Intro
Preface
Organization
Contents
Privacy
Exploring Privacy-Preserving Techniques on Synthetic Data as a Defense Against Model Inversion Attacks
1 Introduction
2 Threat Model
3 Background and Related Work
3.1 Synthetic Data Generation
3.2 Privacy-Preserving Techniques
3.3 Model Inversion Attribute Inference Attacks
3.4 Attribute Disclosure Risk
4 Experimental Setup
4.1 Data Set
4.2 Privacy-Preserving Techniques on Synthetic Training Data
4.3 Target Machine Learning Model
4.4 Model Inversion Attribute Inference Attacks

5 Performance of the Target Models
6 Results of Model Inversion Attribute Inference Attacks
6.1 Attacks on the Model Trained on Original Data
6.2 Attacks on the Model Trained on Protected Synthetic Data
7 Correct Attribution Probability
8 Conclusion and Future Work
References
Privacy-Preserving Medical Data Generation Using Adversarial Learning
1 Introduction
2 Background
2.1 Related Works
2.2 Differential Privacy
2.3 Rényi Differential Privacy
3 Algorithmic Framework
3.1 GAN
3.2 Variational Autoencoders
3.3 Model Architecture
3.4 Privacy Loss

4 Experimental Evaluation
4.1 Datasets
4.2 Comparison
4.3 Synthetic Data Generation
5 Conclusion
References
Balanced Privacy Budget Allocation for Privacy-Preserving Machine Learning
1 Introduction
2 Preliminary
2.1 Notation
2.2 Local Differential Privacy
2.3 Classification Methods Using Machine Learning
3 Related Work
3.1 Unified LDP-Algorithm
3.2 Scalable Unified Privacy-Preserving Machine Learning Framework (SUPM)
4 Contribution-Based Privacy-Budget Allocation
4.1 Contribution-Based Dimension Reduction Using Odds Ratio

4.2 Privacy-Preserving Machine Learning with Balanced Privacy Budget Allocation
5 Experiments Analysis
5.1 Experiment Settings
5.2 Logistic Regression
5.3 Support Vector Machine
6 Conclusion
References
Intrusion Detection and Systems
SIFAST: An Efficient Unix Shell Embedding Framework for Malicious Detection
1 Introduction
2 Problem Definition and Background
2.1 Malicious Unix Shell Operations
2.2 Threat Model
3 Related Works
3.1 Command and Script Detection Using NLP Techniques
3.2 Command and Script Detection Using Hybrid Features
4 Methodology

4.1 Data Preprocessing and AShellTokenizer
4.2 Command Embedding
4.3 Script Embedding
5 Experiment Settings and Results
5.1 Experiment Settings
5.2 Evaluations
6 Conclusion
References
VNGuard: Intrusion Detection System for In-Vehicle Networks
1 Introduction
2 Background and Related Work
2.1 Local Interconnect Network (LIN)
2.2 Automotive Ethernet (AE)
2.3 Intrusion Detection Systems for In-Vehicle Networks
3 Attack Scenarios for LIN
4 Attack Scenarios for AE
5 Methodology
5.1 Data Extraction
5.2 Data Pre-processing
5.3 Model Structure