检索结果-内蒙古大学图书馆

Continual domain adaptation for machine reading comprehension

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Su, Lixin Guo, Jiafeng Zhang, Ruqing Fan, Yixing Lan, Yanyan Cheng, Xueqi CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China China University of Chinese Academy of Sciences Beijing China

Machine reading comprehension (MRC) has become a core component in a variety of natural language processing (NLP) applications such as question answering and dialogue systems. It becomes a practical challenge that an MRC model needs to learn in non-stationary environments, in which the underlying data distribution changes over time. A typical scenario is the domain drift, i.e. different domains of data come one after another, where the MRC model is required to adapt to the new domain while maintaining previously learned ability. To tackle such a challenge, in this work, we introduce the Continual Domain Adaptation (CDA) task for MRC. So far as we know, this is the first study on the continual learning perspective of MRC. We build two benchmark datasets for the CDA task, by re-organizing existing MRC collections into different domains with respect to context type and question type, respectively. We then analyze and observe the catastrophic forgetting (CF) phenomenon of MRC under the CDA setting. To tackle the CDA task, we propose several BERT-based continual learning MRC models using either regularization-based methodology or dynamic-architecture paradigm. We analyze the performance of different continual learning MRC models under the CDA task and show that the proposed dynamic-architecture based model achieves the best performance. Copyright © 2020, The Authors. All rights reserved.

关键词： Memory architecture

L2R2: Leveraging ranking for abductive reasoning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhu, Yunchang Pang, Liang Lan, Yanyan Cheng, Xueqi CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China China University of Chinese Academy of Sciences Beijing China

The abductive natural language inference task (αNLI) is proposed to evaluate the abductive reasoning ability of a learning system. In the αNLI task, two observations are given and the most plausible hypothesis is asked to pick out from the candidates. Existing methods simply formulate it as a classification problem, thus a cross-entropy log-loss objective is used during training. However, discriminating true from false does not measure the plausibility of a hypothesis, for all the hypotheses have a chance to happen, only the probabilities are different. To fill this gap, we switch to a ranking perspective that sorts the hypotheses in order of their plausibilities. With this new perspective, a novel L2R2 approach is proposed under the learning-to-rank framework. Firstly, training samples are reorganized into a ranking form, where two observations and their hypotheses are treated as the query and a set of candidate documents respectively. Then, an ESIM model or pre-trained language model, e.g. BERT or RoBERTa, is obtained as the scoring function. Finally, the loss functions for the ranking task can be either pair-wise or list-wise for training. The experimental results on the ART dataset reach the state-of-the-art on the public leaderboard. © 2020, CC BY.

关键词： Learning systems

Towards 6G wireless communication networks:vision, enabling technologies, and new paradigm shifts

学校读者我要写书评

暂无评论

science China(Information sciences) 2021年第1期64卷 5-78页

作者： Xiaohu YOU Cheng-Xiang WANG Jie HUANG Xiqi GAO Zaichen ZHANG Mao WANG Yongming HUANG Chuan ZHANG Yanxiang JIANG Jiaheng WANG Min ZHU Bin SHENG Dongming WANG Zhiwen PAN Pengcheng ZHU Yang YANG Zening LIU Ping ZHANG Xiaofeng TAO Shaoqian LI Zhi CHEN Xinying MA Chih-Lin I Shuangfeng HAN Ke LI Chengkang PAN Zhimin ZHENG Lajos HANZO Xuemin (Sherman) SHEN Yingjie Jay GUO Zhiguo DING Harald HAAS Wen TONG Peiying ZHU Ganghua YANG Jun WANG Erik G.LARSSON Hien Quoc NGO Wei HONG Haiming WANG Debin HOU Jixin CHEN Zhe CHEN Zhangcheng HAO Geoffrey Ye LI Rahim TAFAZOLLI Yue GAO H.Vincent POOR Gerhard P.FETTWEIS Ying-Chang LIANG National Mobile Communications Research Laboratory School of Information Science and EngineeringSoutheast University Purple Mountain Laboratories Shanghai Institute of Fog Computing Technology (SHIFT) ShanghaiTech University Research Center for Network Communication Peng Cheng Laboratory State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications National Engineering Laboratory for Mobile Network Technologies Beijing University of Posts and Telecommunications National Key Laboratory of Science and Technology on Communications University of Electronic Science and Technology of China (UESTC) China Mobile Research Institute School of Electronics and Computer Science University of Southampton Department of Electrical and Computer Engineering University of Waterloo Global Big Data Technologies Centre (GBDTC) University of Technology Sydney School of Electrical and Electronic Engineering The University of Manchester LiFi Research and Development Centre Institute for Digital CommunicationsSchool of EngineeringThe University of Edinburgh Huawei Technologies Canada Co. Ltd. Huawei Technologies Department of Electrical Engineering (ISY) Link?ping University Institute of Electronics Communications & Information Technology (ECIT)Queen's University Belfast State Key Laboratory of Millimeter Waves School of Information Science and EngineeringSoutheast University School of Electrical and Computer Engineering Georgia Institute of Technology 5G Innovation Centre University of Surrey Princeton University Vodafone Chair Mobile Communications Systems Technische Universit?t Dresden Center for Intelligent Networking and Communications (CINC) University of Electronic Science and Technology of China (UESTC)

The fifth generation(5G) wireless communication networks are being deployed worldwide from 2020 and more capabilities are in the process of being standardized, such as mass connectivity, ultra-reliability,and guaranteed low latency. However, 5G will not meet all requirements of the future in 2030 and beyond, and sixth generation(6G) wireless communication networks are expected to provide global coverage, enhanced spectral/energy/cost efficiency, better intelligence level and security, etc. To meet these requirements, 6G networks will rely on new enabling technologies, i.e., air interface and transmission technologies and novel network architecture, such as waveform design, multiple access, channel coding schemes, multi-antenna technologies, network slicing, cell-free architecture, and cloud/fog/edge computing. Our vision on 6G is that it will have four new paradigm shifts. First, to satisfy the requirement of global coverage, 6G will not be limited to terrestrial communication networks, which will need to be complemented with non-terrestrial networks such as satellite and unmanned aerial vehicle(UAV) communication networks, thus achieving a space-airground-sea integrated communication network. Second, all spectra will be fully explored to further increase data rates and connection density, including the sub-6GHz, millimeter wave(mmWave), terahertz(THz),and optical frequency bands. Third, facing the big datasets generated by the use of extremely heterogeneous networks, diverse communication scenarios, large numbers of antennas, wide bandwidths, and new service requirements, 6G networks will enable a new range of smart applications with the aid of artificial intelligence(AI) and big data technologies. Fourth, network security will have to be strengthened when developing 6G networks. This article provides a comprehensive survey of recent advances and future trends in these four aspects. Clearly, 6G with additional technical requirements beyond those of 5G will enable f

关键词： 6G vision network architecture air interface and transmission technologies space-air-ground-sea integrated network all spectra artificial intelligence network security

Model-Based Reinforcement Learning for Quantized Federated Learning Performance Optimization

学校读者我要写书评

暂无评论

Model-Based Reinforcement Learning for Quantized Federated L...

GLOBECOM 2022 - 2022 IEEE Global Communications Conference

作者： Nuocheng Yang Sihua Wang Mingzhe Chen Christopher G. Brinton Changchuan Yin Walid Saad Shuguang Cui Beijing Laboratory of Advanced Information Network Beijing University of Posts and Telecommunications Beijing China State Key Laboratory Of Networking And Switching Technology Beijing University of Posts and Telecommunications Beijing China Department of Electrical and Computer Engineering Institute for Data Science and Computing University of Miami Coral Gables FL USA School of Electrical and Computer Engineering Purdue University West Lafayette IN USA Bradley Department of Electrical and Computer Engineering Virginia Tech Arlington VA USA Shenzhen Research Institute of Big Data (SRIBD) and the Future Network of Intelligence Institute (FNii) Chinese University of Hong Kong Shenzhen China

ISBN: (纸本)9781665435413

This paper considers improving wireless communication and computation efficiency in federated learning (FL) via model quantization. In the proposed bitwidth FL scheme, edge devices train and transmit quantized versions of their local FL model parameters to a coordinating server, which, in turn, aggregates them into a quantized global model and synchronizes the devices. With the goal of jointly determining the set of participating devices in each training iteration and the bitwidths employed at the devices, we pose an optimization problem for minimizing the training loss of quantized FL under a device sampling budget and delay requirement. Our analytical results show that the improvement of FL training loss between two consecutive iterations depends on not only the device selection and quantization scheme, but also on several parameters inherent to the model being learned. As a result, we propose, a model-based reinforcement learning (RL) method to optimize action selection over iterations. Compared to model-free RL, the proposed approach leverages the derived mathematical characterization of the FL training process to discover an effective device selection and quantization scheme without imposing additional device communication overhead. Numerical evaluations show that the proposed FL framework can achieve the same classification performance while reducing the number of training iterations needed for convergence by 20% compared to model-free RL-based FL.

关键词： Training Performance evaluation Wireless communication Quantization (signal) Computational modeling Reinforcement learning Mathematical models

Real-Time Near-Term Iterative Assimilation and Forecasting System of Terrestrial Ecosystem Carbon Cycle (Cafs 1.0)

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Ren, Xiaoli He, Honglin Wan, Meng Nie, Ningming Liu, Weihua Xu, Qian Shan, Rui Lin, Zining Cao, Rongqiang Wang, Yangang Cao, Naixun Wu, Xiaojing Ge, Rong Yang, Qinmeng Tang, Xinzhai Zhang, Li Zhang, Qianmei Key Laboratory of Ecosystem Network Observation and Modeling Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing China National Ecosystem Science Data Center Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing China College of Resources and Environment University of Chinese Academy of Sciences Beijing China Computer Network Information Center Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Technology Co. Ltd China Institute of Natural Resources and Environmental Audits School of Government Audit Nanjing Audit University Nanjing China Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystem South China Botanical Garden Chinese Academy of Sciences Guangzhou China

The recent development of iterative forecasting provides an opportunity to inform management by forecasting ecosystem dynamics in advance, yet the iterative prediction focusing on ecosystem carbon dynamics is still poorly studied. In this study, we developed a real-time near-term iterative Assimilation and Forecasting System of terrestrial ecosystem Carbon cycle (CAFS 1.0) to realize the near-term iterative prediction of the site scale carbon dynamics (in the next 7 days), combining the pathway of real-time field observation, model simulation, and iterative forecating to provide support for carbon management. The system has been successfully deployed and applied in four forest field sites in China, helping assess the upcoming carbon sink under climatic changes. This provides an example of the transformation of field site scientific research mode from observation to forecasting, and can be applied in more sites, thereby facilitating regional real-time near-term carbon forecasting in the future. © 2023, The Authors. All rights reserved.

关键词： Forecasting

RadioGAT: A Joint Model-based and data-driven Framework for Multi-band Radiomap Reconstruction via Graph Attention networks

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Xiaojie Zhang, Songyang Li, Hang Li, Xiaoyang Xu, Lexi Xu, Haigao Mei, Hui Zhu, Guangxu Qi, Nan Xiao, Ming Shenzhen Research Institute of Big Data The Chinese University of Hong Kong-Shenzhen Guangdong China College of Physics Nanjing University of Aeronautics and Astronautics Nanjing210016 China Department of Electrical and Computer Engineering University of Louisiana at Lafayette LafayetteLA70503 United States Research Institute China United Network Communications Corporation Beijing100048 China China Mobile Communications Group Jiangxi Co. Ltd 330038 China Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space Nanjing University of Aeronautics and Astronautics Nanjing210016 China State Key Laboratory of Integrated Services Networks Xidian University Xi'An710071 China Department of Information Science and Engineering School of Electrical Engineering of KTH Royal Institute of Technology Stockholm Sweden

Multi-band radiomap reconstruction (MB-RMR) is a key component in wireless communications for tasks such as spectrum management and network planning. However, traditional machine-learning-based MB-RMR methods, which rely heavily on simulated data or complete structured ground truth, face significant deployment challenges. These challenges stem from the differences between simulated and actual data, as well as the scarcity of real-world measurements. To address these challenges, our study presents RadioGAT, a novel framework based on Graph Attention network (GAT) tailored for MBRMR within a single area, eliminating the need for multiregion datasets. RadioGAT innovatively merges model-based spatial-spectral correlation encoding with data-driven radiomap generalization, thus minimizing the reliance on extensive data sources. The framework begins by transforming sparse multiband data into a graph structure through an innovative encoding strategy that leverages radio propagation models to capture the spatial-spectral correlation inherent in the data. This graph-based representation not only simplifies data handling but also enables tailored label sampling during training, significantly enhancing the framework's adaptability for deployment. Subsequently, The GAT is employed to generalize the radiomap information across various frequency bands. Extensive experiments using raytracing datasets based on real-world environments have demonstrated RadioGAT's enhanced accuracy in supervised learning settings and its robustness in semi-supervised scenarios. These results underscore RadioGAT's effectiveness and practicality for MBRMR in environments with limited data availability. Copyright © 2024, The Authors. All rights reserved.

关键词： Encoding (symbols)

Strong quantum nonlocality without entanglement in multipartite quantum systems

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Yuan, Pei Tian, Guojing Sun, Xiaoming CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

In this paper, we generalize the concept of strong quantum nonlocality from two aspects. Firstly in Cd ⊗ Cd ⊗ Cd quantum system, we present a construction of strongly nonlocal quantum states containing 6(d−1)2 orthogonal product states, which is one order of magnitude less than the number of basis states d3. Secondly, we give the explicit form of strongly nonlocal orthogonal product basis in C3 ⊗ C3 ⊗ C3 ⊗ C3 quantum system, where four is the largest known number of subsystems in which there exists strong quantum nonlocality up to now. Both the two results positively answer the open problems in [Halder, et al., PRL, 122, 040403 (2019)], that is, there do exist and even smaller number of quantum states can demonstrate strong quantum nonlocality without entanglement. Copyright © 2020, The Authors. All rights reserved.

关键词： Quantum communication

Meta Mask Correction for Nuclei Segmentation in Histopathological Image

学校读者我要写书评

暂无评论

Meta Mask Correction for Nuclei Segmentation in Histopatholo...

2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021

作者： Shi, Jiangbo Jia, Chang Gao, Zeyu Gong, Tieliang Wang, Chunbao Li, Chen Xi'An Jiao Tong University National Engineering Lab for Big Data Analytics Shaanxi Xi'an710049 China School of Computer Science and Technology Xi'An Jiao Tong University Shaanxi Xi'an710049 China Shaanxi Province Key Laboratory of Satellite and Terrestrial Network Tech.RandD Xi'An Jiao Tong University Shaanxi Xi'an710049 China Department of Pathology The First Affiliated Hospital of xi'An Jiaotong University Shaanxi Xi'an710061 China

ISBN: (纸本)9781665401265

Nuclei segmentation is a fundamental task in digital pathology analysis and can be automated by deep learning-based methods. However, the development of such an automated method requires a large amount of data with precisely annotated masks which is hard to obtain. Training with weakly labeled data is a popular solution for reducing the workload of annotation. In this paper, we propose a novel meta-learning-based nuclei segmentation method which follows the label correction paradigm to leverage data with noisy masks. Specifically, we design a fully conventional meta-model that can correct noisy masks using a small amount of clean meta-data. Then the corrected masks can be used to supervise the training of the segmentation model. Meanwhile, a bi-level optimization method is adopted to alternately update the parameters of the main segmentation model and the meta-model in an end-to-end way. Extensive experimental results on two nuclear segmentation datasets show that our method achieves the state-of-the-art result. It even achieves comparable performance with the model training on supervised data in some noisy settings. © 2021 IEEE.

关键词： Deep learning

Renal Clearing Carbon Dots-Based Near-Infrared Fluorescent Super-Small Nanoprobe for Renal Imaging

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Gao, Pengli Hui, Hui Guo, Chanjuan Liu, Yu Su, Ya Huang, Xiazi Shang, Wenting Jiang, Jingying Tian, Jie Guo, Kunxiong CAS Key Laboratory of Molecular Imaging Beijing Key Laboratory of Molecular Imaging The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China School of Biological Science and Medicine Engineering School of Engineering Medicine Beihang University No. 37 Xueyuan Road Beijing100191 China Key Laboratory of Big Data-Based Precision Medicine Beihang University Ministry of Industry and Information Technology of the People’s Republic of China No. 37 Xueyuan Road Beijing100191 China School of Engineering Medicine Beihang University No. 37 Xueyuan Road Beijing100191 China Department of Hepatobiliary Surgery Zhujiang Hospital Southern Medical University Guangzhou China Beijing Advanced Innovation Center for Big Data-Based Precision Medicine China

Real-time renal imaging is indispensable for monitoring kidney-related disease processes and evaluating the nephrotoxicity of nanoprobes. Owing to their small size, tunable photoluminescence properties, and low biotoxicity, carbon dots (CDs) are widely employed as nanoprobes for imaging-guided disease diagnosis. However, the preparation of near-infrared CDs is a stupendous challenging. Here, CDs with near-infrared fluorescence (NIR-CDs ) were prepared by one-pot pyrolysis and were comprehensively evaluated for real-time renal imaging. The NIR-CDs possessed super-small size (1.8 ± 0.2 nm) and high quantum yield (14.3% in aqueous solution) at 696 nm. NIR-CDs exhibited high-efficiency renal clearance at 24 h post-treatment (97.47-99.27% ID) via glomerular filtration. The irreversible clearance of NIR-CDs enables effective real-time monitoring of kidney function in the normal and acute kidney injury (AKI) mouse models. Consequently, NIR-CDs can be utilized for imaging analyses of impaired kidney function, pioneering a serviceable guideline for neoapplications of NIR-CDs. © 2022, The Authors. All rights reserved.

关键词： Nanoprobes