Policies & Projects

Prerequisites

• Required: Basic mathematical statistics (e.g., STOR 555 or equivalent)
• Strongly encouraged: Graduate probability (e.g., STOR 634)

Students without this background should be prepared to fill gaps independently.

Course format

• Regular reading outside class is expected.
• Over time, class sessions will emphasize discussion, synthesis, and student-driven exploration (not pure lecturing).
• Active participation is strongly encouraged.

Expectations (non-negotiable)

• Attend class regularly (email in advance to excuse absences).
• Participate in discussions.
• Work toward formulating a publication-worthy problem.
• Make progress toward a paper (or at least the beginnings of one).

You are expected to treat everyone with respect, uphold academic integrity, and fully participate in any group work.

Mid-semester proposal (end of February)

By the end of February, each student (or team of at most two) will:

• Propose a paper / set of papers / coherent research direction,
• Actively read and analyze the material,
• Give a short in-class presentation outlining their direction.

Final project (end of semester)

• Deliverable: written report.
• Stretch goal: beginnings of a conference/journal submission to a strong ML or applied probability venue.
• Projects may be individual or in teams of two.

Scope note

This course does not focus in depth on social/ethical/legal/security dimensions of generative AI, all topics worthy of their own course and super important in their own right as we increasingly see the impact of these technologies on our lives. These may arise tangentially, but the course is centered on statistical/probabilistic/modeling principles as I do not have the expertise to cover these topics.