检索结果-内蒙古大学图书馆

CWLP:coordinated warp scheduling and locality-protected cache allocation on GPUs

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2018年第2期19卷 206-220页

作者： Yang ZHANG Zuo-cheng XING Cang LIU Chuan TANG National Laboratory for Parallel and Distributed Processing National University of Defense Technology

As we approach the exascale era in supercomputing, designing a balanced computer system with a powerful computing ability and low power requirements has becoming increasingly important. The graphics processing unit（GPU） is an accelerator used widely in most of recent supercomputers. It adopts a large number of threads to hide a long latency with a high energy efficiency. In contrast to their powerful computing ability, GPUs have only a few megabytes of fast on-chip memory storage per streaming multiprocessor（SM）. The GPU cache is inefficient due to a mismatch between the throughput-oriented execution model and cache hierarchy design. At the same time, current GPUs fail to handle burst-mode long-access latency due to GPU＇s poor warp scheduling ***, benefits of GPU＇s high computing ability are reduced dramatically by the poor cache management and warp scheduling methods, which limit the system performance and energy efficiency. In this paper, we put forward a coordinated warp scheduling and locality-protected（CWLP） cache allocation scheme to make full use of data locality and hide latency. We first present a locality-protected cache allocation method based on the instruction program counter（LPC） to promote cache performance. Specifically, we use a PC-based locality detector to collect the reuse information of each cache line and employ a prioritised cache allocation unit（PCAU） which coordinates the data reuse information with the time-stamp information to evict the lines with the least reuse possibility. Moreover, the locality information is used by the warp scheduler to create an intelligent warp reordering scheme to capture locality and hide latency. Simulation results show that CWLP provides a speedup up to 19.8% and an average improvement of 8.8% over the baseline methods.

关键词： Locality Graphics processing unit （GPU） Cache allocation Warp scheduling

VISUAL CONFUSION LABEL TREE FOR IMAGE CLASSIFICATION

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Liu, Yuntao Dou, Yong Jin, Ruochun Li, Rongchun National University of Defense Technology National Laboratory for Parallel and Distributed Processing Changsha410073 China

Convolution neural network models are widely used in image classification tasks. However, the running time of such models is so long that it is not the conforming to the strict real-time requirement of mobile devices. In order to optimize models and meet the requirement mentioned above, we propose a method that replaces the fully-connected layers of convolution neural network models with a tree classifier. Specifically, we construct a Visual Confusion Label Tree based on the output of the convolution neural network models, and use a multi-kernel SVM plus classifier with hierarchical constraints to train the tree classifier. Focusing on those confusion subsets instead of the entire set of categories makes the tree classifier more discriminative and the replacement of the fully-connected layers reduces the original running time. Experiments show that our tree classifier obtains a significant improvement over the state-of-the-art tree classifier by 4.3% and 2.4% in terms of top-1 accuracy on CIFAR-100 and ImageNet datasets respectively. Additionally, our method achieves 124× and 115× speedup ratio compared with fully-connected layers on AlexNet and VGG16 without accuracy decline. Copyright © 2019, The Authors. All rights reserved.

关键词： Image classification

Improving unsupervised domain adaptation by reducing bi-level feature redundancy

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Wang, Mengzhu Zhang, Xiang Lan, Long Wang, Wei Tan, Huibin Luo, Zhigang Science and Technology on Parallel and Distributed Laboratory College of Computer National University of Defense Technology Changsha China Institute for Quantum State Key Laboratory of High Performance Computing National University of Defense Technology Changsha China DUT-RU International School of Information Science & Engineering Dalian University of Technology DalianLiaoning116000 China Department of Science and Technology on Parallel and Distributed Processing National University of Defense Technology Changsha China

Reducing feature redundancy has shown beneficial effects for improving the accuracy of deep learning models, thus it is also indispensable for the models of unsupervised domain adaptation (UDA). Nevertheless, most recent efforts in the field of UDA ignore this point. Moreover, main schemes realizing this in general independent of UDA purely involve a single domain, thus might not be effective for cross-domain tasks. In this paper, we emphasize the significance of reducing feature redundancy for improving UDA in a bi-level way. For the first level, we try to ensure compact domain-specific features with a transferable decorrelated normalization module, which preserves specific domain information whilst easing the side effect of feature redundancy on the sequel domain-invariance. In the second level, domain-invariant feature redundancy caused by domain-shared representation is further mitigated via an alternative brand orthogonality for better generalization. These two novel aspects can be easily plugged into any BN-based backbone neural networks. Specifically, simply applying them to ResNet50 has achieved competitive performance to the state-of-the-arts on five popular benchmarks. Our code will be available at https://***/dreamkily/gUDA. © 2020, CC BY.

关键词： Deep learning

Succinct representations in collaborative filtering: A case study using wavelet tree on 1,000 cores 20

学校读者我要写书评

暂无评论

Succinct representations in collaborative filtering: A case ...

20th International Conference on parallel and distributed Computing, Applications and Technologies, PDCAT 2019

作者： Peng, Xiangjun Wang, Qingfeng Sun, Xu Gong, Chunye Wang, Yaohua User-Centric Computing Group University of Nottingham Ningbo China China Department of Computer Science and Technology National University of Defense Technology China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology China

ISBN: (纸本)9781728126166

User-Item (U-I) matrix has been used as the dominant data infrastructure of Collaborative Filtering (CF). To reduce space consumption in runtime and storage, caused by data sparsity and growing need to accommodate side information in CF design, one needs to go beyond the U-I Matrix. In this paper, we took a case study of Succinct Representations in Collaborative Filtering, rather than using a U-I Matrix. Our key insight is to introduce Succinct Data Structures as a new infrastructure of CF. Towards this, we implemented a User-based K-Nearest-Neighbor CF prototype via Wavelet Tree, by first designing a Accessible Compressed Documents (ACD) to compress U-I data in Wavelet Tree, which is efficient in both storage and runtime. Then, we showed that ACD can be applied to develop an efficient intersection algorithm without decompression, by taking advantage of ACD's characteristics. We evaluated our design on 1,000 cores of Tianhe-II supercomputer, with one of the largest public data set ml-20m. The results showed that our prototype could achieve 3.7 minutes on average to deliver the results. © 2019 IEEE.

关键词： Collaborative filtering

Visual Tree Convolutional Neural Network in Image Classification

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Liu, Yuntao Dou, Yong Jin, Ruochun Qiao, Peng National University of Defense Technology National Laboratory for Parallel and Distributed Processing Hunan Changsha410073 China

In image classification, Convolutional Neural Network(CNN) models have achieved high performance with the rapid development in deep learning. However, some categories in the image datasets are more difficult to distinguished than others. Improving the classification accuracy on these confused categories is benefit to the overall performance. In this paper, we build a Confusion Visual Tree(CVT) based on the confused semantic level information to identify the confused categories. With the information provided by the CVT, we can lead the CNN training procedure to pay more attention on these confused categories. Therefore, we propose Visual Tree Convolutional Neural Networks(VT-CNN) based on the original deep CNN embedded with our CVT. We evaluate our VT-CNN model on the benchmark datasets CIFAR-10 and CIFAR-100. In our experiments, we build up 3 different VT-CNN models and they obtain improvement over their based CNN models by 1.36%, 0.89% and 0.64%, respectively. Copyright © 2019, The Authors. All rights reserved.

关键词： Image classification

Exploring frame segmentation networks for temporal action localization

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Yang, Ke Shen, Xiaolong Qiao, Peng Li, Shijie Li, Dongsheng Dou, Yong National Laboratory for Parallel and Distributed Processing College of Computer National University of Defense Technology Changsha China

Temporal action localization is an important task of computer vision. Though many methods have been proposed, it still remains an open question how to predict the temporal location of action segments precisely. Most state-of-the-art works train action classifiers on video segments pre-determined by action proposal. However, recent work found that a desirable model should move beyond segment-level and make dense predictions at a fine granularity in time to determine precise temporal boundaries. In this paper, we propose a Frame Segmentation Network (FSN) that places a temporal CNN on top of the 2D spatial CNNs. Spatial CNNs are responsible for abstracting semantics in spatial dimension while temporal CNN is responsible for introducing temporal context information and performing dense predictions. The proposed FSN can make dense predictions at frame-level for a video clip using both spatial and temporal context information. FSN is trained in an end-to-end manner, so the model can be optimized in spatial and temporal domain jointly. We also adapt FSN to use it in weakly supervised scenario (WFSN), where only video level labels are provided when training. Experiment results on public dataset show that FSN achieves superior performance in both frame-level action localization and temporal action localization. Copyright © 2019, The Authors. All rights reserved.

关键词： Forecasting

Container-Based Complex Programming Skills Training Platform

学校读者我要写书评

暂无评论

Container-Based Complex Programming Skills Training Platform

IEEE International Conference on Software Engineering and Service Sciences (ICSESS)

作者： Wei Wang Tao Wang Gang Yin National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha Hunan Province China

ISBN: (数字)9781728109459

ISBN: (纸本)9781728109466

In recent years, online programming education and evaluation have developed rapidly. The traditional online education platform focuses on the teaching of knowledge, lacking a good evaluation platform to assess the learning outcomes, while the existing online evaluation platform only focuses on the evaluation of algorithms or languages. Until now, no platform can support the training of engineering-level complex technologies. In this paper, we present our solution - Educoder. Educoder is an one-stop service platform that integrates “teaching”, “learning” and “use”. It not only supports simple programming language training type, but also supports the training of complex technology such as Hadoop, Spring, Machine Learning, Docker, etc. The code environment for student learning and evaluation is managed by Gitlab, and the runtime environment is provided by Kubernetes pod, which guarantees good isolation. For complex types of training, we provide the Docker Web Terminal module for SSH connection to enter the pod for operation and use docker in docker technology to provide support for training that requires multiple services or multiple nodes. In Educoder, we use scripts to serialize the training evaluation steps, and we formed a private evaluation script template library for training teacher to reference, which greatly reduced the time cost of writing a new training course.

关键词：

Quantitative analysis of collaborative-based annotation in software engineering education 8

学校读者我要写书评

暂无评论

Quantitative analysis of collaborative-based annotation in s...

2018 8th International Workshop on Computer Science and Engineering, WCSE 2018

作者： Yu, Jie Wang, RenMin Zeng, LingBin Wang, Tao Yin, Gang Fan, Qiang Yu, Yue National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha China

ISBN: (纸本)9789811178610

The primary goal of software engineering education is to develop the ability of students, a common method is to let students to read excellent open source code, so that students can improve themselves through this behavior. When reading the code, we tend to add comments to the code, in the past, the labeling of students tend to be an independent behavior, in this paper we analyze the quantitative of collaborative-based code annotation method, and the result shows the collaborative-based code annotation method is working well in improving students programming skills. © Proceedings of 2018 the 8th International Workshop on Computer Science and Engineering, WCSE 2018. All rights reserved.

关键词： Open source software

Audio Tagging by Cross Filtering Noisy Labels

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhu, Boqing Xu, Kele Kong, Qiuqiang Wang, Huaimin Peng, Yuxing Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha410073 China Centre for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

—High quality labeled datasets have allowed deep learning to achieve impressive results on many sound analysis tasks. Yet, it is labor-intensive to accurately annotate large amount of audio data, and the dataset may contain noisy labels in the practical settings. Meanwhile, the deep neural networks are susceptive to those incorrect labeled data because of their outstanding memorization ability. In this paper, we present a novel framework, named CrossFilter, to combat the noisy labels problem for audio tagging. Multiple representations (such as, Logmel and MFCC) are used as the input of our framework for providing more complementary information of the audio. Then, though the cooperation and interaction of two neural networks, we divide the dataset into curated and noisy subsets by incrementally pick out the possibly correctly labeled data from the noisy data. Moreover, our approach leverages the multitask learning on curated and noisy subsets with different loss function to fully utilize the entire dataset. The noisy-robust loss function is employed to alleviate the adverse effects of incorrect labels. On both the audio tagging datasets FSDKaggle2018 and FSDKaggle2019, empirical results demonstrate the performance improvement compared with other competing approaches. On FSDKaggle2018 dataset, our method achieves state-of-the-art performance and even surpasses the ensemble models. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks