A few-shot deep learning approach for improved intrusion detection

Chowdhury, Moin Uddin; Hammond, Frederick; Konowicz, Glenn; Xin, Chunsheng; Wu, Hongyi; Jiang, Li

doi:10.1109/uemcon.2017.8249084

Cited by 94 publications

(36 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Deep learning models achieved superb results in different areas in recent years such as object detection and tracking [5][6][7][8], image classification [9][10][11][12][13], remote sensing [12,[14][15][16][17][18], speech recognition [19,20], autonomous driving [21,22], cybersecurity [23,24], and medical imaging [25]. Among various structures, CNNs are the most popular models for different applications.…”

Section: Deep Learningmentioning

confidence: 99%

“…In [52], the authors demonstrated that the performance of a deep belief network (DBN) for post-traumatic stress disorder (PTSD) diagnosis could be significantly improved using transfer learning. Chowdhury et al [24] developed a few-shot deep learning approach for intrusion detection, in which they extracted features of a small dataset from some trained CNN and DBN models, and trained a simple classifier to improve the intrusion detection performances. Transfer learning with CNN for regression problems has been studied in [49,53].…”

Section: Transfer Learningmentioning

confidence: 99%

See 1 more Smart Citation

Quantifying Seagrass Distribution in Coastal Water with Deep Learning Models

et al. 2020

Self Cite

View full text Add to dashboard Cite

Coastal ecosystems are critically affected by seagrass, both economically and ecologically. However, reliable seagrass distribution information is lacking in nearly all parts of the world because of the excessive costs associated with its assessment. In this paper, we develop two deep learning models for automatic seagrass distribution quantification based on 8-band satellite imagery. Specifically, we implemented a deep capsule network (DCN) and a deep convolutional neural network (CNN) to assess seagrass distribution through regression. The DCN model first determines whether seagrass is presented in the image through classification. Second, if seagrass is presented in the image, it quantifies the seagrass through regression. During training, the regression and classification modules are jointly optimized to achieve end-to-end learning. The CNN model is strictly trained for regression in seagrass and non-seagrass patches. In addition, we propose a transfer learning approach to transfer knowledge in the trained deep models at one location to perform seagrass quantification at a different location. We evaluate the proposed methods in three WorldView-2 satellite images taken from the coastal area in Florida. Experimental results show that the proposed deep DCN and CNN models performed similarly and achieved much better results than a linear regression model and a support vector machine. We also demonstrate that using transfer learning techniques for the quantification of seagrass significantly improved the results as compared to directly applying the deep models to new locations.

show abstract

Section: Deep Learningmentioning

confidence: 99%

Section: Transfer Learningmentioning

confidence: 99%

Quantifying Seagrass Distribution in Coastal Water with Deep Learning Models

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…The results on the Kyoto2013 dataset show that the performance of the GRU_SVM model achieved significant improvement over the traditional GRU model. Chowdhury et al [ 30 ] trained a DCNN as a feature extractor to extract the output of each layer of the CNN as the input of the SVM. This method was inspired by small sample learning and solves the sample imbalance problem.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Intrusion Detection Model Combining SAE with Kernel Approximation in Internet of Things

Lee

Gong

et al. 2020

Sensors

View full text Add to dashboard Cite

Owing to the constraints of time and space complexity, network intrusion detection systems (NIDSs) based on support vector machines (SVMs) face the “curse of dimensionality” in a large-scale, high-dimensional feature space. This study proposes a joint training model that combines a stacked autoencoder (SAE) with an SVM and the kernel approximation technique. The training model uses the SAE to perform feature dimension reduction, uses random Fourier features to perform kernel approximation, and then random Fourier mapping is explicitly applied to the sub-sample to generate the random feature space, making it possible to apply a linear SVM to uniformly approximate to the Gaussian kernel SVM. Finally, the SAE performs joint training with the efficient linear SVM. We studied the effects of an SAE structure and a random Fourier feature on classification performance, and compared that performance with that of other training models, including some without kernel approximation. At the same time, we compare the accuracy of the proposed model with that of other models, which include basic machine learning models and the state-of-the-art models in other literatures. The experimental results demonstrate that the proposed model outperforms the previously proposed methods in terms of classification performance and also reduces the training time. Our model is feasible and works efficiently on large-scale datasets.

show abstract

“…This is mainly due to two factors, ie, using balanced dataset through training and testing stages and using adaptive tuned neurons of high selectivity in RBF network where neurons of high activation to specific attacks respond with high activation to it and responds with almost zero activation to other others. 54 Self-Taught Machine (STL) NSL-KDD 79.1 Niyaz et al 54 Soft-max Regression (SMR) NSL-KDD 75.23 Chowdhury et al 45 Deep Neural Network NSL-KDD 94.62 Bamakan et al 30 Time In order to have better insight into the multi-class experimental results obtained from our hybrid model, many comparisons with other existing (single and hybrid) schemes using KDD Cup'99 and NSL-KDD datasets have been conducted against overall detection accuracy as presented in Table 10 and against detection rate per class, as shown in Figure 8.…”

Section: Referring Tomentioning

confidence: 99%

Cascaded hybrid intrusion detection model based on SOM and RBF neural networks

Almiani

AbuGhazleh

Al‐Rahayfeh

et al. 2019

Concurrency and Computation

View full text Add to dashboard Cite

Cybercriminal activities over computer network systems are considered one of the preponderant issues that humanity will face in the coming two decades. The development steps in the design of intrusion detection systems must be carried out in analogous manner to sophistication levels of intrusions developed by hackers. This work proposes a layered hybrid intrusion detection model uses cascaded layers of Clustered Self-Organized Map (CSOM) and Radial Basis Function (RBF) neural networks to improve the efficiency of detecting frequent and least frequent intrusions. K-means clustered SOM was used to filter attacks as a first layer, whereas RBF-based neural network worked as second filtering and attacked classification layer leading to significance reduction in time required to process connection records and notable improvements in the performance of intrusion detection. A new balanced version of cleansed NSL-KDD dataset was used to validate and evaluate the proposed model. Compared with other existing schemes; the proposed model shows high detection performance in terms of accuracy 97.73% and false positive rate as low as 0.023%. In particular, for detecting least and most harmful attacks, U2R and R2L, the system achieved detection rate of 88.6% with false positive rate of 0.016. Comparative results showed that CSOM-RBF model is more suitable for real-life implementation than other many existing state-of-the-art intrusion detection models.

show abstract

A few-shot deep learning approach for improved intrusion detection

Cited by 94 publications

References 11 publications

Quantifying Seagrass Distribution in Coastal Water with Deep Learning Models

Quantifying Seagrass Distribution in Coastal Water with Deep Learning Models

A Hybrid Intrusion Detection Model Combining SAE with Kernel Approximation in Internet of Things

Cascaded hybrid intrusion detection model based on SOM and RBF neural networks

Contact Info

Product

Resources

About