Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep Learning
The ever-evolving cybersecurity environment has given rise to sophisticated adversaries who constantly explore new ways to attack cyberinfrastructure. Recently, the use of deep learning-based intrusion detection systems has been on the rise. This rise is due to deep neural networks (DNN) complexity...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/13/6451 |
| _version_ | 1827655519833161728 |
|---|---|
| author | Kudzai Sauka Gun-Yoo Shin Dong-Wook Kim Myung-Mook Han |
| author_facet | Kudzai Sauka Gun-Yoo Shin Dong-Wook Kim Myung-Mook Han |
| author_sort | Kudzai Sauka |
| collection | DOAJ |
| description | The ever-evolving cybersecurity environment has given rise to sophisticated adversaries who constantly explore new ways to attack cyberinfrastructure. Recently, the use of deep learning-based intrusion detection systems has been on the rise. This rise is due to deep neural networks (DNN) complexity and efficiency in making anomaly detection activities more accurate. However, the complexity of these models makes them black-box models, as they lack explainability and interpretability. Not only is the DNN perceived as a black-box model, but recent research evidence has also shown that they are vulnerable to adversarial attacks. This paper developed an adversarial robust and explainable network intrusion detection system based on deep learning by applying adversarial training and implementing explainable AI techniques. In our experiments with the NSL-KDD dataset, the PGD adversarial-trained model was a more robust model than DeepFool adversarial-trained and FGSM adversarial-trained models, with a ROC-AUC of 0.87. The FGSM attack did not affect the PGD adversarial-trained model’s ROC-AUC, while the DeepFool attack caused a minimal 9.20% reduction in PGD adversarial-trained model’s ROC-AUC. PGD attack caused a 15.12% reduction in the DeepFool adversarial-trained model’s ROC-AUC and a 12.79% reduction in FGSM trained model’s ROC-AUC. |
| first_indexed | 2024-03-09T22:08:50Z |
| format | Article |
| id | doaj.art-d11145a82960455084cc517e10227a57 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T22:08:50Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-d11145a82960455084cc517e10227a572023-11-23T19:36:51ZengMDPI AGApplied Sciences2076-34172022-06-011213645110.3390/app12136451Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep LearningKudzai Sauka0Gun-Yoo Shin1Dong-Wook Kim2Myung-Mook Han3School of Computing, Gachon University, Seongnam-si 13120, KoreaSchool of Computing, Gachon University, Seongnam-si 13120, KoreaSchool of Computing, Gachon University, Seongnam-si 13120, KoreaSchool of Computing, Gachon University, Seongnam-si 13120, KoreaThe ever-evolving cybersecurity environment has given rise to sophisticated adversaries who constantly explore new ways to attack cyberinfrastructure. Recently, the use of deep learning-based intrusion detection systems has been on the rise. This rise is due to deep neural networks (DNN) complexity and efficiency in making anomaly detection activities more accurate. However, the complexity of these models makes them black-box models, as they lack explainability and interpretability. Not only is the DNN perceived as a black-box model, but recent research evidence has also shown that they are vulnerable to adversarial attacks. This paper developed an adversarial robust and explainable network intrusion detection system based on deep learning by applying adversarial training and implementing explainable AI techniques. In our experiments with the NSL-KDD dataset, the PGD adversarial-trained model was a more robust model than DeepFool adversarial-trained and FGSM adversarial-trained models, with a ROC-AUC of 0.87. The FGSM attack did not affect the PGD adversarial-trained model’s ROC-AUC, while the DeepFool attack caused a minimal 9.20% reduction in PGD adversarial-trained model’s ROC-AUC. PGD attack caused a 15.12% reduction in the DeepFool adversarial-trained model’s ROC-AUC and a 12.79% reduction in FGSM trained model’s ROC-AUC.https://www.mdpi.com/2076-3417/12/13/6451machine learningadversarial attacksexplainablenetwork intrusion detection systemdeep neural networks (DNN)adversarial robust |
| spellingShingle | Kudzai Sauka Gun-Yoo Shin Dong-Wook Kim Myung-Mook Han Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep Learning Applied Sciences machine learning adversarial attacks explainable network intrusion detection system deep neural networks (DNN) adversarial robust |
| title | Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep Learning |
| title_full | Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep Learning |
| title_fullStr | Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep Learning |
| title_full_unstemmed | Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep Learning |
| title_short | Adversarial Robust and Explainable Network Intrusion Detection Systems Based on Deep Learning |
| title_sort | adversarial robust and explainable network intrusion detection systems based on deep learning |
| topic | machine learning adversarial attacks explainable network intrusion detection system deep neural networks (DNN) adversarial robust |
| url | https://www.mdpi.com/2076-3417/12/13/6451 |
| work_keys_str_mv | AT kudzaisauka adversarialrobustandexplainablenetworkintrusiondetectionsystemsbasedondeeplearning AT gunyooshin adversarialrobustandexplainablenetworkintrusiondetectionsystemsbasedondeeplearning AT dongwookkim adversarialrobustandexplainablenetworkintrusiondetectionsystemsbasedondeeplearning AT myungmookhan adversarialrobustandexplainablenetworkintrusiondetectionsystemsbasedondeeplearning |