Interpretability Practitioners

<p><a href="http://www.rayhong.net/">Sungsoo Ray Hong</a> joins us to discuss the paper <a href= "https://arxiv.org/abs/2004.11440">Human Factors in Model Interpretability: Industry Practices, Challenges, and Needs</a>.</p>

Guest

Sungsoo Ray Hong: Ray Hong will start his Assistant Professorship at the Department of Information Sciences and Technology at George Mason University in Summer 2020. He received his Ph.D. in the Department of Human-Centered Design & Engineering at the University of Washington in 2018, then worked as a Visualization for Machine Learning Fellow in the CSE at NYU. His mission is to devise practical techniques and tools that help people to better leverage Machine Learning models in drawing insights and decisions using data. His current research projects are related to human-in-the-loop, a way to leverage humans to improve models, and human-AI collaboration, a way to apply models to improve human performance of data-centric tasks, using theoretical constructs in Human-Computer Interaction (HCI), Information Visualization (InfoVis), and Computer-Supported Cooperative Work (CSCW). So far, he has published 20 papers in top-tier conferences closely related to his research, including ACM SIGCHI (7 papers) and CSCW (3 papers). One of his recent CSCW papers discuss industry practice, challenges, and needs specifically related to model interpretability: https://github.com/rayhong/model_interpretability He has 6 years of industry experience at Samsung Research where he worked on commercializing new digital products adopted in Samsung’s millions of mobile and home devices.