Deep learning applications for cyber security / Mamoun Alazeb, MingJian Tang, editors.

Intro; Foreword; Acknowledgements; Contents; About the Authors; Adversarial Attack, Defense, and Applications with Deep Learning Frameworks; 1 Introduction; 2 Methods for Generating Adversarial Samples; 2.1 Box-Constrained L-BFGS Approach; 2.2 Fast Gradient Sign Method; 2.3 Iterative Gradient Sign Methods; 2.4 Jacobian-Based Saliency Map Attack; 2.5 DeepFool; 2.6 Carlini & Wagner Attack; 2.7 Transferability Based Approach; 2.8 Houdini; 3 Methods for Defending Adversarial Attacks; 3.1 Adversarial Training; 3.2 Defensive Distillation; 3.3 Game Theory Based Approach; 3.4 MagNet; 3.5 Defense-GAN

4 Adversarial Learning Applications for Cyber Security4.1 Attack Commercial Web Services; 4.2 Attack Automatic Speech Recognition System; 4.3 Attack Malware Classifier; 5 Conclusion; References; Intelligent Situational-Awareness Architecture for Hybrid Emergency Power Systems in More Electric Aircraft; 1 Introduction; 2 Problem Setting; 3 Situation-Aware Intelligent Security Control Architecture for Energy Management Strategy; 3.1 Deep Learning-Based Cyber Attack Detection Scheme; 3.2 Adaptive Neuro-Fuzzy Inference System (ANFIS)-Based Estimation Scheme; 4 Simulation Results

4.1 Deep Learning Implementation and Results4.2 ANFIS Implementation and Results; 5 Conclusion; References; Deep Learning in Person Re-identification for Cyber-Physical Surveillance Systems; 1 Introduction; 2 Background; 2.1 Supervised Learning in Person Re-identification; 2.2 Unsupervised Learning in Person Re-identification; 3 Deep Learning in Person Re-identification for Cyber-Physical Surveillance Systems: New Trending Methodologies; 3.1 Image-Based Person Re-identification; 3.1.1 Supervised Deep Representation Learning; 3.1.2 Deep Hashing Learning for Fast Person Re-identification

3.1.3 Unsupervised Deep Representation Learning with Generative Models3.2 Video-Based Person Re-identification; 4 Conclusions; 5 Open Problems; Problems; Problem 1; Problem 2; References; Deep Learning-Based Detection of Electricity Theft Cyber-Attacks in Smart Grid AMI Networks; 1 Introduction; 2 Review of State-of-the-Art Detection Techniques; 2.1 Hardware-Based Solutions; 2.2 Software-Based Solutions; 3 Dataset Used and Cyber Attacks Injection; 3.1 Energy Consumption Dataset; 3.2 Electricity Theft Cyber-Attacks; 4 Proposed Deep Electricity Theft Detectors

4.1 Deep Customer-Specific Electricity Theft Detector4.2 Feedforward Detector; 4.3 Gated Recurrent Detector; 4.4 Deep Generalized Electricity Theft Detector; 5 Hyper-Parameters Tunning; 5.1 Sequential Grid Search Tunning for Hyper-Parameters; 5.2 Random Grid Search Tunning for Hyper-Parameters; 5.3 Genetic Algorithm-Based Tunning for Hyper-Parameters; 6 Numerical Results and Discussion; 6.1 Customer-Specific Detector Evaluation with Sequential Grid Search for Hyper-Parameters Tuning; 6.2 Generalized Detector Evaluation with Random Grid Search for Hyper-Parameters Tuning; 6.3 Generalized Detector Evaluation with Genetic-Based Search for Hyper-Parameters Tuning