CUIDS Distinguished Speaker Series 2022-23
Light refreshments will be provided!
Explainable AI for Classification and Decision Systems (In-Person)
Leopoldo Bertossi, Skema Business School, Canada, Montreal
Mar. 1, 2023, 1:00 p.m. – 2:00 p.m.
Description: Explainable AI (XAI) has become an effervescent area of research in AI and Machine Learning (ML). It is interesting per se, but also touches several aspects of what is nowadays called Ethical AI. In this presentation we will introduce some key concepts and ideas related to score-based explanation methods in ML-systems for classification and decision making. We will also emphasize: (a) the need for reasoning with explanations, (b) the connection with other more traditional explanation methods in AI; and (c) the fact that XAI is at the very core of AI, as opposed to being a kind of meta-analysis of AI methods.
Biography: Leopoldo Bertossi is a Professor at the Skema Business School -a French private business school- and its R&D Lab for Business AI, Montreal, Canada (since July 2022).
Since 2019, he is a Professor Emeritus of the School of Computer Science, Carleton University (Ottawa, Canada), with Grad Supervision Status. Until August 2022 he was a Full Professor at the Faculty of Engineering and Sciences, “Universidad Adolfo Ibanez” (UAI, Santiago, Chile), where he was the Director of the PhD and MSc. Programs in Data Science. He is a Senior Fellow of the UAI. He is also a Senior Researcher at the “Millennium Institute for Foundational Research on Data” (IMFD, Chile).
His broad research interests are related to Data Science and Artificial Intelligence, with focus on explainable AI, causality, knowledge representation, data management, computational logic, ontologies, uncertainty management, and machine learning.
Using Data from Various Sources to Facilitate Learning and Inference (In-Person)
Dr. Junfeng Wen
Oct. 19, 2022, 10:30 a.m. – 11:30 a.m.
Description: Nowadays, we have access to millions of data examples from various sources. However, not all of them are relevant to the problem we are trying to solve at hand. In order to train an effective machine learning model, one needs to select the most useful data to conduct learning and inference, which is Junfeng’s research focus. In this talk, Junfeng will first provide an overview of his research in transfer learning, reinforcement learning and federated learning. Then he will showcase a few developed algorithms, including multi-source domain adaptation and personalized federated learning.
Biography: Junfeng Wen is an Assistant Professor in the School of Computer Science at Carleton University. He obtained his Ph.D. in Statistical Machine Learning from the Department of Computing Science at the University of Alberta. His research focuses on how to leverage data collected from various sources to build effective models for a different but related target domain in the fields of machine learning (ML) and reinforcement learning (RL).