Deanna Needell, University of California at Los Angeles

Watch Needell talk on YouTube here.

Towards Fairer-ness in Machine Learning

In this talk, we will address several areas of recent work centered around the themes of transparency and fairness in machine learning as well as highlight the challenges in this area. We will discuss recent results involving linear algebraic tools for learning, such as methods in non-negative matrix factorization that include tailored approaches for fairness. We will showcase our derived theoretical guarantees as well as practical applications of those approaches. These methods allow for natural transparency and human interpretability while still offering strong performance. Then, we will discuss new challenges and directions in fairness including an example in large-scale optimization that allows for population subgroups to have better predictors than when treated within the population as a whole.  Throughout the talk, we will include example applications from collaborations with community partners, using machine learning to help organizations with fairness and justice goals.

Citation: Needell earned a B.S. in mathematics from the University of Nevada, Reno, and a M.A. and PhD in mathematics from University of California, Davis. After serving as a Postdoctoral Fellow at Stanford University and a faculty position at Claremont McKenna College, Needell is currently at the University of California, Los Angeles where she serves as Professor of Mathematics (since 2017), Dunn Family Endowed Chair in Data Theory (since 2020), and Executive Director of the Institute for Digital Research and Education (as of 2021). Widely recognized for her contributions and service to mathematics, Needell has earned many awards including the Alfred P. Sloan fellowship, an NSF CAREER award, the 2016 IMA Prize in Mathematics and its Applications, and was elected an American Mathematical Society Fellow in 2022. As of Fall 2023, she was the Eisenbud Professor at SLMath (formerly MSRI), where she participated in the Semester Program on Algorithms, Fairness, and Equity. 

 Needell is an exceptional researcher. She is a leading expert making fundamental research contributions both to theoretical problems in data science and to their applications, including to important societal issues. In particular, she has been recognized for contributions to sparse approximation, signal processing, and stochastic optimization. It is also noteworthy that she has more than 80 publications in research journals, more than 70 publications in conference proceedings, and more than 20 papers currently under review. She has presented numerous invited lectures, conference presentations, and seminar talks. She was keynote speaker at several prominent conferences, including speaking at the Fields Institute and the SIAM Conference on Applied Linear Algebra.

Needell has mentored and advised many mathematicians, including 10 PhD students and 11 postdocs. Since 2017, Needell’s AWM involvement includes being a Women in Math and Data Science (WISDM) organizer and leader. Her efforts in the community include involvement with the IChicas program for Latina youth, San Diego Refugee Tutoring (for refugee K–12 students), and many others. She is an active member of the professional mathematics community, serving on numerous boards and committees for the AMS, SIAM, IPAM, and IEEE. Through her active involvement in national committees, work on societal issues and mentoring of students of various backgrounds, Professor Needell is dedicated to broadening access to mathematics. Her ongoing excellence in mathematics research, teaching, and outreach underscores her selection as the distinguished 2024 Etta Z. Falconer Lecturer.