(Click Category to List Courses)
37 - ITC - Information Technology - Miscellaneous
ITC 137 - Differential Privacy for Machine Learning (7 Days)
| Code | Start Date | Duration | Venue | |
|---|---|---|---|---|
| ITC 137 | 02 December 2024 | 7 Days | Istanbul | Registration Form Link |
| ITC 137 | 20 January 2025 | 7 Days | Istanbul | Registration Form Link |
| ITC 137 | 17 March 2025 | 7 Days | Istanbul | Registration Form Link |
| ITC 137 | 12 May 2025 | 7 Days | Istanbul | Registration Form Link |
| ITC 137 | 07 July 2025 | 7 Days | Istanbul | Registration Form Link |
| ITC 137 | 01 September 2025 | 7 Days | Istanbul | Registration Form Link |
| ITC 137 | 27 October 2025 | 7 Days | Istanbul | Registration Form Link |
| ITC 137 | 22 December 2025 | 7 Days | Istanbul | Registration Form Link |
Please contact us for fees
Course Description
Differential Privacy for Machine Learning is a comprehensive course that explores the intersection of privacy and machine learning. The course covers the fundamental concepts of differential privacy, a rigorous framework for quantifying the privacy guarantees of algorithms. Participants will learn how to design machine learning models that preserve privacy while maintaining utility and accuracy.
Course Objectives
- Understand the fundamental concepts of differential privacy and its importance in protecting sensitive data in machine learning applications.
- Learn how to design and implement privacy-preserving machine learning algorithms that provide strong privacy guarantees.
- Explore techniques for privacy-preserving data analysis and machine learning model training.
- Gain practical experience in applying differential privacy principles to real-world machine learning projects.
- Understand the trade-offs between privacy, utility, and accuracy in machine learning models.
Who Should Attend?
- Machine Learning Engineers
- Cybersecurity professionals
- Privacy Professionals
- Policy Makers
Course Details/Schedule
Day 1
- Privacy in AI
- AI Learning Threats
- How data is processed inside Machine Learning algorithms
- Why privacy protection in Machine Learning is important?
- Privacy-Preserving Machine Learning
- Differential Privacy Overview
- How Differential Privacy Works?
- Cryptography for Privacy-Preserving Computation
- Advanced Concepts of Differential Privacy for Machine Learning
Day 2
- Applying Differential Privacy in Machine Learning
- Privacy-Preserving Synthetic Data Generation
- Fundamentals of Synthetic Data Generation
- Differential Privacy for Privacy-Preserving Synthetic
Day 3
- Data Generation
- Coding Practices for Privacy-Preserving Machine Learning
- Robustness of Machine Learning Applications via Privacy-Preserving
Day 4
- Hacking AI
- AI Security Overview
- How to hack AI?
- Adversarial Attacks
- Attack Types for Machine Learning
- Reconstruction Attacks
- Model Inversion Attacks
- Membership Inference Attacks
- Re-Identification Attacks
- AI Hacking (Adversarial Attack) Demos with PyTorch and TensorFlow
Day 5
- Secure and Encrypted Machine Learning
- Secure AI
- How to protect AI from Attackers?
- Requirements of Secure Machine Learning
- Federated Learning
- Federated Learning Overview
- Federated Learning Architecture
- Federated Learning Application Development with TensorFlow
Day 6
- Encrypted Machine Learning Overview
- How it works?
- Encrypted Machine Learning Tools
- Tenseal
- Concrete ML
- How to Choose a Homomorphic Encryption Library/Tools
- Homomorphic Encryption
- Homomorphic Encryption Overview
- Partially Homomorphic Encryption with RSA in Python from Scratch
Day 7
- Fully Homomorphic Encryption with RSA in Python from Scratch(concreteML)
- Fully Homomorphic Encryption with Tenseal-
- TFHE with Concrete ML
- Project development of Covid dataset using pretrained model in ConcreteML and Demo
- Project development of heart disease dataset using pretrained model in concreteML and demo
