Privacy-Preserving Machine Learning with Federated Learning Training is an advanced engineering and research-oriented program focused on developing distributed machine learning models in environments where collecting sensitive data in a central location is restricted or impractical.

Federated Learning (FL) enables multiple organizations or devices to collaboratively train a machine learning model without transferring their raw datasets to a central server. Instead, models are trained where the data resides, while model updates are shared to improve a common global model.

The course begins with the FedAvg algorithm and explores Cross-Silo and Cross-Device architectures, centralized and decentralized learning approaches, Non-IID Data (Statistical Heterogeneity), and differences in computing and network resources.

Participants also examine methods such as FedProx, designed to address challenges associated with heterogeneous data, as well as Privacy-Enhancing Technologies (PETs) including Differential Privacy (DP), Homomorphic Encryption (HE), and Secure Multi-Party Computation (SMPC).

Through hands-on activities with Flower and TensorFlow Federated (TFF), participants gain practical exposure to distributed model training approaches that prioritize data privacy.

For corporate groups, we can organize this training as an on-site private group.

Prerequisites

There are no mandatory prerequisites for attending this training.

Who Should Attend

Data Privacy Engineers and Security Architects
Infrastructure and MLOps Engineers
Embedded Systems Specialists and Mobile Developers
Senior AI and Machine Learning Engineers
Machine Learning Researchers
Technical professionals working with sensitive data in healthcare, finance, and telecommunications
CTOs, Technology Leaders, and R&D Directors

What You Will Learn

Upon completing this training, you will be able to:

Explain Federated Learning architectures and how they differ from centralized machine learning.
Develop collaborative machine learning models across distributed datasets.
Understand the FedAvg training cycle and the purpose of FedProx.
Compare Cross-Silo and Cross-Device Federated Learning architectures.
Evaluate challenges related to Non-IID data and system heterogeneity.
Identify privacy-enhancing methods such as Differential Privacy, Homomorphic Encryption, and SMPC.
Relate the data privacy requirements of GDPR, HIPAA, and KVKK to architectural decisions.
Work with Flower and TensorFlow Federated to explore distributed model training.
Design approaches that share model updates rather than centralizing raw data.
Evaluate privacy-preserving AI solutions for industries that process sensitive information.

Training Outline

1. Foundations of Privacy-Preserving AI and Requirements Analysis

Limitations of Traditional Centralized Machine Learning

Examining the centralized machine learning approach
Understanding data privacy risks
- Data breaches
- Secondary use of data
- Identity disclosure
The impact of regulations such as GDPR, HIPAA, and KVKK on centralized data processing
Challenges associated with data silos, collaboration, and competitive opportunities

Privacy-Enhancing Technologies (PETs)

Data anonymization techniques and their limitations
Re-identification attacks
Differential Privacy (DP)
Homomorphic Encryption (HE)
Secure Multi-Party Computation (SMPC)
Zero-Knowledge Proofs (ZKP)
The relationship between Federated Learning and other privacy-enhancing technologies

2. Federated Learning Architecture and Core Algorithms

How Federated Learning Works

Fundamental components of Federated Learning
Interaction between the central server and distributed clients
Distributing models to participating clients
Training models on local datasets
Sharing model weights and gradient updates
Aggregating updates to improve the global model

FedAvg Algorithm

Understanding Federated Averaging (FedAvg)
The local training and global aggregation cycle
Collaborative model training without sharing raw data

Federated Learning Architectures

Cross-Silo Federated Learning: Collaborative training across organizations
Cross-Device Federated Learning: Training across large numbers of edge devices
Centralized Federated Learning
Decentralized (Peer-to-Peer / P2P) Federated Learning
Hierarchical Federated Learning

3. Key Challenges in Federated Learning: Heterogeneity

Understanding Non-IID Data
Statistical heterogeneity across participating clients
The effects of different data distributions on model training
System heterogeneity
- Differences in computational resources
- Network connectivity variations
- Device and infrastructure limitations
Communication bottlenecks and network costs
The role of FedProx in heterogeneous learning environments

4. Enhancing Security: Differential Privacy and Cryptographic Methods

Differential Privacy

Introduction to Differential Privacy (DP)
Adding mathematical noise to model updates
Reducing the risk of exposing individual data contributions

Cryptographic Privacy Techniques

Homomorphic Encryption (HE) and computation on encrypted data
Aggregating encrypted model updates
Secure Multi-Party Computation (SMPC)
Combining Federated Learning with privacy-enhancing technologies

5. Hands-On Workshop: Flower and TensorFlow Federated (TFF)

Introduction to the Flower library
Python-based, framework-agnostic Federated Learning approaches
Exploring TensorFlow Federated (TFF)
Practical evaluation of distributed model training
Working with local model training and global model updates

Why Choose Us

Leading UK-based global training provider since 1995.

Experience Privacy-Preserving Machine Learning with Federated Learning in Australia through Bilginç IT Academy's live and interactive virtual classroom environment. Join a Public course as an individual delegate or arrange a dedicated Private / In-house online training program exclusively for your organization.

Delivery Method: Online Instructor-Led
Participation Model: Public / Private (In-house)
Live and Interactive Training: Connect with your instructor in real time and actively participate through discussions, Q&A sessions, practical exercises, and group activities.
Flexible Participation: Join the training from your home, office, or any location with a suitable internet connection.
Expert Trainer Network: Learn from experienced trainers with strong industry backgrounds and practical field expertise.
Over 30 Years of Training Expertise: Benefit from Bilginç IT Academy's professional training experience since 1995.
Worldwide Access: Join our live virtual classrooms from Australia or anywhere else in the world, or arrange a dedicated online training program for your organization.

Experience Privacy-Preserving Machine Learning with Federated Learning through face-to-face Classroom Based training in Australia. Training can be delivered as a Public course open to individual delegates or as a dedicated Private / In-house class for your organization.

Delivery Method: Classroom Based
Participation Model: Public / Private (In-house)
Face-to-Face Learning: Interact directly with your instructor and fellow delegates in an engaging classroom environment.
Experienced Trainers: Learn from specialists with extensive industry experience and practical real-world knowledge.
Professional Training Environment: Attend training in comfortable, well-equipped classrooms designed to support effective learning.
Practical Learning: Depending on the course, reinforce your knowledge through hands-on exercises, scenarios, case studies, and instructor-led activities.

Arrange Privacy-Preserving Machine Learning with Federated Learning in Australia as a dedicated Onsite training program for your organization. Bilginç IT Academy trainers can deliver the training at your office or another location of your choice, with the program planned around your team's requirements and business objectives.

Delivery Method: Onsite
Participation Model: Private (In-house)
Training at Your Preferred Location: Organize the training at your company's office or another location selected by your organization.
Tailored Course Content: Adapt the training program to your projects, team structure, existing skill levels, and specific business requirements.
Team-Focused Learning: Develop your team around a shared knowledge base while strengthening internal collaboration and knowledge transfer.
Flexible Scheduling: Plan the training dates, location, and program according to your organization's operational requirements.
Worldwide Onsite Delivery: Arrange the training in Australia or at another preferred location worldwide. Bilginç IT Academy trainers can travel to your selected location to deliver the dedicated training program.

Privacy-Preserving Machine Learning with Federated Learning Training in Australia