Responsible AI blog - Fall-24

This repository is intedned to contain a collection of notes and paper summaries for each class of the Reponsible AI course tought at UVA in the Fall-24. The course is organized around six topics:

• Fairness
• Safety
• Privacy
• Evaluation
• Unlearning
• Misuse of AI and Governance

Each topic is associated with the corresponding folder in this repository.

Students should should write the report on the papers and topics reviewed in their class by modifying the associated ".md" file.

Syllabus

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

Date	Topic	Subtopic	Blog
Feb 5	Fairness	Intro and bias sources	Group 1
Feb 7	Fairness	Statistical measures	Group 2
Feb 12	Fairness	Tradeoffs	Group 3
Feb 14	Fairness	LLMs: Toxicy and Bias	Group 4
Feb 19	Fairness	LLMs: Fairness	Group 5
Feb 21	Fairness	Policy aspects	Group 6
Feb 26	No class (AAAI)
Feb 28	Safety	Distribution shift	Group 1
Mar 6	Spring break
Mar 11	Spring break
Mar 11	Safety	Poisoning	Group 2
Mar 13	Safety	Adversarial Robustness	Group 3
Mar 18	Safety	Adversarial Robustness	Group 4
Mar 20	Safety	LLMs: Prompt injection	Group 5
Mar 25	Safety	LLMs: Jailbreaking	Group 6
Mar 27	Privacy	Differential Privacy 1	Group 1
Apr 1	Privacy	Differential Privacy 2	Group 2
Apr 3	Privacy	Auditing and Membership inference	Group 3
Apr 8	Privacy	Privacy and Fairness	Group 4
Apr 10	Privacy	LLMs: Private issues in LLMs	Group 5
Apr 15	Privacy	LLMs: Privacy in LLMs	Group 6
Apr 17	Evaluation	Model cards	Group 1
Apr 22	Evaluation	LLMs: evaluation	Group 2
Apr 24	Unlearning	Unlearning	Group 3
Apr 29	Unlearning	LLMs: Targeted unlearning	Group 4

Expectations:

• Each group will reivew all paper from the provided list, and they may propose additional ones for approval.
• Summaries should be written in Markdown format (supporting images and formulas) and committed to the course's GitHub repository.
• The summary should include the following sections: Introduction and Motivations, Methods, Key Findings, and Critical Analysis.
• The Critical Analysis section should evaluate the strengths, weaknesses, potential biases, and ethical considerations of the paper.
• Summaries must be submitted four days prior to the presentation for review and potential feedback.

Assessment Criteria:

• Clarity and coherence of the written summary.
• Depth of critical analysis and understanding of the paper's content.
• Proper use of formatting and adherence to submission guidelines.
• Timeliness of submission.