Differential Privacy |
|
|
|
|
| 9/2 |
Course welcome
Reading: How to Read a Paper |
Justin |
--- |
[slides] |
| 9/4 |
Basic private mechanisms
Reading: Dwork and Roth 3.2-4 |
Justin |
--- |
|
| 9/7 |
NO CLASS: LABOR DAY |
|
|
|
| 9/9 |
Composition and closure properties
Reading: Dwork and Roth 3.5 |
Justin |
--- |
Signups Due |
| 9/11 |
What does differential privacy actually mean?
Reading: Lunchtime for Differential Privacy |
Justin |
--- |
|
| 9/14 |
Private machine learning
Reading: On the Protection of Private Information in Machine Learning Systems: Two Recent Approaches
Reading: Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data |
Nathan/Matt T. |
Saniya/Marcus |
|
| 9/16 |
Privately generating synthetic data
Reading: A Simple and Practical Algorithm for Differentially Private Data Release
Reading: Private Post-GAN Boosting |
Zijian/Yuchen |
Deepan/Kendall |
|
Adversarial Machine Learning |
|
|
|
|
| 9/18 |
Overview and basic concepts |
Justin |
--- |
|
| 9/21 |
Adversarial examples
Reading: Intriguing Properties of Neural Networks
Reading: Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
See also: Explaining and Harnessing Adversarial Examples |
Deepan and Kendall |
Keaton/Anna |
|
| 9/23 |
Data poisoning
Reading: Poisoning Attacks against Support Vector Machines
Reading: Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networks |
Grishma/Lokit |
Amos/Suleman |
|
| 9/25 |
Defenses and detection: challenges
Reading: Towards Evaluating the Robustness of Neural Networks
Reading: Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methods |
Justin |
--- |
|
| 9/28 |
Certified defenses
Reading: Certified Defenses for Data Poisoning Attacks
Reading: Certified Defenses against Adversarial Examples |
Yucheng/Matt W. |
Roger/Zifan |
|
| 9/30 |
Adversarial training
Reading: Towards Deep Learning Models Resistant to Adversarial Attacks
See also: Ensemble Adversarial Training: Attacks and Defenses |
Nikhil/Scott |
Grishma/Lokit |
|
Applied Cryptography |
|
|
|
|
| 10/2 |
Overview and basic constructions
Reading: Boneh and Shoup, 11.6, 19.4
See also: Evans, Kolesnikov, and Rosulek, Chapter 3 |
Justin |
--- |
|
| 10/5 |
Secure data collection at scale
Reading: Prio: Private, Robust, and Scalable Computation of Aggregate Statistics |
Saniya/Marcus |
Jinwoo/Mazharul |
|
| 10/7 |
Verifiable computing
Reading: SafetyNets: Verifiable Execution of Deep Neural Networks on an Untrusted Cloud |
Mike |
Siyang/Dan |
|
| 10/9 |
Side channels and implementation issues
Reading: On Significance of the Least Significant Bits For Differential Privacy |
Siyang/Dan |
Nathan/Matt T. |
|
| 10/12 |
Model watermarking
Reading: Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooring
See also: Protecting Intellectual Property of Deep Neural Networks with Watermarking |
Amos/Suleman |
Sidharth/Martin |
MS1 Due |
Algorithmic Fairness |
|
|
|
|
| 10/14 |
Overview and basic notions
Reading: Barocas, Hardt, and Narayanan, Chapter 1-2
See also: 50 Years of Test (Un)fairness: Lessons for Machine Learning |
Justin |
--- |
|
| 10/16 |
Individual and group fairness
Reading: Fairness through Awarness
Reading: Equality of Opportunity in Supervised Learning |
Sidharth/Martin |
Vishal/Nikita |
|
| 10/19 |
Inherent tradeoffs
Reading: Inherent Trade-Offs in the Fair Determination of Risk Scores |
Shiyu/Rita |
Rishabh/Aaron |
|
| 10/21 |
Fairness and causality
Reading: Barocas, Hardt, and Narayanan, Chapter 4 |
Justin |
--- |
|
| 10/23 |
Fairness in unsupervised learning
Reading: Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
See also: Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints |
Keaton/Anna |
Shiyu/Rita |
|
| 10/26 |
Testing fairness, empirically
Reading: Automated Experiments on Ad Privacy Settings: A Tale of Opacity, Choice, and Discrimination
Reading: Discrimination through optimization: How Facebook’s ad delivery can lead to skewed outcomes
See also: Barocas, Hardt, and Narayanan, Chapter 5 |
Rishabh/Aaron |
Mike |
|
PL and Verification |
|
|
|
|
| 10/28 |
Overview and basic notions |
Justin |
--- |
|
| 10/30 |
Probabilistic programming languages
Reading: Probabilistic Programming |
Vishal/Nikita |
Zijian/Yuchen |
|
| 11/2 |
Verifying probabilistic programs
Reading: A Program Logic for Union Bounds
See also: Advances and Challenges of Probabilistic Model Checking |
Jinwoo/Mazharul |
Yucheng/Matt W. |
|
| 11/4 |
Languages for differential privacy
Reading: Distance Makes the Types Grow Stronger: A Calculus for Differential Privacy
See also: Programming Language Techniques for Differential Privacy |
Ashish/Athena |
Nikhil/Scott |
|
| 11/6 |
Verifying neural networks
Reading: AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation
See also: DL2: Training and Querying Neural Networks with Logic |
Roger/Zifan |
Ashish/Athena |
MS2 Due |
No Lectures: Work on Projects |
|
|
|
|
| 12/7 |
Project Presentations |
|
|
|
| 12/9 |
Project Presentations |
|
|
|
| 12/11 |
PROJECTS DUE |
|
|
Projects Due |