18,555 Hits in 3.1 sec

A Novel Technique for Image Steganalysis Based on Separable Convolution and Adversarial Mechanism

Yuwei Ge, Tao Zhang, Haihua Liang, Qingfeng Jiang, Dan Wang
2021 Electronics  
Although convolutional neural networks (CNNs) can effectively extract the features describing the image content, the difficulty lies in extracting the subtle features that describe the existence of hidden  ...  Considering this concern, this paper introduces separable convolution and adversarial mechanism, and proposes a new network structure that effectively solves the problem.  ...  The separable convolution module and the adversarial mechanism work together to ensure that the network can fully extract the steganographic embedding features and complete the steganalysis task, without  ... 
doi:10.3390/electronics10222742 fatcat:o7exnpjzrbbozizqp7in2durtu

Multi-Channel Convolutional Neural Networks with Adversarial Training for Few-Shot Relation Classification (Student Abstract)

Yuxiang Xie, Hua Xu, Congcong Yang, Kai Gao
In this paper, we use word embedding and position embedding to construct multi-channel vector representation and use the multi-channel convolutional method to extract features of sentences.  ...  Moreover, in order to alleviate few-shot learning to be sensitive to overfitting, we introduce adversarial learning for training a robust model.  ...  All rights reserved. word embedding and position embedding into a multichannel form similar to image representation and use multichannel convolution neural networks for feature extraction.  ... 
doi:10.1609/aaai.v34i10.7256 fatcat:tusk5cjd45hq5pam2akz354gpm

AGCN: Adversarial Graph Convolutional Network for 3D Point Cloud Segmentation

Seunghoi Kim, Daniel Alexander
2021 British Machine Vision Conference  
In this paper, we propose an Adversarial Graph Convolutional Network for 3D point cloud segmentation.  ...  By using an embedding L 2 loss as an adversarial loss, the proposed network is learned to reduce noisy labels by enforcing the consistency between neighbouring labels.  ...  The proposed Adversarial Graph Convolution Network (AGCN) trains two networks, a segmentation network and a discriminator network, in an adversarial manner where the discriminator network calculates a  ... 
dblp:conf/bmvc/KimA21 fatcat:oegq6iglnnchbhoffgnpfk2vva

Security Enhanced Sentence Similarity Computing Model Based on Convolutional Neural Network

Qifeng Sun, Xingzhe Huang, Godfrey Kibalya, Neeraj Kumar, SVN Santhosh Kumar, Peiying Zhang, Dongliang Xie
2021 IEEE Access  
Convolution neural network is able to fully extract the features of high-dimensional word vectors through multi-layer convolution and pooling.  ...  In the extraction of sentence mutual information, we mainly consider word2vec word vector embedding and co-occurrence word position information embedding. 2) Word vector embedding After the word segmentation  ... 
doi:10.1109/access.2021.3099489 fatcat:fpm2kaezazh33fvxvlrrykurcy

An End-to-End Text-Independent Speaker Verification Framework with a Keyword Adversarial Network

Sungrack Yun, Janghoon Cho, Jungyun Eum, Wonil Chang, Kyuwoong Hwang
2019 Interspeech 2019  
Also, with the adversarial gradient of the ASR network, the text-dependency of the speaker embedding vector can be reduced.  ...  In training our speaker verification framework, we consider both the triplet loss minimization and adversarial gradient of the ASR network to obtain more discriminative and text-independent speaker embedding  ...  In training the SE network, we use the adversarial gradient from the ASR to encourage the SE network to extract the embedding vector which is phonetically-independent and contains only speaker characteristics  ... 
doi:10.21437/interspeech.2019-2208 dblp:conf/interspeech/YunCECH19 fatcat:56ejc4sjcncfjbzxkf3p6lhoxi

Sentiment Analysis Method based on Piecewise Convolutional Neural Network and Generative Adversarial Network

Changshun Du, Lei Huang
2019 International Journal of Computers Communications & Control  
However, the original convolution neural network model ignores sentence structure information which is very important for text sentiment analysis.  ...  In order to alleviate the sparsity of the data, this paper also uses the generative adversarial network to make common feature extractions, so that the model can obtain the common features associated with  ...  common sentiment feature extraction with generative adversarial network Inspired by GAN, we propose a adversarial-sharing-private model for multi-domain common sentiment feature extraction.  ... 
doi:10.15837/ijccc.2019.1.3374 fatcat:jsjkwh2ehjhidbik3ajgjppb64

A Survey on Deep Convolutional Neural Networks for Image Steganography and Steganalysis

2020 KSII Transactions on Internet and Information Systems  
Steganalysis & steganography have witnessed immense progress over the past few years by the advancement of deep convolutional neural networks (DCNN).  ...  The result of this study opens new approaches for upcoming research and may serve as source of hypothesis for further significant research on deep learning-based image steganography and steganalysis.  ...  Their network composed of two networks, hiding network called U-Net and an extraction network. Extraction network subsist of six convolutional layers with filter size of 3×3.  ... 
doi:10.3837/tiis.2020.03.017 fatcat:7ci7bfbjsfd2nn5yagnv2h3ora

On the combination of data augmentation method and gated convolution model for building effective and robust intrusion detection

Yixiang Wang, Shaohua lv, Jiqiang Liu, Xiaolin Chang, Jinqiang Wang
2020 Cybersecurity  
Following this work, we propose a gated convolutional neural network (GCNN) model to thoroughly extract the potential information of augmented sequences.  ...  Adversarial examples used in adversarial training are generated by the proposed adversarial sequence generation algorithm.  ...  For convolutional neural networks and recurrent neural networks, it is necessary to use word embedding to convert system calls into word vector for calculation.  ... 
doi:10.1186/s42400-020-00063-5 fatcat:foakfaojyvfb5ipsg6cuj2hm64

An Efficient Method for Generating Adversarial Malware Samples

Yuxin Ding, Miaomiao Shao, Cai Nie, Kunyang Fu
2022 Electronics  
To address this issue, we propose a novel method to generate adversarial malware samples. Different from gradient-based methods, we extract feature byte sequences from benign samples.  ...  These methods generate adversarial samples case-by-case, which is very time-consuming to generate a large number of adversarial samples.  ...  So, they can be more widely used to generated adversarial samples for different neural networks, such as BP network, CNN [16] , and RNN [20] .  ... 
doi:10.3390/electronics11010154 fatcat:uvbjys6v2zcgflmcfgdzxp6die

Joint Character-Level Convolutional and Generative Adversarial Networks for Text Classification

Tianshi Wang, Li Liu, Huaxiang Zhang, Long Zhang, Xiuxiu Chen
2020 Complexity  
In the framework, we first quantify the texts by a character-level convolutional neural network and input the textual features into an adversarial network and a classifier, respectively.  ...  Our framework introduces generative networks to enrich corpus and utilizes a character-level convolutional network to extract latent semantic.  ...  Self-Attentive [35] : a model for extracting an interpretable sentence embedding by introducing self-attention. e method uses a 2D matrix to represent the embedding and proposes a self-attention mechanism  ... 
doi:10.1155/2020/8516216 fatcat:jpliqzndizblndfvmh6ebrzcsu

NLP Technique for Malware Detection Using 1D CNN Fusion Model

Paul Ntim Yeboah, Haruna Balle Baz Musah
2022 Security and Communication Networks  
The proposed model automatically extracts features from semantically embedded n-grams of raw static operation code (opcodes) sequences to determine the maliciousness of a binary file.  ...  In this article, we employ a deep learning-based model consisting of 1-dimensional convolutional neural network (1D CNN) to automate the detection of Android malware.  ...  [21] for example represented Android API method sequences in semantic vector forms using word2vec embedding and further employed a 2-dimensional convolutional neural network to extract features to train  ... 
doi:10.1155/2022/2957203 doaj:33488b946a024bf7affa002ac7294395 fatcat:uwgqxp27d5b4nnhflnqx37mlnm

Subverting Privacy-Preserving GANs: Hiding Secrets in Sanitized Images [article]

Kang Liu, Benjamin Tan, Siddharth Garg
2020 arXiv   pre-print
Currently, state-of-the-art approaches use privacy-preserving generative adversarial networks (PP-GANs) for this purpose, for instance, to enable reliable facial expression recognition without leaking  ...  We show that it is possible to hide the sensitive identification data in the sanitized output images of such PP-GANs for later extraction, which can even allow for reconstruction of the entire input images  ...  An accompanying reveal network extracts the secret message from the a secret-embedded image.  ... 
arXiv:2009.09283v1 fatcat:6tclxdr6rjgr7dksrrqpwbq2f4

Multi-View Brain HyperConnectome AutoEncoder For Brain State Classification [article]

Alin Banka, Inis Buzi, Islem Rekik
2020 arXiv   pre-print
For each subject, we further regularize the hypergraph autoencoding by adversarial regularization to align the distribution of the learned hyperconnectome embeddings with that of the input hyperconnectomes  ...  Our experiments showed that the learned embeddings by HCAE yield to better results for brain state classification compared with other deep graph embedding methods methods.  ...  It utilizes conventional graph convolutional layers to learn the embeddings and adversarial regularizing network for original-encoded distribution alignment.  ... 
arXiv:2009.11553v1 fatcat:pkdsaihiqrarvfjs4wep3wdw3m

Semi-Supervised Adversarial Semantic Segmentation Network Using Transformer and Multiscale Convolution for High-Resolution Remote Sensing Imagery

Yalan Zheng, Mengyuan Yang, Min Wang, Xiaojun Qian, Rui Yang, Xin Zhang, Wen Dong
2022 Remote Sensing  
In this study, a novel semi-supervised adversarial semantic segmentation network is developed for remote sensing information extraction.  ...  A multiscale input convolution module (MICM) is designed to extract sufficient local features, while a Transformer module (TM) is applied for long-range dependency modeling.  ...  Acknowledgments: We would like to sincerely thank the editors and reviewers for their time. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/rs14081786 fatcat:dh3ilaolr5cp7i45p66rhr5k4u

Convolutional Neural Network-Based Digital Image Watermarking Adaptive to the Resolution of Image and Watermark

Jae-Eun Lee, Young-Ho Seo, Dong-Wook Kim
2020 Applied Sciences  
It is a convolutional neural network (CNN)-based scheme that consists of pre-processing networks for both host image and watermark, a watermark embedding network, an attack simulation for training, and  ...  First, the proposed network is as simple as the most profound path consists of only 13 CNN layers, which is through the pre-processing network, embedding network, and extraction network.  ...  WM extraction network and the adversary network also consist of 1 × 1 and 3 × 3 convolution layers.  ... 
doi:10.3390/app10196854 fatcat:xetpxgllqjdabi7stv7cae2u5a
« Previous Showing results 1 — 15 out of 18,555 results