检索结果-内蒙古大学图书馆

Binary neural networks: A survey

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Qin, Haotong Gong, Ruihao Liu, Xianglong Bai, Xiao Song, Jingkuan Sebe, Nicu State Key Lab of Software Development Environment Beihang University Beijing China Beijing Advanced Innovation Center for Big Data-Based Precision Medicine Beihang University Beijing China Center for Future Media School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Department of Information Engineering and Computer Science University of Trento Trento Italy School of Computer Science and Engineering Beijing Advanced Innovation Center for Big Data and Brain Computing Jiangxi Research Institute Beihang University Beijing China

The binary neural network, largely saving the storage and computation, serves as a promising technique for deploying deep models on resource-limited devices. However, the binarization inevitably causes severe information loss, and even worse, its discontinuity brings difficulty to the optimization of the deep network. To address these issues, a variety of algorithms have been proposed, and achieved satisfying progress in recent years. In this paper, we present a comprehensive survey of these algorithms, mainly categorized into the native solutions directly conducting binarization, and the optimized ones using techniques like minimizing the quantization error, improving the network loss function, and reducing the gradient error. We also investigate other practical aspects of binary neural networks such as the hardware-friendly design and the training tricks. Then, we give the evaluation and discussions on different tasks, including image classification, object detection and semantic segmentation. Finally, the challenges that may be faced in future research are prospected. Copyright © 2020, The Authors. All rights reserved.

关键词： Surveys

Learning to augment expressions for few-shot fine-grained facial expression recognition

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Wang, Wenxuan Fu, Yanwei Sun, Qiang Chen, Tao Cao, Chenjie Zheng, Ziqi Xu, Guoqiang Qiu, Han Jiang, Yu-Gang Xue, Xiangyang School of Computer Science Shanghai Key Lab of Intelligent Information Processing Fudan University Shanghai China School of Data Science Shanghai Key Lab of Intelligent Information Processing Fudan University Shanghai China Academy for Engineering & Technology Fudan University Shanghai China School of Information Science and Technology Fudan University Shanghai China Ping An OneConnect Shanghai China

—Affective computing and cognitive theory are widely used in modern human-computer interaction scenarios. Human faces, as the most prominent and easily accessible features, have attracted great attention from researchers. Since humans have rich emotions and developed musculature, there exist a lot of subtle and fine-grained expressions in real-world applications. However, it is extremely time-consuming to collect and annotate a large number of facial images, of which some subtle expressions may even require psychologists to correctly categorize them. To the best of our knowledge, the existing expression datasets are only limited to several basic facial expressions, which are not sufficient to support our ambitions in developing successful human-computer interaction systems. To this end, a novel Fine-grained Facial Expression database – F2ED is contributed in this paper;and such a dataset costs one year to be collected by us and annotated with the help of psychological annotators. Totally, it includes more than 200k images with 54 facial expressions from 119 persons. So far as we know, this is the first large dataset to label faces with subtle emotion changes for the recognition of facial expressions. Considering the phenomenon of uneven data distribution and lack of samples is common in real-world scenarios, we further evaluate several tasks of few-shot expression learning by virtue of our F2ED, which are to recognize the facial expressions given only few training instances. These tasks mimic human performance to learn robust and general representation from few examples. To address such few-shot tasks, we propose a unified task-driven framework – Compositional Generative Adversarial Network (Comp-GAN) learning to synthesize facial images and thus augmenting the instances of few-shot expression classes. Essentially, Comp-GAN consists of two generators: one for editing faces with desired expression and the other for changing the face posture;so it can generate many re

关键词： Generative adversarial networks

Climatic and Biotic Controls of Evapotranspiration Across Grassland Ecosystems on Tibetan Plateau

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Peng, Siyuan Yang, Yue Luo, Dengnan Zeng, Xiang Liang, Minqi Tao, Long Zhang, Guangru Li, Pan Liao, Weijie Guo, Qun Cao, Ruochen Li, Yuzhe Zhang, Weirong Hu, Zhongmin School of Geography South China Normal University Shipai Campus Guangzhou510631 China Key Laboratory of Agro-Forestry Environmental Processes and Ecological Regulation School of Ecology and Environment Hainan University Haikou570228 China Center for Eco-Environment Restoration Engineering of Hainan Province Hainan University Haikou570228 China Facility of Geographical Science Beijing Normal University Beijing100875 China Zhuhai Branch of State Key Laboratory of Earth Surface Processes and Resource Ecology Beijing Normal University Zhuhai519087 China School of Atmospheric Sciences Guangdong Province Data Center of Terrestrial and Marine Ecosystems Carbon Cycle Sun Yat-sen University Guangdong Zhuhai510245 China Key Lab of Ecosystem Network Observation and Modeling Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing100101 China College of Resources and Environment University of Chinese Academy of Sciences Beijing101408 China International Institute for Earth System Sciences Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application Nanjing University Jiangsu210023 China Jiangsu Provincial Key Laboratory of Geographic Information Science and Technology Key Laboratory for Land Satellite Remote Sensing Applications of Ministry of Natural Resources School of Geography and Ocean Science Nanjing University Jiangsu210023 China Key Laboratory of Land Surface Pattern and Simulation Institute of Geographical Sciences and Natural Resources Research Chinese Academy of Sciences Beijing China

Accurately quantifying spatiotemporal variations in evapotranspiration (ET) and its components on Tibetan Plateau (TP) is crucial for understanding the regional water cycle and energy balance. However, the determinant of spatial variations in ET and its component processes in grassland ecosystems of TP remain unclear. In this study, we used 14 sites observation data and applied Shuttleworth-Wallace-Hu (SWH) model to investigate the spatial relationships between ET, its components, and influencing factors. Our results highlighted the good performance of SWH model. There were corresponding drops in ET on alpine meadow and on alpine steppe. And the ratio of ET and precipitation (ET/PPT) of most sites was less than 1 in alpine meadow. Furthermore, the ratio of transpiration (T) and ET (T/ET) in alpine meadow generally exceeded that in alpine steppe. PPT emerged as the primary driver of interannual spatial variations in ET, while LAI played a crucial role in regulating T/ET. Our results further demonstrated that in the grassland ecosystem of TP, as the aridity index increased, the ET gradually transitioned from being primarily influenced by water supply to energy demand. This transition occurred at an aridity index of 0.73 approximately. The results of this study imply that in the context of global warming, with aridity index in TP decreasing, the ET of more ecosystems will shift from energy-led to water-led. Such findings are of immense significance in enhancing our understanding of the intricate interactions of the water cycle between land and the atmosphere. © 2023, The Authors. All rights reserved.

关键词： Water supply

On dynamic time division duplex transmissions for small cell networks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ding, Ming López-Pérez, David Xue, Ruiqi Vasilakos, Athanasios V. Chen, Wen Data 61 Australia Bell Labs Alcatel- Lucent Ireland Shanghai Jiao Tong University China Department of Computer Science Electrical and Space Engineering Lule17University of Technology Sweden Shanghai Key Lab of Navigation and Location Based Services Shanghai Jiao Tong University School of Electronic Engineering and Automation Guilin University of Electronic Technology China

Motivated by the promising benefits of dynamic Time Division Duplex (TDD), in this paper, we use a unified framework to investigate both the technical issues of applying dynamic TDD in homogeneous small cell networks (HomSCNs), and the feasibility of introducing dynamic TDD into heterogeneous networks (HetNets). First, HomSCNs are analyzed, and a small cell BS scheduler that dynamically and independently schedules DL and UL subframes is presented, such that load balancing between the DL and the UL traffic can be achieved. Moreover, the effectiveness of various inter-link interference mitigation (ILIM) schemes as well as their combinations, is systematically investigated and compared. Besides, the interesting possibility of partial interference cancellation (IC) is also explored. Second, based on the proposed schemes, the joint operation of dynamic TDD together with cell range expansion (CRE) and almost blank subframe (ABS) in HetNets is studied. In this regard, scheduling polices in small cells and an algorithm to derive the appropriate macrocell traffic off-load and ABS duty cycle under dynamic TDD operation are proposed. Moreover, the full IC and the partial IC schemes are investigated for dynamic TDD in HetNets. The user equipment (UE) packet throughput performance of the proposed/discussed schemes is benchmarked using system-level simulations. Copyright © 2020, The Authors. All rights reserved.

关键词： Heterogeneous networks

ZSTAD: Zero-Shot Temporal Activity Detection

学校读者我要写书评

暂无评论

ZSTAD: Zero-Shot Temporal Activity Detection

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Lingling Zhang Xiaojun Chang Jun Liu Minnan Luo Sen Wang Zongyuan Ge Alexander Hauptmann School of Computer Science and Technology Xi'an Jiaotong University Xian China Ministry of Education Key Lab For Intelligent Networks and Network Security Xian China Faculty of Information Technology Monash University Australia National Engineering Lab for Big Data Analytics Xi'an Jiaotong University Xian China School of Information Technology and Electrical Engineering The University of Queensland Australia School of Computer Science Carnegie Mellon University USA

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

An integral part of video analysis and surveillance is temporal activity detection, which means to simultaneously recognize and localize activities in long untrimmed videos. Currently, the most effective methods of temporal activity detection are based on deep learning, and they typically perform very well with large scale annotated videos for training. However, these methods are limited in real applications due to the unavailable videos about certain activity classes and the time-consuming data annotation. To solve this challenging problem, we propose a novel task setting called zero-shot temporal activity detection (ZSTAD), where activities that have never been seen in training can still be detected. We design an end-to-end deep network based on R-C3D as the architecture for this solution. The proposed network is optimized with an innovative loss function that considers the embeddings of activity labels and their super-classes while learning the common semantics of seen and unseen activities. Experiments on both the THUMOS'14 and the Charades datasets show promising performance in terms of detecting unseen activities.

关键词： Semantics Proposals Training Task analysis Feature extraction Microsoft Windows Three-dimensional displays

On-edge multi-task transfer learning: Model and practice with data-driven task allocation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zheng, Zimu Chen, Qiong Hu, Chuang Wang, Dan Liu, Fangming The National Engineering Research Center for Big Data Technology and System Key Laboratory of Services Computing Technology and System Ministry of Education School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China The Edge Cloud Innovation Lab. Technical Innovation Department Cloud BU Huawei Technologies Co. Ltd. Shenzhen China The Department of Computing Hong Kong Polytechnic University Kowloon Hong Kong Hong Kong

On edge devices, data scarcity occurs as a common problem where transfer learning serves as a widely-suggested remedy. Nevertheless, transfer learning imposes heavy computation burden to the resource-constrained edge devices. Existing task allocation works usually assume all submitted tasks are equally important, leading to inefficient resource allocation at a task level when directly applied in Multi-task Transfer Learning (MTL). To address these issues, we first reveal that it is crucial to measure the impact of tasks on overall decision performance improvement and quantify task importance. We then show that task allocation with task importance for MTL (TATIM) is a variant of NP-complete Knapsack problem, where the complicated computation to solve this problem needs to be conducted repeatedly under varying contexts. To solve TATIM with high computational efficiency, we propose a data-driven Cooperative Task Allocation (DCTA) approach. Finally, we evaluate the performance of DCTA by not only a trace-driven simulation, but also a new comprehensive real-world AIOps case study which bridges model and practice via a new architecture and main components design within AIOps system. Extensive experiments show that our DCTA reduces 3.24 times of processing time, and saves 48.4% energy consumption compared with the state-of-the-art when solving TATIM. Copyright © 2021, The Authors. All rights reserved.

关键词： Energy utilization

A study on the uncertainty of convolutional layers in deep neural networks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Shen, Haojing Chen, Sihong Wang, Ran Big Data Institute College of Computer Science and Software Engineering Guangdong Key Lab. of Intelligent Information Processing Shenzhen University ShenzhenGuangdong518060 China College of Mathematics and Statistics Shenzhen University Shenzhen518060 China Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China

This paper shows a Min-Max property existing in the connection weights of the convolutional layers in a neural network structure, i.e., the LeNet. Specifically, the Min-Max property means that, during the back propagation-based training for LeNet, the weights of the convolutional layers will become far away from their centers of intervals, i.e., decreasing to their minimum or increasing to their maximum. From the perspective of uncertainty, we demonstrate that the Min-Max property corresponds to minimizing the fuzziness of the model parameters through a simplified formulation of convolution. It is experimentally confirmed that the model with the Min-Max property has a stronger adversarial robustness, thus this property can be incorporated into the design of loss function. This paper points out a changing tendency of uncertainty in the convolutional layers of LeNet structure, and gives some insights to the interpretability of convolution. © 2020, CC-BY.

关键词： Convolution

Syntax-enhanced Pre-trained model

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Xu, Zenan Guo, Daya Tang, Duyu Su, Qinliang Shou, Linjun Gong, Ming Zhong, Wanjun Quan, Xiaojun Jiang, Daxin Duan, Nan School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Microsoft Research Asia Beijing China Microsoft Search Technology Center Asia Beijing China Guangdong Key Laboratory Big Data Analysis and Processing Guangzhou China Key Lab. of Machine Intelligence and Advanced Computing Ministry of Education China

We study the problem of leveraging the syntactic structure of text to enhance pre-trained models such as BERT and RoBERTa. Existing methods utilize syntax of text either in the pre-training stage or in the fine-tuning stage, so that they suffer from discrepancy between the two stages. Such a problem would lead to the necessity of having human-annotated syntactic information, which limits the application of existing methods to broader scenarios. To address this, we present a model that utilizes the syntax of text in both pre-training and fine-tuning stages. Our model is based on Transformer with a syntax-aware attention layer that considers the dependency tree of the text. We further introduce a new pre-training task of predicting the syntactic distance among tokens in the dependency tree. We evaluate the model on three downstream tasks, including relation classification, entity typing, and question answering. Results show that our model achieves state-of-the-art performance on six public benchmark datasets. We have two major findings. First, we demonstrate that infusing automatically produced syntax of text improves pre-trained models. Second, global syntactic distances among tokens bring larger performance gains compared to local head relations between contiguous tokens. Copyright © 2020, The Authors. All rights reserved.

关键词： Syntactics

ReS2TIM: Reconstruct Syntactic Structures from Table Images

学校读者我要写书评

暂无评论

ReS2TIM: Reconstruct Syntactic Structures from Table Images

International Conference on Document Analysis and Recognition

作者： Wenyuan Xue Qingyong Li Dacheng Tao Beijing Key Lab of Transportation Data Analysis and Mining Beijing Jiaotong University China Faculty of Engineering The University of Sydney Australia

Tables often represent densely packed but structured data. Understanding table semantics is vital for effective information retrieval and data mining. Unlike web tables, whose semantics are readable directly from markup language and contents, the full analysis of tables published as images requires the conversion of discrete data into structured information. This paper presents a novel framework to convert a table image into its syntactic representation through the relationships between its cells. In order to reconstruct the syntactic structures of a table, we build a cell relationship network to predict the neighbors of each cell in four directions. During the training stage, a distance-based sample weight is proposed to handle the class imbalance problem. According to the detected relationships, the table is represented by a weighted graph that is then employed to infer the basic syntactic table structure. Experimental evaluation of the proposed framework using two datasets demonstrates the effectiveness of our model for cell relationship detection and table structure inference.

关键词： Syntactics Visualization Semantics Feature extraction data mining Training Magnetic resonance imaging