Responsible AI: Privacy, Fairness, and Robustness Seminar

Course Description

This seminar-style course delves into the ethical dimensions of Artificial Intelligence (AI), with a particular focus on the intersectionality of privacy, fairness, and robustness. The course is structured around reading, discussing, and critically analyzing seminal and state-of-the-art papers in the field. Participants will engage in intellectual discourse to understand the challenges, methodologies, and emerging trends related to responsible AI. The course is designed for graduate students with good ML, stats, and optimization background.

Course Objectives

Prerequisites

Syllabus

This is a tentative calendar and it is subject to change.

Date Topic Subtopic Papers Presenting
Wed Jan 17 Intro to class   class slides Fioretto
Mon Jan 22 Intro to class Safety and Alignment class slides Fioretto
Wed Jan 24 Intro to class Privacy (settings and attacks) class slides Fioretto
Mon Jan 29 Intro to class Privacy (cont) class slides Fioretto
Wed Jan 31 Intro to class Privacy and Fairness class slides Fioretto
Mon Feb 5 Fairness Intro and bias sources [1] – [4] Group 1
Wed Feb 7 Fairness Statistical measures [5] – [8] Group 2
Mon Feb 12 Fairness Tradeoffs [9] – [12] Group 3
Wed Feb 14 Fairness LLMs: Toxicy and Bias [13] – [16] Group 4
Mon Feb 19 Fairness LLMs: Fairness [17] – [19] Group 5
Wed Feb 21 Fairness Policy aspects [20] – [22] Group 6
Mon Feb 26 No class (AAAI)      
Wed Feb 28 Safety Distribution shift [23] – [25] Group 1
Mon Mar 4 Spring break      
Wed Mar 6 Spring break      
Mon Mar 11 Safety Poisoning [26] – [29] Group 2
Wed Mar 13 Safety Adversarial Robustness [30] – [34] Group 3
Mon Mar 18 Safety Adversarial Robustness [35] – [39] Group 4
Wed Mar 20 Safety LLMs: Prompt injection [40] – [45] Group 5
Mon Mar 25 Safety LLMs: Jailbreaking [46] – [50] Group 6
Wed Mar 27 Privacy Differential Privacy 1   Group 1
Mon Apr 1 Privacy Differential Privacy 2   Group 2
Wed Apr 3 Privacy Auditing and Membership inference   Group 3
Mon Apr 8 Privacy Privacy and Fairness   Group 4
Wed Apr 10 Privacy LLMs: Private issues in LLMs   Group 5
Mon Apr 15 Privacy LLMs: Privacy in LLMs   Group 6
Wed Apr 17 Evaluation Model cards   Group 1
Mon Apr 22 Evaluation LLMs: evaluation   Group 2
Wed Apr 24 Unlearning Unlearning 1   Group 3
Mon Apr 29 Unlearning LLMs: Targeted unlearning   Group 4


Bibliography

Assessment

Each group will be assessed through the following activities:

1. Paper Summaries (Blogging) – 33.3%

Objective: To develop the ability to critically analyze and summarize AI research papers in a clear and accessible manner.

Expectations:

Assessment Criteria:

2. Presentation – 33.3%

Objective: To enhance students’ ability to communicate complex AI concepts and engage in public speaking.

Expectations:

Assessment Criteria:

3. Discussion Lead – 33.3%

Objective: To cultivate skills in leading intellectual discourse and fostering collaborative learning.

Expectations:

Assessment Criteria:

General Notes:

Groups

Group Members
Group 1 Lei Gong, Archit Uniyal, Luke Benham, Chien-Chen Huang, Stuart Paine
Group 2 Saswat Das, Wenqian Ye, Benny Bigler-Wang, Parker Hutchinson, Linyun Wei, Zhiyang Yuan
Group 3 Nibir Mandal, Guangzhi Xiong, Neh Joshi, Sree Esshaan Mahajan, Esshaan Mahajan
Group 4 Sarvin Motamen, Parth Kandharkar, Ellery Yu, Hongyan Wu, Kefan Song,
Group 5 Mati Ur Rehman, Jeffrey Chen, Candace Chen, Kaylee Liu, Robert Bao
Group 6 Stephanie Schoch, Aidan Hesselroth, Joseph Moretto, Jonathan McGee, ShiHe Wang

Instructor

Ferdinando Fioretto Assistant Professor in Computer Science University of Virgina


This syllabus is subject to changes to meet the learning needs of the course participants.