Potato Quality Grading Based on Depth Imaging and Convolutional Neural Network by Qinghua Su, Naoshi Kondo, Dimas Firmanda Al Riza, Harshana Habaragamuwa

“…The machine learning system, which is composed of a softmax regression model and a convolutional neural network model, can grade a potato tube into six different quality levels based on tube appearance and size. …”
Get full text

Studying Forgetting in Faster R-CNN for Online Object Detection: Analysis Scenarios, Localization in the Architecture, and Mitigation by Baptiste Wagner, Denis Pellerin, Sylvain Huet

“…CR also masks the logits of old objects in the softmax classification layer to mitigate forgetting. …”
Get full text

FSErasing: Improving Face Recognition with Data Augmentation Using Face Parsing by Hiroya Kawai, Koichi Ito, Hwann-Tzong Chen, Takafumi Aoki

“…ResNet-50 with Softmax using CASIA-WebFace improves the average accuracy by 0.442 points and the average EER by 0.452 points, and with ArcFace, the average accuracy and EER improve by 0.228 points and 0.500 points, respectively. …”
Get full text

Person Re-Identification Using Additive Distance Constraint With Similar Labels Loss by Guofa Li, Lisha Huang, Liangwen Tang, Chunli Han, Yaoyu Chen, Heng Xie, Shen Li, Gang Xu

“…Most of the current studies used the traditional Softmax loss for solutions, but its discriminative capability encounters a bottleneck. …”
Get full text

Hvordan sikre mer pensjon til sliterne? by Axel West Pedersen

“…Kommentaren støtter denne problembeskrivelsen, men reiser spørsmål ved de forslag til løsninger som artikkelforfatterne skisserer i artikkelens siste del.…”
Get full text

Intelligent Auxiliary Artificial Wood Plank Pattern Design Based on the Subject Search Algorithm of Multimedia Resources by Lihua Zhao

“…At the same time, use the self-learning method to optimize the convergence efficiency and reduce the design time. Finally, pass the softmax designer extracts design schemes for patterns and straight lines.…”
Get full text

An approach of Bagging ensemble based on feature set and application for traffic classification by Yaguan QIAN, Xiaohui GUAN, Shuhui WU, Bensheng YUN, Dongxiao REN

“…Bagging is a classic ensemble approach,whose effectiveness depends on the diversity of component base classifiers.In order to gain the largest diversity,employing genetic algorithms to get independent feature subset for each base classifier was proposed.Meanwhile,for better generalization,the optimal weights for the base classifiers according to their predictive performance were selected.Finally,refined Bagging ensemble based on simple Softmax regression was applied successfully in traffic classification.The experiment result shows that the proposed approach can get more improvement than the original Bagging ensemble in classification performance,and is better than the random-forests to a certain extent.…”
Get full text

Light weighted privacy protection ViT inference framework based on secret sharing by MA Min, FU Yu, HUANG Kai, JIA Xiaofeng

“…Two edge servers collaborated to execute secure computing protocols based on secret sharing design, such as SSoftmax, SLayerNorm, SGeLU, etc. While maintaining the original framework structure of ViT-B/16, the problem of large inference overhead in privacy protection framework was solved. …”
Get full text

Le biopic du sportif américain by Rémi Fontanel

“…Most of these biographical pictures are characterized by the figure of the sportsman represented as a committed figure, driven by a sense of revolt and fighting. …”
Get full text

Deep belief network-based link quality prediction for wireless sensor network by Lin-lan LIU, Jiang-bo XU, Yue LI, Zhi-yong YANG

“…After analyzing the existing link quality prediction models,a link quality prediction model for wireless sensor network was proposed,which was based on deep belief network.Support vector classification was employed to estimate link quality,so as to get link quality levels.Deep belief network was applied in extracting the features of link quality,and softmax was taken to predict the next time link quality.In different scenarios,compared with the model of link quality prediction based on logistic regression,BP neural network and Bayesian network methods,the experimental results show that the proposed prediction model achieves better precision.…”
Get full text

End-to-end audiovisual speech recognition based on attention fusion of SDBN and BLSTM by Yiming WANG, Ken CHEN, Aihaiti ABUDUSALAMU

“…An end-to-end audiovisual speech recognition algorithm was proposed.In algorithm,a sparse DBN was constructed by introducing mixed l<sub>1/2</sub>norm and l<sub>1</sub>norm into Deep Belief Network with bottleneck structure to extract the sparse bottleneck features,so as to reduce the dimension of data features,and then a BLSTM was used to model the feature in time series.Then,a attention mechanism was used to align and fuse the lip visual information and audio auditory information automatically.Finally,the fused audiovisual information was classified and identified by a BLSTM with a Softmax layer attached.Experiments show that the algorithm can effectively identify visual and auditory information,and has good recognition rate and robustness in similar algorithms.…”
Get full text

ReluformerN: Lightweight High-Low Frequency Enhanced for Hyperspectral Agricultural Lancover Classification by LIU Yi, ZHANG Yanjun

“…Reluformer replaced the softmax function with a function of quadratic computational complexity, and through theoretical and graphical analysised, Relu function, LeakRelu function, and Gelu function were compared, it was found that the ReLU function and the softmax function both had non-negativity, which could be used for feature relevance analysis. …”
Get full text

Deep learning Chinese input method with incremental vocabulary selection by Huajian REN, Xiulan HAO, Wenjing XU

“…The core task of an input method is to convert the keystroke sequences typed by users into Chinese character sequences.Input methods applying deep learning methods have advantages in learning long-range dependencies and solving data sparsity problems.However, the existing methods still have two shortcomings: the separation structure of pinyin slicing in conversion leads to error propagation, and the model is complicated to meet the demand for real-time performance of the input method.A deep-learning input method model incorporating incremental word selection methods was proposed to address these shortcomings.Various softmax optimization methods were compared.Experiments on People’s Daily data and Chinese Wikipedia data show that the model improves the conversion accuracy by 15% compared with the current state-of-the-art model, and the incremental vocabulary selection method makes the model 130 times faster without losing conversion accuracy.…”
Get full text

CNID: Research of Network Intrusion Detection Based on Convolutional Neural Network by Guojie Liu, Jianbiao Zhang

“…First, the data is preprocessed, the original one-dimensional network intrusion data is converted into two-dimensional data, and then the effective features are learned using optimized convolutional neural networks, and, finally, the final test results are produced in conjunction with the Softmax classifier. In this paper, KDD-CUP 99 and NSL-KDD standard network intrusion detection dataset were used to carry out the multiclassification network intrusion detection experiment; the experimental results show that the multiclassification network intrusion detection model proposed in this paper improves the accuracy and check rate, reduces the false positive rate, and also obtains better test results for the detection of unknown attacks.…”
Get full text

Neural Linguistic Steganalysis via Multi-Head Self-Attention by Sai-Mei Jiao, Hai-feng Wang, Kun Zhang, Ya-qi Hu

“…Then, we utilize multi-head self-attention to model the interactions between words in carrier. Finally, a softmax layer is utilized to categorize the input text as cover or stego. …”
Get full text

New method of text representation model based on neural network by Shui-fei ZENG, Xiao-yan ZHANG, Xiao-feng DU, Tian-bo LU

“…Method of text representation model was proposed to extract word-embedding from text feature.Firstly,the word-embedding of the dual word-embedding list based on dictionary index and the corresponding part of speech index was created.Then,feature vectors was obtained further from these extracted word-embeddings by using Bi-LSTM recurrent neural network.Finally,the sentence vectors were processed by mean-pooling layer and text categorization was classified by softmax layer.The training effects and extraction performance of the combination model of Bi-LSTM and double word-embedding neural network were verified.The experimental results show that this model not only performs well in dealing with the high-quality text feature vector and the expression sequence,but also significantly outperforms other three kinds of neural networks,which includes LSTM,LSTM+context window and Bi-LSTM.…”
Get full text

FAULT DIAGNOSIS OF WIND TURBINE BEARING BASED ON SENET-RESNEXT-LSTM by DU HaoFei, ZHANG Chao, LI JianJun

“…Finally, the two extracted features are fused and input to the Softmax layer for fault classification. The experimental results show that, compared with the current bearing fault diagnosis method based on deep learning, the proposed method performs better in bearing fault classification accuracy.…”
Get full text

A lightweight privacy-preserving truth discovery mechanism for IoT by Yingjie HE, Qiwei LI, Han SUN, Di GAO, Jianfeng DONG, Shuhua YANG

“…In order to solve diverse quality and privacy leakage of perceived data in fog-cloud integrated internet of things (IoT), a streaming encryption-based privacy-preserving truth discovery mechanism for IoT was proposed.Firstly, by utilizing shuffling and streaming encryption algorithms, the ground truths and the weights were anonymously updated on the cloud server and fog server, respectively, so that the collusion attacks between malicious attackers and cloud or fog servers could be resisted to defend against privacy leakage of IoT devices.Secondly, by adopting the Softmax function, the device weights were calculated on fog server, which reduces the error rate for calculating the ground truths.Finally, the theoretical analysis proved that the mechanism could protect privacy of the devices.And, the experimental results demonstrate that the proposed mechanism is an effective privacy-preserving trust discovery mechanism for large-scale IoT devices, which can outperform existing ones in computing efficiency.…”
Get full text

Identités professionnelles des formateurs d’enseignants d’EPS en UFR STAPS lors de l’année de préparation au CAPEPS by Youcef Alanbagi, Ghislain Carlier, Jacques Mikulovic

“…The study shows three antagonists poles of these trainers (Scientists, Didacticians, Sportmen) and Sport science teachers’ trainers difficulties to represent their own position within the University. …”
Get full text

OSVRT NA DESETLJETNO SUSTAVNO ETNOLOŠKO ISTRAŽIVANJE PRIMORSKO-LIČKIH BUNJEVACA by MILANA ČERNELIĆ

“…Ante Glavičića 1999. godine, nastavila se putem projekta Ministarstava znanosti obrazovanja i sporta Republike Hrvatske Identitet i etnogeneza primorskih Bunjevaca (2002.-1006) u suradnji s Gradskim muzejom Senja. …”
Get full text