Speaker: Dr. Yu Tian
From: University of Pennsylvania
Abstract
In today's world, deep neural networks drive machine learning systems that pervade every aspect of our daily lives. Yet, their deployment often raises concerns about trustworthiness, including safety, fairness, and other issues. This is particularly important and concerning in adapting AI systems in medicine. This talk will delve into the trustworthiness of machine learning algorithms, focusing on two main areas: 1) Developing machine learning systems that can detect anomalies or Out-of-Distribution (OOD) samples across various domains, from computer vision to medical imaging. We will highlight the unique challenges when adapting these systems from general computer vision applications to those specific to medicine; and 2) Enhancing fairness in algorithms to reduce biases across different demographic groups, including race, ethnicity, and gender. We will address the specific fairness challenges that arise within the field of medical imaging.
For more info, please follow this link.
Read More