发布求助
文献互助 智能选刊 最新文献

2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)最新文献

筛选
英文 中文
Fast QBE: Towards Real-Time Spoken Term Detection with Separable Model 快速QBE:用可分离模型实现实时口语词检测
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00035
Ziwei Tian, Shiqing Yang, Minqiang Xu
{"title":"Fast QBE: Towards Real-Time Spoken Term Detection with Separable Model","authors":"Ziwei Tian, Shiqing Yang, Minqiang Xu","doi":"10.1109/MLISE57402.2022.00035","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00035","url":null,"abstract":"State-of-the-art spoken term detection or query-by-example networks depend on recurrent neural network (RNN), which extract fixed-dimensional vectors (embedded vectors) from both spoken query and the search content, and then calculate cosine distances over the vectors. However, these methods depend on time sequence, so it is a computational cost task, can not meet the requirements of both the query accuracy and search speed. In this work, we introduce a fast Spoken term detection system based on a separable model—RepVGG. Because of the trick of reparameterization, it has a faster speed in inference. Secondly, we use non maximum suppression and norm in the step of inference to improve it performance. Thirdly, we use multilanguage training to improve both accuracy and robustness of the system. Corresponding experiments are designed to verify these ideas. It show that proposed methods can import the GPU real-time factor (RTF) from 150 to 2300, and outperforms the state-of-the art method.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132462979","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
Design of a YOLO Model Accelerator Based on PYNQ Architecture 基于PYNQ体系结构的YOLO模型加速器设计
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00011
Lin Wang, Tianyong Ao, Le Fu, Jian Liu, Yang Liu, Yingjie Zhou
{"title":"Design of a YOLO Model Accelerator Based on PYNQ Architecture","authors":"Lin Wang, Tianyong Ao, Le Fu, Jian Liu, Yang Liu, Yingjie Zhou","doi":"10.1109/MLISE57402.2022.00011","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00011","url":null,"abstract":"The application requirements of object detection models based on deep learning are very extensive. However, high computing power requirements often seriously restrict the application of these models on resource-constrained devices with high energy efficiency requirements. To address this problem, a YOLO model accelerator architecture is proposed based on PYNQ. Based on the FPGA hardware platform, the hardware accelerator is designed by making full use of pipeline, loop unrolling, data reordering and other methods to accelerate the computationally intensive units in the YOLOv2 model such as the convolution and pooling layers. In order to reduce the delay in the data transmission process, the multi-channel transmission architecture combined with the ping-pong buffer is designed, and block-by-block reading strategy is adopted to read the off-chip data. The proposed YOLO model accelerator has been implemented and verified on Xilinx PYNQ-z2 platform. The experimental results show that the system has high detection accuracy and far lower power consumption than CPU and GPU. It can also be deployed on mobile devices to detect the surrounding environment.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130989721","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
Image Segmentation Algorithm Based on Attention Mechanism and Jump Connection 基于注意机制和跳跃连接的图像分割算法
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00058
Zhaotong Cui, Yanjun Wei, Tianping Li, Guanxing Li
{"title":"Image Segmentation Algorithm Based on Attention Mechanism and Jump Connection","authors":"Zhaotong Cui, Yanjun Wei, Tianping Li, Guanxing Li","doi":"10.1109/MLISE57402.2022.00058","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00058","url":null,"abstract":"With the development of deep learning, convolutional neural networks have become the mainstream of computer vision algorithms. In recent years, the biggest problem of applying convolutional neural networks to image segmentation is that they cannot achieve accurate segmentation of the last layer, also cause resolution loss when extracting features, and cannot meet the demand of different pixels requiring different context dependencies. To address these issues, we add an attention mechanism and a jump feature fusion method to deeplabv3+ so that features are extracted without severe feature loss and a broader range of contextual information can be encoded into local features. The feature map is further enriched by adding a module combining bilinear upsampling and deconvolution in the process of feature restoration. Compared to previous algorithms, the results of this algorithm are superior. A performance of 85.73% is achieved on PASCAL VOC2012 using the proposed model.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116094042","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
An Improved Anchor-Free Object Detection Method 一种改进的无锚目标检测方法
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00009
YuHu Han, Tonghe Ding, Tianping Li, Meng Li
{"title":"An Improved Anchor-Free Object Detection Method","authors":"YuHu Han, Tonghe Ding, Tianping Li, Meng Li","doi":"10.1109/MLISE57402.2022.00009","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00009","url":null,"abstract":"Object detection plays an important role in various industries. However, a fully convolutional one-stage detector (FCOS) has high computational cost and low detection accuracy. Therefore, in this paper, we fused the attention module (CBAM) with feature pyramid (FPN) to extract important feature information, suppress useless information and improve detection accuracy. Finally, we use inverted residual convolution block to replace the detection head of the original method, and the improved detection head reduces the calculation cost and amount of calculation. We use PASCAL VOC to train and evaluate our network. Experimental results show that, compared with the traditional method, the detection accuracy is improved by 1.2%, and the number of parameters is reduced by 4.2m.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123862282","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
A Domain Adaptive Adversarial Training Method Based on Self-Supervised Learning 基于自监督学习的领域自适应对抗训练方法
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00070
Chuqing Sun
{"title":"A Domain Adaptive Adversarial Training Method Based on Self-Supervised Learning","authors":"Chuqing Sun","doi":"10.1109/MLISE57402.2022.00070","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00070","url":null,"abstract":"Image classification technology based on neural network is an important task in computer vision, and the introduction of transfer learning can solve the problems of lack of data sets and long training time. To address this problem, this paper proposes a self-supervised domain-adaptive adversarial network approach. The algorithm uses the VGG network to extract image features, realizes the transfer learning of different image styles through domain adversarial training, and introduces a data augmentation model and self-supervised learning method based on pseudo-label to improve the accuracy of model classification. The experimental results show that the model can effectively improve the accuracy of image transfer learning of different styles in the image classification problem. When the number of pseudo-labels is 10, the classification effect is the best, and the accuracy rate is improved to 12.99%, which greatly saves training time and computing power while solving the problem of missing training data.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131387461","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
Parameter optimization algorithm for quantum particle swarm-based i-vector identification systems 基于量子粒子群的i向量识别系统参数优化算法
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00065
Guangqi Liu, Wushour Silamu
{"title":"Parameter optimization algorithm for quantum particle swarm-based i-vector identification systems","authors":"Guangqi Liu, Wushour Silamu","doi":"10.1109/MLISE57402.2022.00065","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00065","url":null,"abstract":"For the noise robustness problem in i-vector: Based on the theoretical principle of i-vector speaker recognition system, the extraction principle and scoring calculation method of i-vector and the process of channel compensation algorithm based on PLDA (Probabilistic Linear Discriminant Analysis) with PLDA model are studied. The matching principle is studied. A statistical averaging i-vector extraction algorithm based on speech fragmentation is proposed to extract more robust i-vector features by weakening the statistical parameters of bad speech fragments to improve the recognition performance of the system. After that, the i-vector system is designed to improve the recognition performance of the i-vector.l Then, a Quantum Particle Swarm Optimization is designed to optimize the parameters of the i-vector recognition system to avoid the degradation of the system performance caused by artificial empirical values. Experimental analysis shows that the proposed algorithm has improved performance over the traditional i-vector recognition algorithm, especially in the case of noise interference, and has better recognition performance","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127787471","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
A Peach Maturity Detection Algorithm Based on YOLOV5 Attention Mechanism 基于YOLOV5注意机制的桃子成熟度检测算法
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00063
Jiong Zhang, Caizeng Ye
{"title":"A Peach Maturity Detection Algorithm Based on YOLOV5 Attention Mechanism","authors":"Jiong Zhang, Caizeng Ye","doi":"10.1109/MLISE57402.2022.00063","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00063","url":null,"abstract":"With the rapid development of computer vision technology, the original method of rapid non-destructive testing of peach maturity has been unable to meet the management requirements, and the management efficiency and accuracy cannot meet the daily needs. Based on the above reasons, this paper proposes a peach maturity detection algorithm based on YOLOV5 attention mechanism. The algorithm shows the whole process of management through data collection, model training, and model application. It adopts the prototype method and the object-oriented system development method. The external camera network configuration is used to connect the image display interface of the web display screen, and the server is used as the support and the bearer network is used as the system basis. Through the experimental results, it is found that the new algorithm proposed in this paper has obvious advantages compared with other comparison algorithms, and the experimental results more precise.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124475928","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
Multiple solutions for a Class of p-Kirchhoff type equations 一类p-Kirchhoff型方程的多重解
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/mlise57402.2022.00024
Beibei Wang
{"title":"Multiple solutions for a Class of p-Kirchhoff type equations","authors":"Beibei Wang","doi":"10.1109/mlise57402.2022.00024","DOIUrl":"https://doi.org/10.1109/mlise57402.2022.00024","url":null,"abstract":"In this paper, An improved Mountain Pass Theorem is used to study the existence of multiple solutions for the nonlinear p-kirchhoff Dirichlet boundary value problems under some natural conditions on f(x, v), and some known results are generalized.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115831860","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: 1
Stock Market Predictability Using Machine Learning Techniques 利用机器学习技术预测股票市场
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00075
Jiuye Wu
{"title":"Stock Market Predictability Using Machine Learning Techniques","authors":"Jiuye Wu","doi":"10.1109/MLISE57402.2022.00075","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00075","url":null,"abstract":"The purpose of this article is to examine stock market analysis using machine learning techniques unique and attractive way to reassess products according to the changing times, and to publish information to create textual information from one or more sources in order to identify different threats. This is accomplished by controlling the reader to estimate the value of the stock by determining the need to view chaotic data and estimate the impact of Microsoft. The product uses the Naive Bayes classifier. First, the newsletters and archives have been placed in a free file in the “Date” column after making some significant changes. It also has a structure similar to the data distribution that protects the nearest-neighbor (Well-NN) experiment. Lately, natural language processing is used in the Journal of Captions and has a guide with 20,000 commonly used words and how to convert them to vectorized text. Subsequently, Naive Bayes models were also developed using dropout datasets, 80 based on training data and 20 based on control data. Overall, the neural-based model was found to be slightly better and the macro means for F1 than the standard and configured model, and slightly better. The results of F1-score validate effectiveness of the proposed method, which achieves a satisfying performance.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124779889","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
Unteady aerodynamics modeling method based on dendrite-based gated recurrent neural network model 基于树突门控递归神经网络模型的非稳态空气动力学建模方法
2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) Pub Date : 2022-08-01 DOI: 10.1109/MLISE57402.2022.00093
Ke Liu, Jun Huang, Zhiqin Liu, Qingfeng Wang
{"title":"Unteady aerodynamics modeling method based on dendrite-based gated recurrent neural network model","authors":"Ke Liu, Jun Huang, Zhiqin Liu, Qingfeng Wang","doi":"10.1109/MLISE57402.2022.00093","DOIUrl":"https://doi.org/10.1109/MLISE57402.2022.00093","url":null,"abstract":"System identification-based aerodynamic reduced-order model (ROM) is an effective method for solving nonlinear unsteady aerodynamics prediction. However, most of the ROMs are shallow neural network architecture which is difficult to capture nonlinear aerodynamic behavior with large samples and Varying Mach Numbers. This paper proposes a dendrite-based gated recurrent neural network (DD-GRU) fusion model from deep learning theory to improve the capability of system identification-based ROM in varying flow conditions. DD-GRU network can process time-series data, which is also suitable for capturing the time-delay effect of unsteady aerodynamics. Unlike other recurrent models, DD-GRU uses the dendrite method to re-extract the logic of information of GRU. The approach is evaluated by predicting NACA64A01 airfoil transonic aerodynamic loads across multiple flow conditions. The example demonstrates that the model accurately predicts the harmonic aerodynamic response in the frequency and time domains with different flow conditions.","PeriodicalId":350291,"journal":{"name":"2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134216768","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: 1
首页 上一页 下一页 尾页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信