Embracing Change: Tackling In-the-Wild Shifts in Machine Learning
Friday, March 10, 2023 11 AM
About this Event
Huaxiu Yao
Postdoctoral Scholar
Computer Science Department
Standford University
The real-world deployment of machine learning algorithms often poses challenges due to shifts in data distributions and tasks. These shifts can lead to a degradation in model performance, as the model may not have seen such changes during training. It can also make it difficult for the model to generalize to new scenarios and can lead to poor performance in real-world applications. In this talk, I will present our research on building machine learning models that are unbiased, widely generalizable, and easily adaptable to different shifts. Specifically, I will first discuss our approach to learning unbiased models through selective augmentation for scenarios with subpopulation shifts. Second, I will also delve into the utilization of domain relational information to enhance model generalizability for arbitrary domain shifts. Then, I will present our techniques for quickly adapting models to new tasks with limited labeled data. Additionally, I will show our success practices for addressing shifts in real-world applications, such as in the healthcare, e-commerce, and transportation industries. The talk will also cover the remaining challenges and future research directions in this area.
Event Details
See Who Is Interested
Dial-In Information
Join Zoom Meeting
https://wustl.zoom.us/j/99687348688?pwd=emE0WVNLWHk1S0dmZS9wenJBV1ZBZz09
About this Event
Huaxiu Yao
Postdoctoral Scholar
Computer Science Department
Standford University
The real-world deployment of machine learning algorithms often poses challenges due to shifts in data distributions and tasks. These shifts can lead to a degradation in model performance, as the model may not have seen such changes during training. It can also make it difficult for the model to generalize to new scenarios and can lead to poor performance in real-world applications. In this talk, I will present our research on building machine learning models that are unbiased, widely generalizable, and easily adaptable to different shifts. Specifically, I will first discuss our approach to learning unbiased models through selective augmentation for scenarios with subpopulation shifts. Second, I will also delve into the utilization of domain relational information to enhance model generalizability for arbitrary domain shifts. Then, I will present our techniques for quickly adapting models to new tasks with limited labeled data. Additionally, I will show our success practices for addressing shifts in real-world applications, such as in the healthcare, e-commerce, and transportation industries. The talk will also cover the remaining challenges and future research directions in this area.