检索结果-内蒙古大学图书馆

High-efficiency optical frequency mixing in an all-dielectric metasurface enabled by multiple bound states in the continuum

学校读者我要写书评

暂无评论

Physical Review B 2023年第7期107卷 075441-075441页

作者： Tingting Liu Meibao Qin Feng Wu Shuyuan Xiao Institute of Photonic Chips University of Shanghai for Science and Technology Shanghai 200093 China Centre for Artificial-Intelligence Nanophotonics School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai 200093 China School of Physics and Materials Science Nanchang University Nanchang 330031 China School of Optoelectronic Engineering Guangdong Polytechnic Normal University Guangzhou 510665 China Institute for Advanced Study Nanchang University Nanchang 330031 China Jiangxi Key Laboratory for Microscale Interdisciplinary Study Nanchang University Nanchang 330031 China

We present nonlinear optical four-wave mixing in a silicon nanodisk dimer metasurface. Under the oblique incident plane waves, the designed metasurface exhibits a multiresonant feature with simultaneous excitations of three quasi-bound states in the continuum (BIC). Through employing these quasi-BICs with maximizing electric field energy at the input bump wavelengths, significant enhancement of third-order nonlinear processes including third-harmonic generation, degenerate and nondegenerate four-wave mixing are demonstrated, giving rise to ten frequencies in the visible wavelengths. This work may lead to a new frontier of ultracompact optical mixer for applications in optical circuitry, ultrasensitive sensing, and quantum nanophotonics.

关键词： Nanophotonics Metasurfaces Finite-element method

Transforming GP-CNN Tree Search Into Trainable Architectures for Image Classification

学校读者我要写书评

暂无评论

Transforming GP-CNN Tree Search Into Trainable Architectures...

IEEE International Conference on Systems, Man and Cybernetics

作者： Feng Sun Yan Ke Yue-Jiao Gong Yun Li Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen China School of Computer Science and Engineering South China University of Technology Guangzhou China i4AI Ltd London WCIN3AX United Kingdom

ISBN: (数字)9781665410205

ISBN: (纸本)9781665410212

Data-efficient image classification poses a challenge in achieving effectiveness with limited data, as evidenced by the current methods based on convolutional neural networks (CNNs) and genetic programming (GP). Existing works employing these two methods encounter limitations, such as a lack of flexibility and an inability to effectively explore the latent features of the data. To tackle these challenges, this paper introduces a genetic programming method for data-efficient image recognition, leveraging novel function sets, terminal sets, and program structures. This method transforms tree-based data structures in GP into trainable CNN architectures. Further, by employing block structures instead of single operations in the search space, the search space is reduced and the stability of the search structures enhanced. Comparative experiments with state-of-the-art neural network methods and GP-based methods on data-efficient classification datasets validate the GP-CNN method offering higher performance.

关键词： Accuracy Image recognition Neural networks Genetic programming Transforms Feature extraction Data structures Convolutional neural networks Periodic structures Image classification

Constituency Parsing using LLMs

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Bai, Xuefeng Wu, Jialong Chen, Yulong Wang, Zhongqing Zhang, Yue School of Engineering Westlake University China School of Computer Science and Engineering Southeast University Bangladesh Natural Language Processing Lab Soochow University China Institute of Advanced Technology Westlake Institute for Advanced Study China

Constituency parsing is a fundamental yet unsolved natural language processing task. In this paper, we explore the potential of recent large language models (LLMs) that have exhibited remarkable performance across various domains and tasks to tackle this task. We employ three linearization strategies to transform output trees into symbol sequences, such that LLMs can solve constituency parsing by generating linearized trees. We conduct experiments using a diverse range of LLMs, including ChatGPT, GPT-4, OPT, LLaMA, and Alpaca, comparing their performance against the state-of-the-art constituency parsers. Our experiments encompass zero-shot, few-shot, and full-training learning settings, and we evaluate the models on one in-domain and five out-of-domain test datasets. Our findings reveal insights into LLMs’ performance, generalization abilities, and challenges in constituency parsing. © 2023, CC BY.

关键词： Linearization

S-INF: Towards Realistic Indoor Scene Synthesis via Scene Implicit Neural field

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liang, Zixi Xu, Guowei Wu, Haifeng Huang, Ye Li, Wen Duan, Lixin Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China China School of Computer Science and Engineering University of Electronic Science and Technology of China China Sichuan Provincial Key Laboratory for Human Disease Gene Study the Center for Medical Genetics Department of Laboratory Medicine Sichuan Academy of Medical Sciences Sichuan Provincial People’s Hospital UESTC China

Learning-based methods have become increasingly popular in 3D indoor scene synthesis (ISS), showing superior performance over traditional optimization-based approaches. These learning-based methods typically model distributions on simple yet explicit scene representations using generative models. However, due to the oversimplified explicit representations that overlook detailed information and the lack of guidance from multimodal relationships within the scene, most learning-based methods struggle to generate indoor scenes with realistic object arrangements and styles. In this paper, we introduce a new method, Scene Implicit Neural field (S-INF), for indoor scene synthesis, aiming to learn meaningful representations of multimodal relationships, to enhance the realism of indoor scene synthesis. S-INF assumes that the scene layout is often related to the object-detailed information. It disentangles the multimodal relationships into scene layout relationships and detailed object relationships, fusing them later through implicit neural fields (INFs). By learning specialized scene layout relationships and projecting them into S-INF, we achieve a realistic generation of scene layout. Additionally, S-INF captures dense and detailed object relationships through differentiable rendering, ensuring stylistic consistency across objects. Through extensive experiments on the benchmark 3D-FRONT dataset, we demonstrate that our method consistently achieves state-of-the-art performance under different types of ISS. © 2024, CC BY.

关键词： Contrastive Learning

P-I2Prange: An Automatic Construction Architecture for Scenarios in I2P Ranges

学校读者我要写书评

暂无评论

P-I2Prange: An Automatic Construction Architecture for Scena...

International Joint Conference on Neural Networks (IJCNN)

作者： Runnan Tan Qingfeng Tan Haiyan Wang Yushun Xie Peng Zhang Cyberspace Institute of Advanced Technology Guangzhou University Guangzhou China Department of New Networks PengCheng Laboratory Shenzhen China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen China The School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Anonymous networks play a crucial role in preserving information privacy, yet simultaneously spark technical conflicts and gamesmanship between their maintainers and regulatory authorities. To solve the above conflicts and economic losses in real networks, it is imminent to realize adversarial exercises in anonymous networks and validate new technologies and scenarios in the cyber range. It is well known that the two dominant types of anonymity networks are Tor and Invisible Internet Project (I2P). This paper establishes a cyber range based on I2P to realize a task-driven automated scenario construction technique. The proposed task-driven automated scenario construction technique can solve the problem of decoupling between business scenarios and I2P cyber range network infrastructure. Specifically, our approach allows users to upload models/executable code to extend flexibly and fine-grained control I2P nodes, protocols, and traffic behavior characteristics. Then it empowers users to define application scenarios as needed, enabling them to handle intricate business scenarios programmatically. Therefore, it further supports diverse adversarial exercises and the validation of new technologies and scenarios within the dark web.

关键词： Measurement Privacy Costs Protocols Dark Web Scalability Neural networks

Polychromatic full-polarization control in mid-infrared light

学校读者我要写书评

暂无评论

Light(science & Applications) 2023年第5期12卷 894-904页

作者： Jin Chen Feilong Yu Xingsi Liu Yanjun Bao Rongsheng Chen Zengyue Zhao Jiuxu Wang Xiuxia Wang Wen Liu Yuzhi Shi Cheng-Wei Qiu Xiaoshuang Chen Wei Lu Guanhai Li State Key Laboratory of Infrared Physics Shanghai Institute of Technical PhysicsChinese Academy of Sciences500 Yu Tian RoadShanghai 200083China Hangzhou Institute for Advanced Study University of Chinese Academy of SciencesNo.1 SubLane XiangshanHangzhou 310024China Shanghai Research Center for Quantum Sciences 99 Xiupu RoadShanghai 201315China University of Chinese Academy of Science No.19 Yuquan RoadBeijing 100049China Department of Electrical and Computer Engineering National University of Singapore4 Engineering Drive 3Singapore 117583Singapore Institute of Nanophotonics Jinan UniversityGuangzhou 511443China Center for Micro-and Nanoscale Research and Fabrication University of Science and Technology of ChinaHefei 230026China Institute of Precision Optical Engineering School of Physics Science and EngineeringTongji UniversityShanghai 200092China National University of Singapore Suzhou Research Institute No.377 Linquan StreetSuzhouJiangsu 215123China

Objects with different shapes,materials and temperatures can emit distinct polarizations and spectral information in mid-infrared band,which provides a unique signature in the transparent window for object ***,the crosstalk among various polarization and wavelength channels prevents from accurate mid-infrared detections at high signal-to-noise ***,we report full-polarization metasurfaces to break the inherent eigen-polarization constraint over the wavelengths in *** recipe enables to select arbitrary orthogonal polarization basis at individual wavelength independently,therefore alleviating the crosstalk and efficiency degradation.A six-channel all-silicon metasurface is specifically presented to project focused mid-infrared light to distinct positions at three wavelengths,each with a pair of arbitrarily chosen orthogonal *** isolation ratio of 117 between neighboring polarization channels is experimentally recorded,exhibiting detection sensitivity one order of magnitude higher than existing infrared ***,the high aspect ratio~30 of our meta-structures manufactured by deep silicon etching technology at temperature−150℃ guarantees the large and precise phase dispersion control over a broadband from 3 to 4.5μ*** believe our results would benefit the noise-immune mid-infrared detections in remote sensing and space-to-ground communications.

关键词： polarization light noise

Marine Video Cloud: A Cloud-Based Video Analytics Platform for Collaborative Marine Research

学校读者我要写书评

暂无评论

Marine Video Cloud: A Cloud-Based Video Analytics Platform f...

OCEANS

作者： Ziqiang Zheng Tan-Sang Ha Yingshu Chen Haixin Liang Apple Pui-Yi Chui Yue-Him Wong Sai-Kit Yeung Department of Computer Science and Engineering The Hong Kong University of Science and Technology Division of Integrative Systems and Design Hong Kong University of Science and Technology School of Life Sciences The Chinese University of Hong Kong Institute for Advanced Study Shenzhen University

With the increasing attention and corresponding developments of data collection of underwater surveying/diving videos, a large number of underwater videos have been yielded for different purposes. However, it is time-consuming and requires the corresponding expertise to yield a hierarchical and comprehensive biological analysis report based on the collected underwater videos. The development of the underwater video analysis platform is an emergency and users of such a platform should not be limited to some specific groups (e.g., biological experts). The existing VIAME provides an intelligent solution for fishery and marine organism analysis with an installable platform. Differently, in this paper, we develop a novel cloud-based underwater video analysis platform, which focuses on more diverse marine species such as coral reefs and sea urchins. The proposed platform could provide hierarchical and comprehensive services for a wider user group. We aim to build a one-stop platform, which could allow every user to upload their surveying or diving videos and build a centralized database for more robust and accurate data analysis.

关键词：

A Digital Twin-based Intelligent Network Architecture for Underwater Acoustic Sensor Networks

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Song, Shanshan Huangfu, Bingwen Guo, Jiani Liu, Jun Cui, Junhong Shen, Xuemin The College of Computer Science and Technology Jilin University Changchun130012 China The School of Electronic and Information Engineering Beihang University Beijing100191 China Shenzhen Institute for Advanced Study UESTC Shenzhen China The College of Computer Science and Technology Jilin University Changchun China The Department of Electrical and Computer Engineering University of Waterloo WaterlooONN2L 3G1 Canada

Underwater acoustic sensor networks (UASNs) drive toward strong environmental adaptability, intelligence, and multifunctionality. However, due to unique UASN characteristics, such as long propagation delay, dynamic channel quality, and high attenuation, existing studies present untimeliness, inefficiency, and inflexibility in real practice. Digital twin (DT) technology is promising for UASNs to break the above bottlenecks by providing high-fidelity status prediction and exploring optimal schemes. In this article, we propose a Digital Twin-based Network Architecture (DTNA), enhancing UASNs’ environmental adaptability, intelligence, and multifunctionality. By extracting real UASN information from local (node) and global (network) levels, we first design a layered architecture to improve the DT replica fidelity and UASN control flexibility. In local DT, we develop a resource allocation paradigm (RAPD), which rapidly perceives performance variations and iteratively optimizes allocation schemes to improve real-time environmental adaptability of resource allocation algorithms. In global DT, we aggregate decentralized local DT data and propose a collaborative Multi-agent reinforcement learning framework (CMFD) and a task-oriented network slicing (TNSD). CMFD patches scarce real data and provides extensive DT data to accelerate AI model training. TNSD unifies heterogeneous tasks’ demand extraction and efficiently provides comprehensive network status, improving the flexibility of multi-task scheduling algorithms. Finally, practical and simulation experiments verify the high fidelity of DT. Compared with the original UASN architecture, experiment results demonstrate that DTNA can: (i) improve the timeliness and robustness of resource allocation;(ii) greatly reduce the training time of AI algorithms;(iii) more rapidly obtain network status for multi-task scheduling at a low cost. Copyright © 2024, The Authors. All rights reserved.

关键词： Resource allocation

Motion Transformer for Unsupervised Image Animation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Tao, Jiale Wang, Biao Ge, Tiezheng Jiang, Yuning Li, Wen Duan, Lixin School of Computer Science and Engineering Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China China Alibaba Group China

Image animation aims to animate a source image by using motion learned from a driving video. Current state-of-the-art methods typically use convolutional neural networks (CNNs) to predict motion information, such as motion keypoints and corresponding local transformations. However, these CNN based methods do not explicitly model the interactions between motions;as a result, the important underlying motion relationship may be neglected, which can potentially lead to noticeable artifacts being produced in the generated animation video. To this end, we propose a new method, the motion transformer, which is the first attempt to build a motion estimator based on a vision transformer. More specifically, we introduce two types of tokens in our proposed method: i) image tokens formed from patch features and corresponding position encoding;and ii) motion tokens encoded with motion information. Both types of tokens are sent into vision transformers to promote underlying interactions between them through multi-head self attention blocks. By adopting this process, the motion information can be better learned to boost the model performance. The final embedded motion tokens are then used to predict the corresponding motion keypoints and local transformations. Extensive experiments on benchmark datasets show that our proposed method achieves promising results to the state-of-the-art baselines. Our source code will be public available. Copyright © 2022, The Authors. All rights reserved.

关键词： Animation