This is an online seminar. Registration is required.
【 Imperfect Information Learning Team】
【Date】2024/January/30(Tue) 17:00-18:00(JST)
【Speaker】Chen Feng, Queen Mary University of London, School of Electronic Engineering and Computer Science
Title: Towards robust and efficient weakly supervised learning
Abstract:
In this seminar, we delve into the realm of weakly supervised learning, focusing on the challenges and innovations required for effective and robust training in computer vision tasks. Specifically, our research addresses the complexities associated with learning from datasets where annotations or labels exhibit various degrees of noise, coarseness, or absence. We scrutinize three distinct challenges within the domain of weakly supervised learning: learning with noisy labels, learning from coarse labels, and self-supervised learning in the absence of any labels.
