Copyright © 2021, The Authors. All rights reserved.

关键词： Machine learning

Player dominance adjustment in games

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Xu, Junjie H. Bachelor Program in Information Science and Engineering Department of Human and Computer Intelligence College of Information Science and Engineering Ritsumeikan University

Video Games are boring when they are too easy, and frustrating when they are too hard. In terms of providing game experience such as enjoyment to the player by match players with different levels of ability to player ability, We assume that implementing DDA for providing matches between player ability and overall game difficulty to the game, especially the modern game, has limitations in terms of increasing computational cost and complexities in the design of modeling the difficulty in modern games. To overcome limitations underlying the method of providing static difficulty changes to player, and DDA, we proposed a novel idea, "Player Domination adjustment (PDA)." The proposed idea is that to control the AI’s actions based on the player’s inputs so as to adjust the player’s dominant power (e.g. the AI recognizes the player’s attack actions but defends it in a wrong side to let the player incur damage to itself), which was proved as it leads to promotion of game-related self-efficacy in our work. Several pieces of research on were conducted on a social deduction game and a fighting game respectively, show our proposed idea has its potential of promoting User Experience(UX). As in an another study, outperforms DDA in two conducted experiments in terms of health promotion. PDA also suitable in many cases of promoting the health of the player by increasing health metrics of the player such as balancedness (Bal, a term of in use of body parts in body movement). The mechanism is AIs or the whole game environment that selects next actions of AIs/ changes based on player’s health metrics analyzed from his/her input action before such action is executed. With this mechanism we can stealthy encourage players performing the motions considering more healthy. All experimental results demonstrate that PDA is capable of entertaining player, besides, benefits to the design of serious game which uses for health promotion. There is a promising direction for future research in PDA su

关键词： Health