Federated Learning : Security and Privacy (Hardcover)

1. Introduction to Machine Learning a. Types of Learning b. Learning Tasks c. Cost Function d. Optimization e. Evaluation Metrics f. Artificial Neural Network g. Implementation 2. Federated Learning a. Importance of FL b. Types of FL c. Applications in FL d. Challenges in FL e. Security and Privacy Issues f. Defense Techniques g. Privacy-Preserving Byzantine-Robust FL h. Implementation 3. Poisoning Attacks in FL a. Attacker b. Label flipping attack c. Gaussian attack d. LIE attack e. Krum attack f. Trim attack g. Shejwalkar attack h. Scaling attack i. Edge attack j. Vulnerabilities in Cosine Similarity-based Defenses k. Implementation 4. Inference Attacks in FL a. Attacker goal b. Data reconstruction attacks c. Membership inference attacks d. Property inference attacks e. Implementation 5. Byzantine Robust Defenses a. Design goals b. Krum c. Median and Trimmed Mean d. Bulyan e. FoolsGold f. FLTrust g. Moat h. DeFL i. RDFL j. FLTC k. Implementation 6. Privacy-Preserving FL a. Differential Privacy b. DPFL: A Client Level c. Homomorphic d. BatchCrypt: HE-based Scheme e. Threshold Multi-key HE Scheme f. Secure Multi-Party Computation g. Practical Secure Aggregation h. Summary i. Implementation