检索结果-内蒙古大学图书馆

12th International Conference on Language Resources and Evaluation, LREC 2020

作者： Tahmid Rahman Laskar, Md. Huang, Jimmy Hoque, Enamul Department of Electrical Engineering and Computer Science York University TorontoON Canada School of Information Technology York University TorontoON Canada Information Retrieval and Knowledge Management Research Lab York University TorontoON Canada

ISBN: (纸本)9791095546344

Word embeddings that consider context have attracted great attention for various natural language processing tasks in recent years. In this paper, we utilize contextualized word embeddings with the transformer encoder for sentence similarity modeling in the answer selection task. We present two different approaches (feature-based and fine-tuning-based) for answer selection. In the feature-based approach, we utilize two types of contextualized embeddings, namely the Embeddings from Language Models (ELMo) and the Bidirectional Encoder Representations from Transformers (BERT) and integrate each of them with the transformer encoder. We find that integrating these contextual embeddings with the transformer encoder is effective to improve the performance of sentence similarity modeling. In the second approach, we fine-tune two pre-trained transformer encoder models for the answer selection task. Based on our experiments on six datasets, we find that the fine-tuning approach outperforms the feature-based approach on all of them. Among our fine-tuning-based models, the Robustly Optimized BERT Pretraining Approach (RoBERTa) model results in new state-of-the-art performance across five datasets. © European Language Resources Association (ELRA), licensed under CC-BY-NC

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Query Focused Abstractive Summarization via Incorporating Query Relevance and Transfer Learning with Transformer Models 33rd

Query Focused Abstractive Summarization via Incorporating Qu...

引用

33rd Canadian Conference on Artificial Intelligence, Canadian AI 2020

作者： Laskar, Md Tahmid Rahman Hoque, Enamul Huang, Jimmy Department of Electrical Engineering and Computer Science York University Toronto Canada School of Information Technology York University Toronto Canada Information Retrieval and Knowledge Management Research Lab York University Toronto Canada

ISBN: (纸本)9783030473570

In the Query Focused Abstractive Summarization (QFAS) task, the goal is to generate abstractive summaries from the source document that are relevant to the given query. In this paper, we propose a new transfer learning technique by utilizing the pre-trained transformer architecture for the QFAS task in the Debatepedia dataset. We find that the Diversity Driven Attention model (DDA), which was the first model applied on this dataset, only performs well when the dataset is augmented by creating more training instances. In contrast, without requiring any in-domain data augmentation, our proposed approach outperforms the DDA model as well as sets a new state-of-the-art result. © Springer Nature Switzerland AG 2020.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Towards Feature Distribution Alignment and Diversity Enhancement for Data-Free Quantization

Towards Feature Distribution Alignment and Diversity Enhance...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Yangcheng Gao Zhao Zhang Richang Hong Haijun Zhang Jicong Fan Shuicheng Yan School of Computer Science and Information Engineering Hefei University of Technology Hefei China Key Laboratory of Knowledge Engineering with Big Data (Ministry of Education) & Intelligent Interconnected Systems Laboratory of Anhui Province Hefei University of Technology Hefei China Department of Computer Science Harbin Institute of Technology (Shenzhen) Xili University Town Shenzhen China School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China National University of Singapore Singapore

To obtain lower inference latency and less memory footprint of deep neural networks, model quantization has been widely employed in deep model deployment, by converting the floating points to low-precision integers. However, previous methods (such as quantization aware training and post training quantization) require original data for the fine-tuning or calibration of quantized model, which makes them inapplicable to the cases that original data are not accessed due to privacy or security. This gives birth to the data-free quantization method with synthetic data generation. While current data-free quantization methods still suffer from severe performance degradation when quantizing a model into lower bit, caused by the low inter-class separability of semantic features. To this end, we propose a new and effective data-free quantization method termed ClusterQ, which utilizes the feature distribution alignment for synthetic data generation. To obtain high inter-class separability of semantic features, we cluster and align the feature distribution statistics to imitate the distribution of real data, so that the performance degradation is alleviated. Moreover, we incorporate the diversity enhancement to solve class-wise mode collapse. We also employ the exponential moving average to update the centroid of each cluster for further feature distribution improvement. Extensive experiments based on different deep models (e.g., ResNet-18 and MobileNet-V2) over the ImageNet dataset demonstrate that our proposed ClusterQ model obtains state-of-the-art performance.

关键词： Degradation Training Quantization (signal) Semantics Neural networks Memory management Data models

来源：评论

学校读者我要写书评

暂无评论

Robust Cultivated Land Extraction Using Encoder-Decoder

引用

Journal of New Media 2020年第4期2卷 149-155页

作者： Aziguli Wulamu Jingyue Sang Dezheng Zhang and Zuxian Shi Department of Computer School of Computer and Communication EngineeringUniversity of Science and Technology Beijing(USTB)Beijing100083China Beijing Key Laboratory of Knowledge Engineering for Materials Science Beijing100083China 不详

Cultivated land extraction is essential for sustainable development and *** this paper,the network we propose is based on the encoder-decoder structure,which extracts the semantic segmentation neural network of cultivated land from satellite images and uses it for agricultural automation *** encoder consists of two part:the first is the modified Xception,it can used as the feature extraction network,and the second is the atrous convolution,it can used to expand the receptive field and the context information to extract richer feature *** decoder part uses the conventional upsampling operation to restore the original *** addition,we use the combination of BCE and Loves-hinge as a loss function to optimize the Intersection over Union(IoU).Experimental results show that the proposed network structure can solve the problem of cultivated land extraction in Yinchuan City.

关键词： Semantic segmentation encoder-decoder cultivated land extraction atrous convolution

来源：评论

学校读者我要写书评

暂无评论

Multi-site MRI classification using Weighted federated learning based on Mixture of Experts domain adaptation

Multi-site MRI classification using Weighted federated learn...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Tian Bai Yingfang Zhang Yuzhao Wang Yanguo Qin Fa Zhang College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Changchun China The Department of Orthopaedics The Second Hospital of Jilin University Changchun China School of Medical Technology Beijing Institute of Technology Beijing China

ISBN: (数字)9781665468190

ISBN: (纸本)9781665468206

Deep learning often requires large amounts of data from different institutions. Federated learning, as a distributed training framework, enables multiple participants to collaboratively train models without collecting data together and hence protecting data privacy, but the datasets from different institutions usually bring the problem of domain shift, which affects the performance of the model. When addressing domain shift, previous works often use a single global model to share parameters. Therefore, we propose a novel method to train multiple public models with different structures under the federated framework to improve the reliability and robustness of the public models. And each participant keeps its own domain-tuned private model, the private model does not share parameters with other participants. We use Mixture of Experts (MoE) domain adaptation to dynamically combine different public models and private model, which utilizes the similarity between different datasets to update the parameters of the public models. We apply the proposed method to the multi-site Magnetic resonance imaging (MRI) end-to-end classification, and the experiments demonstrate its effectiveness.

关键词： Weight measurement Training Adaptation models Three-dimensional displays Federated learning Magnetic resonance imaging Biological system modeling

来源：评论

学校读者我要写书评

暂无评论

Top-k contrast order-preserving pattern mining

arXiv

引用

arXiv 2023年

作者： Wu, Youxi Meng, Yufei Li, Yan Guo, Lei Zhu, Xingquan Fournier-Viger, Philippe Wu, Xindong School of Artificial Intelligence Hebei University of Technology Tianjin300401 China Shijiazhuang University of Applied Technology Shijiazhuang050081 China School of Economics and Management Hebei University of Technology Tianjin300401 China State Key Laboratory of Reliability and Intelligence of Electrical Equipment Hebei University of Technology Tianjin300401 China Department of Computer & Electrical Engineering and Computer Science Florida Atlantic University FL33431 United States Shenzhen University Shenzhen China Research Center for Knowledge Engineering The Zhejiang Lab Hangzhou311121 China

Recently, order-preserving pattern (OPP) mining, a new sequential pattern mining method, has been proposed to mine frequent relative orders in a time series. Although frequent relative orders can be used as features to classify a time series, the mined patterns do not reflect the differences between two classes of time series well. To effectively discover the differences between time series, this paper addresses the top-k contrast OPP (COPP) mining and proposes a COPP-Miner algorithm to discover the top-k contrast patterns as features for time series classification, avoiding the problem of improper parameter setting. COPP-Miner is composed of three parts: extreme point extraction to reduce the length of the original time series, forward mining, and reverse mining to discover COPPs. Forward mining contains three steps: group pattern fusion strategy to generate candidate patterns, the support rate calculation method to efficiently calculate the support of a pattern, and two pruning strategies to further prune candidate patterns. Reverse mining uses one pruning strategy to prune candidate patterns and consists of applying the same process as forward mining. Experimental results validate the efficiency of the proposed algorithm and show that top-k COPPs can be used as features to obtain a better classification performance. Copyright © 2023, The Authors. All rights reserved.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Deep learning based Identification of Solid Waste Management in Smart Cities through Garbage Separation and Monitoring

Deep learning based Identification of Solid Waste Management...

引用

International Conference on Inventive Research in Computing Applications (ICIRCA)

作者： Abin George Sham H. Mankar N V S Suryanarayana Manish Gupta R Sathish A Priya Knowledge and technology management Division Council of Scientific & Industrial Research National Aerospace Laboratories Bangaluru Karnataka India Department of Mechanical Engineering Pimpri Chinchwad College of Engineering and Research Pune Maharashtra India Administrative Officer Central Tribal University of Andhra Pradesh Vizianagaram Andhra Pradesh India Division of Research and Development Lovely Professional University Phagwara Punjab India Department of Computer Science Shri Nehru Maha Vidayala College of Arts and Science Coimbatore Tamil Nadu India School of Architecture Koneru Lakshmaiah Educational Foundation Gundur Andhra Pradesh India

The solid waste management is the process of proper decomposition of waste materials within a period of time. This includes the collection of garbage's and then proceeded through certain measures for decomposition. There are various methods adopted in the garbage separation process. This includes the artificial intelligence techniques for the estimation and determination of the solid waste through automatic detection and separation of the garbage waste using control and sensing units. They are integrated with the internet of things to enable the two way communication system. This helps to visualize the functioning of the system adopting the digital platform. The proposed system is implemented through the smart dust bin held in every household that automatically senses the non-biodegradable and biodegradable waste materials. The classification of the waste materials are identified through the image processing techniques.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-based Model Acceleration for Cancer Classification in Whole-Slide Images

Uncertainty-based Model Acceleration for Cancer Classificati...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Zeyu Gao Anyu Mao Jialun Wu Yang Li Chunbao Wang Caixia Ding Tieliang Gong Chen Li School of Computer Science and Technology Xi’an Jiaotong University Xi’an China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University Xi’an China Department of Pathology the First Affiliated Hospital of Xi’an Jiaotong University Xi’an China Department of Pathology Shaanxi Provincial Cancer Hospital Xi’an China

ISBN: (数字)9781665468190

ISBN: (纸本)9781665468206

Computational Pathology (CPATH) offers the possibility for highly accurate and low-cost automated pathological diagnosis. However, the high time cost of model inference is one of the main issues limiting the application of CPATH methods. Due to the large size of Whole-Slide Image (WSI), commonly used CPATH methods divided a WSI into a large number of image patches at relatively high magnification, then predicted each image patch individually, which is time-consuming. In this paper, we propose a novel Uncertainty-based Model Acceleration (UMA) method for reducing the time cost of model inference, thereby relieving the deployment burden of CPATH applications. Enlightened by the slide-viewing process of pathologists, only a few high-uncertain regions are regarded as “suspicious” regions that need to be predicted at high magnification, and most of the regions in WSI are predicted at low magnification, thereby reducing the times of image patch extraction and prediction. Meanwhile, uncertainty estimation ensures prediction accuracy at low magnification. We take two fundamental CPATH classification tasks (i.e., cancer region detection and subtyping) as examples. Extensive experiments on two large-scale renal cell carcinoma classification datasets demonstrate that our UMA can significantly reduce the time cost of model inference while maintaining competitive classification performance.

关键词： Pathology Costs Uncertainty Limiting Hospitals Biological system modeling Computational modeling

来源：评论

学校读者我要写书评

暂无评论

Various Deep Learning Methods for Hyperspectral Images

Various Deep Learning Methods for Hyperspectral Images

引用

2020 International Conference on Computing and Information Technology, ICCIT 2020

作者： Sivaram, M. Shobana, S. Jeya Khan, Mujeebudheen Ramakrishnan, Jayabrabu Goel, Pallavi Murghai Maseleno, Andino Lebanese French University Department of Computer Networking Erbil Kurdistan Region Iraq College of Science Knowledge University Department of Computer Science Erbil Kurdistan Region Iraq Ibra College of Technology Department of Electrical and Electronics Oman Jazan University College of Computer Science and Information Technology Department of Information Technology and Security Saudi Arabia Galgotias University School of Computing Science and Engineering Greater Noida India Universiti Tenaga Nasional Institute of Informatics and Computing Energy Department of Business Administration and Information System Malaysia

ISBN: (纸本)9781728126807

Hyperspectral imagery is widely used in remote sensing applications that take into account thousands of spectral channel compositions over a single scene. Hyperspectral imagery requires accurate models of learning to extract the hyperspectral features in an image. Due to the presence of its spatial and spectral resolution, the image learning model presents a core challenge due to its complicated nature of image frames. In order to assist it during the learning process, several attempts have been made to address its complicated nature. However, these methods failed to provide the hyperspectral imagery with a deeper understanding. Because of the presence of mixed pixels, limited training samples and redundant data, the utilization of deep learning techniques addresses the problems. The deep learning process addresses the complex image data relationship. In this paper, various deep learning methods are studied which are used for the learning of hyperspectral imagery. Initially, we present an overview on various deep learning methods for various image processing techniques. A system review is then carried out on various hyperspectral image learning models based on deep learning. © 2020 IEEE.

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Time-resolved Hanbury Brown-Twiss interferometry of on-chip biphoton frequency combs using Vernier phase modulation

arXiv

引用

arXiv 2022年

作者： Myilswamy, Karthik V. Seshadri, Suparna Lu, Hsuan-Hao Alshaykh, Mohammed S. Liu, Junqiu Kippenberg, Tobias J. Weiner, Andrew M. Lukens, Joseph M. School of Electrical and Computer Engineering Purdue Quantum Science and Engineering Institute Purdue University West LafayetteIN47907 United States Quantum Information Science Section Oak Ridge National Laboratory Oak RidgeTN37831 United States Electrical Engineering Department King Saud University Riyadh11421 Saudi Arabia Lausanne1015 Switzerland Knowledge Enterprise Arizona State University TempeAZ85287 United States

Biphoton frequency combs (BFCs) are promising quantum sources for large-scale and high-dimensional quantum information and networking systems. In this context, the spectral purity of individual frequency bins will be critical for realizing quantum networking protocols like teleportation and entanglement swapping. Measurement of the temporal auto-correlation function of the unheralded signal or idler photons comprising the BFC is a key tool for characterizing their spectral purity and in turn verifying the utility of the biphoton state for networking protocols. Yet the experimentally obtainable precision for measuring BFC correlation functions is often severely limited by detector jitter. The fine temporal features in the correlation function—not only of practical value in quantum information, but also of fundamental interest in the study of quantum optics—are lost as a result and have remained unexplored. We propose a scheme to circumvent this challenge through electro-optic phase modulation, experimentally demonstrating time-resolved Hanbury Brown–Twiss characterization of BFCs generated from an integrated 40.5 GHz Si3N4 microring, up to a 3 × 3-dimensional two-qutrit Hilbert space. Through slight detuning of the electro-optic drive frequency from the comb’s free spectral range, our approach leverages Vernier principles to magnify temporal features which would otherwise be averaged out by detector jitter. We demonstrate our approach under both continuous-wave and pulsed pumping regimes, finding excellent agreement with theory. Importantly, our method reveals not only the collective statistics of the contributing frequency bins but also their temporal shapes—features lost in standard fully integrated auto-correlation measurements. Copyright © 2022, The Authors. All rights reserved.

关键词： Phase modulation

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：