Citation

Tan, Jun Wei and Goh, Pey Yun and Tan, Shing Chiang and Chong, Lee Ying (2023) Visual Analysis on Adversarial Images. In: 2023 2nd International Conference on Computer Technologies (ICCTech), 23-25 February 2023, Kuantan, Malaysia.

Text 35.pdf - Published Version Restricted to Repository staff only Download (854kB)

Abstract

Adversarial Machine Learning (ML) is one of the biggest cyber threats to deep neural networks (DNN). Adversarial samples are crafted or created by an attacker to mislead a DNN model in making a decision. On the presence of adversarial samples in a dataset, an interest is to trace their patterns before designing an effective defense algorithm. In this paper, we propose a hybrid method, i.e. Convo t-SNE where convolution neural network (CNN) is used to extract the features of images and t-distributed Stochastic Neighborhood Embedding (t-SNE) is applied to reduce dimensions so that adversarial attack patterns can be visualized. In our work, three attack methods are applied on three public datasets (i.e., MNIST, CIFAR10 and CIFAR100). The graphical results indicate clearly the presence of attack patterns using Convo tSNE. The proposed method is then hybrid with k-mean to further evaluate its effectiveness. Two evaluation metrics i.e. classification accuracy and a clustering normalized mutual information (ClustNMI, a metric to compare between two clustering algorithms, the higher the better) are used. The results are encouraging where Convo t-SNE improves the clustering algorithms with a higher classification accuracy and better ClustNMI than t-SNE.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Convolutional Neural Network, deep learning
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	22 Feb 2024 07:06
Last Modified:	22 Feb 2024 07:06
URII:	http://shdl.mmu.edu.my/id/eprint/12114

Downloads

Edit (login required)