Calendar & Syllabus

Week 1 - Introduction and Privacy Attacks

Jan 29

Introduction & Logistics

Slides   Reading - Sec. 1.1-1.3 of DPAI Book Extra - Narayanan & Shmatikov (2008)

HW 0 released HW0 PDF

Jan 21

Privacy Attacks - Membership Inference

Required - Ch. 5 of DPAI Book Extra - Shokri et al. (2017), Carlini et al. (2021)

Jan 22

Privacy Attacks - Data Reconstruction

Required - Hayes et al. (2025) Extra - Ahmed et al. (2026)

Week 2 - Differential Privacy Basics I

Jan 28

Introduction to differential privacy (DP)

Reading - Sec. 1.4 of DPAI Book Extra - Sec. 1.4-1.6 of Vadhan

Jan 29

Class Canceled

Jan 30

HW 0 due

Week 3 - DP Basics

Feb 4

DP and membership inference

Reading - Sec. 5.4.2 of DPAI book. Proof: Theorem 2.1 of Kairouz et al.

Feb 5

DP - basic composition

Reading - Sec. 3.2 of DPAI book

Week 4 - DP Basics

Feb 9

Lab PyTorch Review

Additional tutorials

Feb 11

Privacy Loss Distribution

Reading - Sec. 3.3 and 3.4 of DPAI book

Feb 12

Concentration of measure

Reading - Duchi’s notes, Sec. 1 - 3 of Rivasplata’s note. Detailed reading (optional) - Chap. 2 of Vershynin’s HDP book

Week 5 - Concentrated DP, Amplification

Feb 16

Lab Privacy Accounting

dp-accounting library

Feb 18

Concentrated DP & advanced composition

Reading - Sec. 3.4.1 and 3.4.2 of DPAI book

Feb 19

Amplification by Subsampling

Reading - Sec. 3.6.1-3.6.3 of DPAI book

Week 6 - Renyi DP and DP-SGD

Feb 23

HW1 Tutorial Doubt Solving and Tutorial Session on HW1.

Feb 25

Renyi DP & Amplification by Subsampling

Reading - Sec. 3.6.4 and 3.6.5 of DPAI book

Feb 26

Subsampled Gaussian Mechanism

Reading - Mironov et al.

Week 7 - Per-sample gradients and Guest Lecture

Mar 2

Lab Per Sample Gradients

Tutorial

Mar 4

Non-instructional Day: Holi

Mar 5

(Online) Guest Lecture by Anjalie Field on applied privacy in NLP domain

Week 8 - Correlated Noise Mechanisms

Mar 9

HW2 Tutorial Doubt Solving and Tutorial Session on HW2.

Mar 11

Stochastic Gradient Descent (SGD) with DP & Practical Considerations

Reading - Sec. 4.2 of Ponomareva et al. (Sec. 5 is also strongly recommended)

Mar 12

Private learning with Correlated Noise Mechanism - I

Reading - Chapter 1 of Correlated Noise Mechanisms Monograph

Week 9 - Exponential Mechanisms and Private Inference in LLMs

Mar 16

Lab DP-SGD

Implementation of Section 3.1 of Paper

Mar 18

Private learning with Correlated Noise Mechanism - II

Reading - Chapter 2 of Correlated Noise Mechanisms Monograph

Mar 19

Private LLM inference (Lecture by Vishnu Vinod)

Reading - Invisible Ink

Week 10 - Advanced Topics

Mar 23

Tutorial Project Office Hours.

Mar 25

Model provenance via membership inference

Reading - Kuditipudi et al.

Mar 26

Model fingerprinting

Reading - Nasery et al.