May the Fuzz Be with You

Übersicht

Explore how LSTM neural networks and Python can generate AI-driven test inputs for software testing and network security, improving upon traditional fuzzing techniques.

Lehrplan

• Introduction to Fuzzing Techniques
Overview of Traditional Fuzzing Methods
Limitations and Challenges of Traditional Fuzzing
• Introduction to LSTM Neural Networks
Understanding Recurrent Neural Networks (RNNs)
Detailed Study of LSTMs: Architecture and Functionality
• Python for AI-Driven Fuzzing
Setting Up Python Environment for AI Development
Key Python Libraries: TensorFlow, Keras
• Implementing LSTMs for Test Input Generation
Preparing Datasets for LSTM Training
Training LSTMs on Existing Software Inputs
Generating AI-Driven Test Inputs with LSTMs
• Application of AI-Driven Fuzzing in Software Testing
Case Studies and Practical Examples
Evaluating the Effectiveness of AI-Generated Inputs
Comparing AI-Driven Fuzzing to Traditional Methods
• Enhancing Network Security with AI-Driven Fuzzing
Identifying Network Vulnerabilities with LSTM Models
Simulating Network Attacks Using Generated Inputs
• Advanced Topics and Future Directions
Improvements in AI Algorithms for Fuzzing
Combining AI-Driven Techniques with Other Testing Methods
• Practical Workshop
Step-by-Step Implementation of LSTM for Fuzzing
Hands-On Projects: Building and Testing AI-Driven Cases
• Course Wrap-Up and Assessments
Summary of Key Learnings
Final Project and Assessment Guidelines

---

Fachgebiete

Conference Talks