Data, Decisions, and Inclusive Social Impact
|
Speaker (s):
Bryan WILDER
PhD Candidate,
University of Southern California
|
|
Date:
Time:
Venue:
|
|
March 14, 2019, Thursday
11:00am - 12:00pm
Meeting Room 5.1, Level 5
School of Information Systems
Singapore Management University
80 Stamford Road
Singapore 178902
|
|
ABSTRACT
Enormous datasets and computational resources have combined to produce highly effective machine learning models for a range of domains. However, the data revolution leaves behind marginalized communities which lack access to technical and financial resources. Including such groups in the benefits of AI progress requires techniques which leverage costly and low-signal data for maximum impact. This talk presents methods at the intersection of optimization and machine learning which enable us to make decisions with limited data, strategically gather additional data when needed, and use this data to improve decisions by integrating combinatorial optimization problems into the training of machine learning models. I will illustrate these techniques through applications to two socially critical domains: HIV prevention for homeless youth and tuberculosis treatment in India. Both projects are undertaken in collaboration with community and governmental partners. Algorithms for the HIV prevention domain have been deployed in field tests, with initial results showing substantial real-world improvements over traditional approaches.
About the Speaker
Bryan WILDER is a PhD student in computer science at the University of Southern California, where he is advised by Milind Tambe. His research focuses on combinatorial optimization and machine learning, driven by applications to interventions for underserved or marginalized communities. He is supported by a NSF Graduate Research Fellowship, and his work has been recognized with best paper and best student paper nominations at AAMAS.
