Fall 2023: COMP 640: Graduate Research Seminar in Machine Learning

Class Timings and location      Wednesday 03:00 PM - 03:50 PM, Duncan Hall 1042.

Course Description:

Our theme is Trustworthy AI and beyond. Despite the remarkable achievement of artificial intelligence in a variety of fields, designing responsible and accountable (trustworthy) ML systems remains a significant problem due to both the practical requirements of the stakeholders as well as the regulations in different domains. Trustworthy AI requires governance and regulatory compliance throughout the AI lifecycle from ideation to design, development and deployment. From this perspective, the ML/DL systems are evaluated in these four dimensions: explainability (transparency), fairness (unbias), robustness (reliability), and respectfulness of privacy (security).

Through this course, we will select some solid work on the trustworthy ML/DL and discuss them.

Learning Outcomes:

The goal of this course is deriving a comprehensive understanding of Trustworthy AI, state-of-the-art techniques, and its potentials in daily life. Moreover, we hope this course could also:

  • give you a stage to brainstorm, explore and express your thoughts about trustworthy-AI.
  • broaden your vision via exchanging your throughts with other students.
  • provide you an opportunity to have collaboration with other people.

Grading and Logistics:

  • For students signning up for 1 credit: Read at least one paper on trustworthy-AI; Give a 30-min presentation on the paper, and lead the discussion during the class.

Prerequisite:

A background of machine learning is required. Each student will be invited to give a talk about advanced papers in trustworthy ML/DL. We will have one talk every week.

Presentations and Summarization Logistics:

  • Each student should read at least one paper on trustworthy-AI (interpretability, fairness, robustness or pravicy of ML).
  • Ten attendences will be randomly taken in the semester and are used to measure the attendance as well. six attendences are required for full score. As long as more than six attendences are received successfully, no extra evidence is needed. Otherwise an excused absence is required.

    • Schedule:

    TBD.

    Instructor Information:

    Name Xia "Ben" Hu
    Email address xia.hu@rice.edu
    Office hours Per request
    Office location Zoom

    Americans with Disabilities Act (ADA):

    The Americans with Disabilities Act (ADA) is a federal anti-discrimination statute that provides comprehensive civil rights protection for persons with disabilities. Among other things, this legislation requires that all students with disabilities be guaranteed a learning environment that provides for reasonable accommodation of their disabilities. If you believe you have a disability requiring an accommodation, please contact Rice’s Disability Resource Center. For additional information, visit https://policy.rice.edu/402

    Academic Integrity:

    "On my honor, I have neither given nor received any unauthorized aid on this (exam, quiz, paper)."

    Upon accepting admission to Rice University, a student immediately assumes a commitment to uphold the Honor Code, to accept responsibility for learning, and to follow the philosophy and rules of the Honor System. Students will be required to state their commitment on examinations, research papers, and other academic work. Ignorance of the rules does not exclude any member of the Rice community from the requirements or the processes of the Honor System. For additional information please visit: https://oaa.rice.edu/policies-and-procedures/honor-code

    Acknowledgement:

    This page has been partially adopted from the previous editions of COMP 640 Spring 2022, instructed by Professor Anshumali Shrivastava of Rice University.