Fault injection attacks on SoftMax function in deep neural networks

Jap, Dirmanto; Won, Yoo-Seung; Bhasin, Shivam

doi:10.1145/3457388.3458870

Cited by 7 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As described in Figure 9, the dot product AM first calculates the point multiplication of query and all keys, divides each by the square root (SQR) dk to prevent overlarge product, and then feeds it to the Softmax function to obtain the corresponding weight (Jap et al. , 2021).…”

Section: Methodsmentioning

confidence: 99%

Analyzing legal education mobile learner's behavior using deep learning under social media

Chen

2022

LHT

View full text Add to dashboard Cite

PurposeUnder emerging social media technology, mobile learners' behavior analysis and legality education have important practical significance. The research aims to detect the mobile learning (M-learning) learners' behavior in legality education under the background of the Internet era and improve the learning and teaching effect of online legality education and law popularization.Design/methodology/approachThis paper proposes a model based on deep learning (DL) fuzzy clustering analysis (FCA), and bidirectional encoder and decoder (ENDEC) of converter model to detect the mobile learners' behaviors in online legality education under the current social media. Then, the effectiveness of the proposed model is tested. The proposed model expects to be applied to multimedia teaching and law popularization activities and provides some theoretical reference and practical value for improving the effectiveness of online teaching.FindingsThe experimental results show that in the learner behavior detection process of M-learning-oriented online legality education, the model's accuracy can reach 99.8%. The response time is shorter than other algorithms. Overall, the application effect of the proposed model and algorithm is good and can be applied in practice.Research limitations/implicationsThe research results may lack universality due to the selected research methods. Therefore, researchers are encouraged to test the proposed methods further. In the future, it is necessary to expand the type and scale of text data to improve the accuracy of data detection.Practical implicationsThe research results provide a specific theoretical reference and practical significance for improving the learning effect of online M-learning-oriented legality education.Originality/valueThis paper meets the needs of mobile learner behavior analysis based on social media.

show abstract

Section: Methodsmentioning

confidence: 99%

Analyzing legal education mobile learner's behavior using deep learning under social media

Chen

2022

LHT

View full text Add to dashboard Cite

show abstract

“…The normalization process is highly susceptible to larger or smaller inputs, and it is easy to map to 0 and 1. After the normalization transformation, when the full probability is assigned to the label corresponding to the maximum value, the model is trained with factor k for scaling to prevent the gradient from disappearing during backpropagation [41].…”

Section: Calculate the Distribution Of Attentionmentioning

confidence: 99%

In-depth Recommendation Model Based on Self-Attention Factorization

2023

KSII TIIS

View full text Add to dashboard Cite

Rating prediction is an important issue in recommender systems, and its accuracy affects the experience of the user and the revenue of the company. Traditional recommender systems use Factorization Machines for rating predictions and each feature is selected with the same weight. Thus, there are problems with inaccurate ratings and limited data representation. This study proposes a deep recommendation model based on self-attention Factorization (SAFMR) to solve these problems. This model uses Convolutional Neural Networks to extract features from user and item reviews. The obtained features are fed into self-attention mechanism Factorization Machines, where the self-attention network automatically learns the dependencies of the features and distinguishes the weights of the different features, thereby reducing the prediction error. The model was experimentally evaluated using six classes of dataset. We compared MSE, NDCG and time for several real datasets. The experiment demonstrated that the SAFMR model achieved excellent rating prediction results and recommendation correlations, thereby verifying the effectiveness of the model.

show abstract

“…SoftMax function [10][11][12] has a wide range of applications in machine learning and deep learning, mainly for multiclassification problems. Most of the time, in the multi-classification problem, we want to output the probability of getting a certain category.…”

Section: Anti-domain Adaptation Of Knowledge Mappingmentioning

confidence: 99%

Research on fault diagnosis model driven by artificial intelligence from domain adaptation to domain generalization

Xu,

Cao

2023

Fourth International Conference on Artificial Intelligence and Electromechanical Automation (AIEA 2023)

View full text Add to dashboard Cite

Transfer learning is studied through inductive research, knowledge transfer, lifelong learning and cumulative learning. The research scope and field of transfer learning are very wide. Transfer learning is a training method that can transfer the network structure and weight originally used to solve the mature task methodology to the new learning task, and can also get better results in the new task. The reason why transfer learning can be realized is that the characteristics of convolutional neural network in shallow learning are universal. In the case of insufficient samples, transfer learning can be used to transfer these general feature learning from other trained networks, so as to save training time and obtain better recognition results.

show abstract

Fault injection attacks on SoftMax function in deep neural networks

Cited by 7 publications

References 4 publications

Analyzing legal education mobile learner's behavior using deep learning under social media

Analyzing legal education mobile learner's behavior using deep learning under social media

In-depth Recommendation Model Based on Self-Attention Factorization

Research on fault diagnosis model driven by artificial intelligence from domain adaptation to domain generalization

Contact Info

Product

Resources

About