AI Roadmaps
← All roadmaps
AdvancedRoadmap 28 · 12 steps · 3h 26m of video

Federated Learning & Privacy-Preserving ML

Train models without collecting the data: federated averaging, Flower simulations, differential privacy, secure aggregation, encryption, and the attacks these defend against.

Before this: Deep Learning (PyTorch training loops). Basic probability from Python & Math for AI helps for the differential privacy steps.

  1. 8 min
    Step 1

    What is federated learning?

    Train on many devices and share model updates instead of raw data.

    What is Federated Learning?

    codebasics

  2. 41 min
    Step 2

    Federated learning in production

    How Google trains phone keyboard models across millions of devices.

    Federated Learning: Machine Learning on Decentralized Data (Google I/O'19)

    TensorFlow

  3. 19 min
    Step 3

    Federated learning algorithms

    University lecture on FedAvg and the algorithms built on it.

    Lecture 11.3 - Federated Learning Algorithms

    UCF CRCV

  4. 10 min
    Step 4

    Your first Flower simulation

    Simulate federated training of a PyTorch model with Flower.

    Launching Your First Simulation - 2025 Tutorial: Federated AI Simulations with Flower

    Flower

    flwrlabs/flower★ 7.2k
  5. 21 min
    Step 5

    Stateful clients in Flower

    Keep per-client state across rounds, as personalized methods need.

    Designing Stateful ClientApps - 2025 Tutorial: Federated AI Simulations with Flower

    Flower

  6. 7 min
    Step 6

    Differential privacy

    Add calibrated noise so no single person changes the result much.

    Differential Privacy - Simply Explained

    Simply Explained

  7. 18 min
    Step 7

    The definition of differential privacy

    Cynthia Dwork, who co-invented it, explains the formal definition.

    The Definition of Differential Privacy - Cynthia Dwork

    Institute for Advanced Study

  8. 14 min
    Step 8

    Differential privacy in deep learning

    Whether DP stops models memorizing data, and tips for using it.

    Differential Privacy in Deep Learning and AI

    Probably Private

  9. 22 min
    Step 9

    Membership inference attacks

    How attackers tell whether a record was in the training set.

    Membership inference attacks from first principles

    IEEE Symposium on Security and Privacy

  10. 22 min
    Step 10

    Secure aggregation

    The server learns the sum of client updates, not each update.

    11 1 17 ACM CCS Aaron Segal Practical Secure Aggregation Machine Learning

    Association of Computing Machinery 2017

  11. 3 min
    Step 11

    Homomorphic encryption

    Compute on encrypted data without decrypting it first.

    A 3-minute introduction to Fully Homomorphic Encryption by a cryptographer

    Zama

  12. 21 min
    Step 12

    Federated fine-tuning of LLMs

    A benchmark for fine-tuning language models across many data owners.

    FlowerTune: A Cross-Domain Benchmark for Federated Fine-Tuning of Large Language Models

    Flower

Build it

Watching isn't learning. Pick one and put it on GitHub.

Browse all roadmaps →