Electrical Engineering Seminars

Towards Fair and Generalizable Machine Learning with Large Models

Samet Oymak, Assistant Professor, University of California, Riverside

Mar 4, 2022
Recent years have witnessed major advances in data science, driven by increasingly large neural networks. While these models hold great promise, modern machine learning applications motivate fine-grained performance control, spanning fairness, robustness, and accuracy. This talk will introduce new strategies to ensure these models behave the way we want.
I will first discuss optimizing fairness objectives on datasets with imbalanced or sensitive groups. We observe that a large model can achieve seemingly perfect fairness on training data but dramatically fail at the test-time. We show that this is due to overfitting as well as ineffectiveness of the traditional approaches. To address this, we propose a new family of fairness-seeking loss functions that better guide the training process and use it to achieve state-of-the-art performance.
As a key ingredient of generalizable learning, I will motivate bilevel problem formulations which help avoid overfitting, select the right model, and overcome distribution shift between training and validation data. Finally, I will introduce new bilevel optimization methods for federated learning across decentralized clients. These methods achieve provable communication and computation efficiency, overcome client heterogeneity, and gracefully specialize to min-max optimization.

I will conclude the talk with a discussion of future research on the theoretical foundations of deep learning, resource-efficient ML architectures, and data-driven control problems.
Speaker Bio

Samet Oymak is an assistant professor of Electrical and Computer Engineering at the University of California, Riverside since 2018. Prior to UCR, he spent three years at Google and in algorithmic finance. During his postdoc, he was at UC Berkeley as a Simons Fellow and a member of AMPLab. He obtained his bachelor's degree from Bilkent University in 2009 and his PhD degree from Caltech in 2015. He received the Charles Wilts Prize for the best departmental PhD thesis at Caltech in 2015 and the NSF CAREER award in 2021. Website:


Na Li


Jessica Brenn