Machine Learning and Friends Lunch (Online)

Title: "Building dialog systems with fewer data and less supervision"

Abstract: Neural network-based models have shown good performance on various NLP tasks. However, they require a lot of data to train. Thus, how to train dialog models with fewer data and less supervision is a practical problem. This talk will introduce some work that utilizes meta-learning, semi-supervised learning, and zero-shot learning algorithms for task-oriented dialog system training.

Bio: Zhou Yu joined the CS department at Columbia University in Jan 2021 as an Assistant Professor. Before that, she was an Assistant Professor at UC Davis. She obtained her Ph.D. from Carnegie Mellon University in 2017. Zhou has built various dialog systems that have a real impact, such as a job interview training system, a depression screening system, and a second language learning system. Her research interest includes dialog systems, language understanding and generation, vision and language, human-computer interaction, and social robots. Zhou received an ACL 2019 best paper nomination, featured in Forbes 2018 30 under 30 in Science, and won the 2018 Amazon Alexa Prize. 

To obtain the Zoom link for this event, please see the event announcements from MLFL on the college email lists or contact Kalpesh Krishna.