A semi-supervised attention model for identifying authentic sneakers

Yang, Yang; Zhu, Nengjun; Wu, Yifeng; Cao, Jian; Zhan, De-Chuan; Xiong, Hui

doi:10.26599/bdma.2019.9020017

Cited by 6 publications

(3 citation statements)

References 28 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nowadays, deep learning has become a powerful technique for big data analysis [27], which can learn the feature representation and the classifier in an end-to-end way. The technique is widely used in many applications, such as flooding process prediction [28], visibility estimation [10], shape recognition [29], object detection [30], visual question answering [31], insect pest recognition [32], advertising click-through rate prediction [33], event extraction [34], sneaker recognition [35], modulation recognition [36], sentiment analysis [37], intrusion detection [38][39][40], climate prediction [41], internet of vehicles [42], healthcare [43], and face clustering [44]. Due to the advantage of deep learning, various researchers apply deep learning to predict credit scores.…”

Section: Credit Scoringmentioning

confidence: 99%

RankXGB-Based Enterprise Credit Scoring by Electricity Consumption in Edge Computing Environment

Shen¹,

Zhang²,

Song³

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

With the rapid development of the internet of things (IoT), electricity consumption data can be captured and recorded in the IoT cloud center. This provides a credible data source for enterprise credit scoring, which is one of the most vital elements during the financial decision-making process. Accordingly, this paper proposes to use deep learning to train an enterprise credit scoring model by inputting the electricity consumption data. Instead of predicting the credit rating, our method can generate an absolute credit score by a novel deep ranking model-ranking extreme gradient boosting net (rankXGB). To boost the performance, the rankXGB model combines several weak ranking models into a strong model. Due to the high computational cost and the vast amounts of data, we design an edge computing framework to reduce the latency of enterprise credit evaluation. Specially, we design a two-stage deep learning task architecture, including a cloud-based weak credit ranking and an edge-based credit score calculation. In the first stage, we send the electricity consumption data of the evaluated enterprise to the computing cloud server, where multiple weak-ranking networks are executed in parallel to produce multiple weak-ranking results. In the second stage, the edge device fuses multiple ranking results generated in the cloud server to produce a more reliable ranking result, which is used to calculate an absolute credit score by score normalization. The experiments demonstrate that our method can achieve accurate enterprise credit evaluation quickly.

show abstract

Section: Credit Scoringmentioning

confidence: 99%

RankXGB-Based Enterprise Credit Scoring by Electricity Consumption in Edge Computing Environment

Shen¹,

Zhang²,

Song³

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

show abstract

“…After that, a mechanism for authentication helped improve the healthcare record-keeping by enhancing reliability. A multiple-wireless sensor network-based authentication was established on the blockchain [ 18 , 19 ]. Several inner and outer blockchains were introduced in this research work that help integrate the local as well as public chain and form a hybrid blockchain design.…”

Section: Related Workmentioning

confidence: 99%

Federated Transfer Learning for Authentication and Privacy Preservation Using Novel Supportive Twin Delayed DDPG (S-TD3) Algorithm for IIoT

Arumugam

Maurya

et al. 2021

Sensors

View full text Add to dashboard Cite

The Industrial Internet of Things (IIoT) has led to the growth and expansion of various new opportunities in the new Industrial Transformation. There have been notable challenges regarding the security of data and challenges related to privacy when collecting real-time and automatic data while observing applications in the industry. This paper proposes an Federated Transfer Learning for Authentication and Privacy Preservation Using Novel Supportive Twin Delayed DDPG (S-TD3) Algorithm for IIoT. In FT-Block (Federated transfer learning blockchain), several blockchains are applied to preserve privacy and security for all types of industrial applications. Additionally, by introducing the authentication mechanism based on transfer learning, blockchains can enhance the preservation and security standards for industrial applications. Specifically, Novel Supportive Twin Delayed DDPG trains the user model to authenticate specific regions. As it is considered one of the most open and scalable interacting platforms of information, it successfully helps in the positive transfer of different kinds of data between devices in more significant and local operations of the industry. It is mainly due to a single authentication factor, and the poor adaptation to regular increases in the number of users and different requirements that make the current authentication mechanism suffer a lot in IIoT. As a result, it has been very clearly observed that the given solutions are very useful.

show abstract

“…It prevented network degradation and improved the performance of deep networks. Yang et al proposed the semisupervised attention (SSA) model [20], which has a semi-supervised attention structure for different small target images. Using unlabeled data in the data can help reduce the change of the same category and achieve more distinguishing feature extraction.…”

Section: Introductionmentioning

confidence: 99%

RT-YOLO: A Residual Feature Fusion Triple Attention Network for Aerial Image Target Detection

Zhang¹,

Deng²,

Chen³

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

In recent years, target detection of aerial images of unmanned aerial vehicle (UAV) has become one of the hottest topics. However, target detection of UAV aerial images often presents false detection and missed detection. We proposed a modified you only look once (YOLO) model to improve the problems arising in object detection in UAV aerial images: (1) A new residual structure is designed to improve the ability to extract features by enhancing the fusion of the inner features of the single layer. At the same time, triplet attention module is added to strengthen the connection between space and channel and better retain important feature information.(2) The feature information is enriched by improving the multi-scale feature pyramid structure and strengthening the feature fusion at different scales. (3) A new loss function is created and the diagonal penalty term of the anchor frame is introduced to improve the speed of training and the accuracy of reasoning. The proposed model is called residual feature fusion triple attention YOLO (RT-YOLO). Experiments showed that the mean average precision (mAP) of RT-YOLO is increased from 57.2% to 60.8% on the vehicle detection in aerial image (VEDAI) dataset, and the mAP is also increased by 1.7% on the remote sensing object detection (RSOD) dataset. The results show that the RT-YOLO outperforms other mainstream models in UAV aerial image object detection.

show abstract

A semi-supervised attention model for identifying authentic sneakers

Cited by 6 publications

References 28 publications

RankXGB-Based Enterprise Credit Scoring by Electricity Consumption in Edge Computing Environment

RankXGB-Based Enterprise Credit Scoring by Electricity Consumption in Edge Computing Environment

Federated Transfer Learning for Authentication and Privacy Preservation Using Novel Supportive Twin Delayed DDPG (S-TD3) Algorithm for IIoT

RT-YOLO: A Residual Feature Fusion Triple Attention Network for Aerial Image Target Detection

Contact Info

Product

Resources

About