Using Machine Learning to Study How Brains Process Natural Language

Abstract – How does the human brain use neural activity to create and represent meanings of words, phrases, sentences and stories? One way to study this question is to give people text to read, while recording their brain activity with fMRI (1 mm spatial resolution) and MEG (1 msec time resolution). By doing this, we have learned intriguing answers to questions such as “Are the neural encodings of word meaning the same in your brain and mine?”, “What sequence of neurally encoded information flows through the brain during the half-second in which the brain comprehends a word?,” “How are meanings of multiple words combined when reading sentences, and stories?,” and “How does our understanding of the brain align with current AI approaches to natural language processing?” This talk will summarize our machine learning approaches, some of what we have learned, and newer questions we are currently studying.

Speaker biography – Tom M. Mitchell is the Founders University Professor at Carnegie Mellon University, where he founded the world’s first Machine Learning Department. His research interests include machine learning, artificial intelligence, cognitive neuroscience, and the impact of AI on society. He has testified to the U.S. Congressional Research Service, and the U.S. House Subcommittee on Veterans’ Affairs regarding potential uses and impacts of artificial intelligence, and is currently co-chairing a U.S. National Academies study on AI and the future of work. Mitchell is an elected member of the U.S. National Academy of Engineering, and the American Academy of Arts and Sciences, and a Fellow and Past President of the Association for the Advancement of Artificial Intelligence (AAAI). Mitchell received an honorary degree from Dalhousie University for his research in machine learning and cognitive neuroscience. Website: www.cs.cmu.edu/~tom