International Conference on Deep Learning Technologies最新文献

Design and Implementation of Convolutional Neural Network Accelerator with Variable Layer-by-layer Debugging 可变逐层调试的卷积神经网络加速器的设计与实现

International Conference on Deep Learning Technologies Pub Date : 2018-06-27 DOI: 10.1145/3234804.3234806

Songpu Huang, Jingfei Jiang, Y. Dou, Liang Bai, Hao Wang, Buyue Qin

{"title":"Design and Implementation of Convolutional Neural Network Accelerator with Variable Layer-by-layer Debugging","authors":"Songpu Huang, Jingfei Jiang, Y. Dou, Liang Bai, Hao Wang, Buyue Qin","doi":"10.1145/3234804.3234806","DOIUrl":"https://doi.org/10.1145/3234804.3234806","url":null,"abstract":"Deep learning algorithms have complex network structures and numerous parameters, and are typical computation-intensive and data-intensive applications. Due to the large amount of data and the difference in accuracy, it is very difficult to realize the function and to adjust the accuracy of FPGA-based deep learning accelerator, which seriously affects the applicability of the accelerator. To this problem, this paper presents a convolutional neural network accelerator with variable layer-by-layer debugging. The accelerator framework consists of host computer, PCIE interface, DDR module, transmission control module, CNN module and variable layer-by-layer debugging module. The variable layer-by-layer debugging module consists of DRAM, FIFO, read DRAM counting module, write DRAM counting module and data alignment module. The debugging module can be assembled in any layer of the convolutional neural network, and effectively achieve layer-by-layer debugging in cooperation with host computer. Supported by this design framework, the paper implements VGG-S on a FPGA board based on the Xilinx XCKU115 chip, achieving an acceleration ratio of 24.78 compared to the CPU platform and a 14.6x performance-to-power ratio. The comprehensive results show that the hardware resource overhead with variable layer-by-layer debugging module is very small and does not affect the frequency of the convolutional network implementation. The experimental process verified the high efficiency of the implementation and debugging of the structure, which can be used as a debugging method for various pipelined convolutional networks in the future.","PeriodicalId":118446,"journal":{"name":"International Conference on Deep Learning Technologies","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123743999","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Improvement of Pruning Method for Convolution Neural Network Compression 卷积神经网络压缩中剪枝方法的改进

International Conference on Deep Learning Technologies Pub Date : 2018-06-27 DOI: 10.1145/3234804.3234824

Chongyang Liu, Qinrang Liu

引用次数: 7

Attention-based neural network for short-text question answering 基于注意力的短文本问答神经网络

International Conference on Deep Learning Technologies Pub Date : 2018-06-27 DOI: 10.1145/3234804.3234813

Yongxin Peng, B. Liu

引用次数: 9

TCR: Temporal-CNN for Reviews Based Recommendation System TCR:基于评论的推荐系统的时态cnn

International Conference on Deep Learning Technologies Pub Date : 2018-06-27 DOI: 10.1145/3234804.3234819

Yelu Mao, Xiaoyu Shi, Mingsheng Shang, Ying Zhang

{"title":"TCR: Temporal-CNN for Reviews Based Recommendation System","authors":"Yelu Mao, Xiaoyu Shi, Mingsheng Shang, Ying Zhang","doi":"10.1145/3234804.3234819","DOIUrl":"https://doi.org/10.1145/3234804.3234819","url":null,"abstract":"In recent year, it has become a popular trend that online stores encouraged their users to write review texts for shopping items. Obviously, these collected text-reviews are helpful for understanding item properties and user preferences, as well as improving the quality of recommendation. However, existing works put considerable attentions on the performance of recommendation without using the temporal information, while the customer inclinations are evolving. In this paper, we propose TCR to model user preferences and item properties by using the convolutional neural network (CNN) combined with temporal information. In details, since the item popularity and user preferences are constantly evolving, we then build a time model that to capture the influence of time evolving on the performance of recommendation and integrate the proposed time model to the original CNN recommender. Furthermore, aiming at building an effective model, we carry out the experimental analysis on the influence of four factors (i.e., word vector embedding dimension, word frequency of comment text, the depth and width of CNN model) on the performance of recommender system. Based on the theoretical analysis, we identify the key factors, and use these factors to optimize our TCR model. Finally, we conduct the experiments on the industrial dataset, i.e., Amazon. It demonstrates that our proposed model has achieved better results than the existing models in terms of prediction accuracy.","PeriodicalId":118446,"journal":{"name":"International Conference on Deep Learning Technologies","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114829913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Multi-Objective Deep CNN for Outdoor Auto-Navigation 用于户外自动导航的多目标深度CNN

International Conference on Deep Learning Technologies Pub Date : 2018-06-27 DOI: 10.1145/3234804.3234823

Wu Wei, Shuai He, Dongliang Wang, Yao Yeboah

引用次数: 1

Application of Improved BP Neural Network in XAJ with Multiple Water Sources 改进BP神经网络在多水源XAJ中的应用

International Conference on Deep Learning Technologies Pub Date : 2018-06-27 DOI: 10.1145/3234804.3234814

Bai Juan, Yong Li, Yao Jun

引用次数: 0

A study of Green Natural Learning Method at the Improvement of Cognitive Function "Take the elderly with mild dementia as example" 绿色自然学习法在认知功能改善中的研究——以老年轻度痴呆患者为例

International Conference on Deep Learning Technologies Pub Date : 2017-06-02 DOI: 10.1145/3094243.3094253

Yu-Che Huang, Tai-Shen Huang

{"title":"A study of Green Natural Learning Method at the Improvement of Cognitive Function \"Take the elderly with mild dementia as example\"","authors":"Yu-Che Huang, Tai-Shen Huang","doi":"10.1145/3094243.3094253","DOIUrl":"https://doi.org/10.1145/3094243.3094253","url":null,"abstract":"Application of Artificial Intelligence in learning has always been an important issue.. However, collapse or lesion on functions of the cerebrum cell is a common phenomenon upon aging for the elderly. Senile Dementia is even more one of common problems. Thus, how to facilitate the activation of the cerebrum cell for deferring aging of the cerebrum through an excellent learning training is the issue of discussion that many scientists and educationalists have been endeavoring to explore for a long time. The study attempt to develop a curriculum of green natural learning(combine landscape therapy with learning therapy) in order to intensify the cerebrum function, alleviate brain collapse, promote mental function and reinforce learning effect by means of horticultural and learning methods on five sense stimulation, experience in person and the application of learning theory. Further reduce the traditional drug treatment have a negative impact (including falls, poisoning, etc.) According to the integrated result of the research, it is verified that the training of green natural learning can improve cognitive function and facilitate the promotion on cognition and learning competency of the elderly. The results of this study hope for the future of the elderly in the wisdom of learning to provide a reference basis and direction.","PeriodicalId":118446,"journal":{"name":"International Conference on Deep Learning Technologies","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131707966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predicting Vulnerable Software Components through Deep Neural Network 基于深度神经网络的易攻击软件组件预测

International Conference on Deep Learning Technologies Pub Date : 2017-06-02 DOI: 10.1145/3094243.3094245

Yulei Pang, Xiaozhen Xue, Huaying Wang

引用次数: 55

Real-Time Illegal Parking Detection System Based on Deep Learning 基于深度学习的违规停车实时检测系统

International Conference on Deep Learning Technologies Pub Date : 2017-06-02 DOI: 10.1145/3094243.3094261

Xuemei Xie, Chenye Wang, Shu Chen, Guangming Shi, Zhifu Zhao

引用次数: 36

The Construction and Practice of Electronic Technology Course Based on ICC 基于ICC的电子技术课程建设与实践

International Conference on Deep Learning Technologies Pub Date : 1900-01-01 DOI: 10.1145/3417188.3417209

N. Wang, Yan Zhang, Xiyou Chen, Xianmin Mu, Guanlin Li

引用次数: 0