- This event has passed.
IDEAS/STAT Optimization Seminar: “Statistics-Powered ML: Building Trust and Robustness in Black-Box Predictions”
March 20 at 12:00 PM - 1:15 PM
Zoom link:
https://upenn.zoom.us/j/98220304722
Abstract:
Modern ML models produce valuable predictions across various applications, influencing people’s lives, opportunities, and scientific advancements. However, these systems can fail in unexpected ways, generating unreliable inferences and perpetuating biases present in the data. These issues are particularly troubling in high-stakes applications, where models are trained on increasingly diverse, incomplete, and noisy data and then deployed in dynamic environments—conditions that often exacerbate test-time failures.
In response to these challenges, this talk explores a key question: How can fundamental statistical principles be harnessed to produce trustworthy predictive inference?
In the first part, I will present a new advancement in conformal prediction—a statistical wrapper for any black-box model that provides precise error bounds on ML predictions. I will focus on scenarios where training data is corrupted or biased, such as through missing features and labels, and introduce a framework for constructing predictive uncertainty estimates that remain valid despite distribution shifts between the available corrupted data and unknown clean data.
In the second part, I will show how sequential statistical testing can enable a novel test-time training scheme, allowing a pre-trained model to adapt online to unfamiliar environments. For instance, consider an image classification task where test images are captured under varying illumination conditions that differ from the training setup. Building on conformal betting martingales, I will first introduce a monitoring tool to detect data drifts. Using this tool, I will derive a rigorous ‘anti-drift correction’ mechanism grounded in (online) optimal transport principles. This mechanism forms the foundation of a self-training scheme that produces robust predictions invariant to dynamically changing environments.
Yaniv Romano
Assistant professor in the Departments of Electrical Engineering and Computer Science, Technion
Yaniv Romano is an assistant professor in the Departments of Electrical Engineering and Computer Science at the Technion. Previously, he was a postdoctoral scholar in the Department of Statistics at Stanford University, advised by Prof. Emmanuel Candès. Yaniv holds a PhD, MSc, and BSc in Electrical Engineering, all from the Technion. The super-resolution technology he invented with Peyman Milanfar has been integrated into Google’s flagship products. His uncertainty quantification technique, developed with Emmanuel Candès, was employed by The Washington Post to estimate outstanding votes during the U.S. presidential election.
Yaniv has received several honors and awards, including the ERC Starting Grant, the Alon Scholarship, the SIAG/IS Early Career Prize, the Sheila Samson Prime Minister’s Prize (Researcher Recruitment Prize), the IEEE Signal Processing Society Best Paper Award, the Krill Prize for Excellence in Scientific Research, and the Henry Taub Prize for Academic Excellence.