CHAI Seminar and Social: Center for Human-Aware Artificial Intelligence

Center for Human-Aware Artificial Intelligence www.rit.edu/chai CHAI Seminar and Social Monday, November 25, 12 noon to 2pm Seminar start at 12:30pm Bamboo Room: Campus Center 2610   Active Learning for Many-class and Multi-label Problems in Knowledge-rich Domains   Dr. Qi Yu, Associate Professor, RIT School of Information and Mr. Weishi Shi, Ph.D. Candidate, GCCIS   Time: Monday, November 25, 12 noon. Seminar start time: 12:30pm Location: Bamboo Room: Campus Center 2610 Abstract: While more labeled data tends to improve the performance of supervised learning, obtaining accurate labels may be highly challenging for many specialized domains, such as medicine and biology, where expert knowledge is required for understanding and extracting the underlying semantics of data. Active Learning (AL) provides a means to reduce human labeling efforts by identifying the most informative data samples. In this talk, we will start by providing a general overview of the major research topics being pursued by the Machine Learning and Data Intensive Computing (Mining) Lab at RIT, including active learning and other directions that support the analysis of large-scale complex data with limited human supervision. We then discuss some of our recent results in developing novel active learning models to achieve better learning from less labeled data for many-class and multi-label problems that can highly benefit knowledge-rich domains. Bios: Dr. Qi Yu is an Associate Professor in the School of Information at RIT. He directs the Machine Learning and Data Intensive Computing (Mining) Lab https://pht180.rit.edu/mining/  Qi’s research interest primarily lies in the areas of machine learning, data mining, and artificial intelligence and his work has been regularly published in major machine learning venues (e.g., ICML and NeurIPS). Weishi Shi is a fourth-year PhD student, studying in the PhD program of computing and information sciences at RIT. Weishi’s research interest is in machine learning and data mining, with a special focus on active learning.