The Frontiers of Machine Learning in Health and Behavioral Science

Abstract

Abstract: Models and algorithms developed within the machine learning community are seeing increasingly widespread use in areas like computational social science, computational health science, and computational ecology. The interaction between machine learning and these emerging areas has begun to reveal new categories of problems that push the boundaries of existing machine learning techniques. This talk will survey some of the problems at the frontiers of machine learning that my research group is currently focusing on. These problems are motivated by a variety of real-world problems including the analysis of physiological time series data from ICU electronic health records, action recognition and prediction from wireless on- body biosensors, and personalization of health communications.

Bio: Benjamin Marlin joined the University of Massachusetts Amherst as an assistant professor of Computer Science in fall 2011 where he co- founded and co-directs the Laboratory for Machine Learning and Data Science. He is also member of the UMass Amherst Center for Intelligent Information Retrieval. He was previously a fellow of both the Pacific Institute for the Mathematical Sciences and the Killam Trusts at the University of British Columbia where he was based in the Laboratory for Computational Intelligence. He completed his PhD in machine learning at the University of Toronto.