Marie-Francine Moens: Can We Build a Machine That Has a Commonsense Memory?
Experiences of translating language to visuals when evaluating machine understanding of human language tell us that comprehensive language understanding needs a large amount of commonsense knowledge. In this talk we go deeper into how to acquire this commonsense knowledge, in what format to represent it, how to store and retrieve it, and how to integrate it in language understanding tasks.
Marie-Francine Moens is a full professor at the Department of Computer Science, KU Leuven. She is author and (co-)editor of several books, and (co-)author of more than 400 peer-reviewed publications in the fields of artificial intelligence, information retrieval, machine learning, and natural language and multimedia processing. She holds the ERC Advanced Grant Commonsense and Anticipation enriched Learning of Continuous representations sUpporting Language UnderStanding (CALCULUS, 2018–2024) granted by the European Research Council. Since 2017 she is associate editor of the journal IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). She is a fellow of the European Laboratory for Learning and Intelligent Systems (ELLIS), which is a pan-European AI network of excellence.