Guest speaker: Dr. Alex Davis from Carnegie Mellon University (Pittsburgh, Pennsylvania, USA)
Date: Monday, October 24
Time: 10:30am to 12:00pm
Location changed to AERL 107/108.
Expert and machine prediction for risk assessment and scientific discovery
Abstract: From radars to radiology, experts have long been asked to detect signals against a noisy and uncertain background. Starting in the middle of the twentieth century, a number of direct comparisons between expert judges and statistical prediction tools found that experts were no better, and most often worse, than even the simplest statistical prediction tools. However, these studies limited experts and tools to the same information, preventing experts from using information in their environment that is not easily translated into statistics. Across a variety of domains, it is not known how much information these tools provide above and beyond assessments made by experts in the natural course of their work. In this talk I discuss ongoing work on predictions by experts about risk and the outcomes of scientific experiments. For the former I discuss a longitudinal cohort study comparing the separate and joint ability of physicians and an electronically integrated statistical prediction tool (the Rothman Index) to predict patient risk of decompensation (defined as a cardiac or respiratory arrest, call to a rapid response team, or transfer to the ICU). For the latter I discuss expert ability to predict optimal combinations of parameters for the development of experimental materials, including nanomaterials, 3D printing of silicone elastomers, and part selection for aerospace metals additive manufacturing.
Photo credit: Fe Ilya from flickr/Creative Commons
Bio: Alex Davis is an Assistant Professor in the Department of Engineering and Public Policy at Carnegie Mellon University. He is a member of the Behavior, Decision, and Policy Group, the Carnegie Electricity Industry Center (CEIC), and the Center for Climate and Energy Decision Making (CEDM). He is currently directing a two year field experiment in collaboration with the Cornell Cooperative Extension examining behavioral approaches to encourage residential energy efficiency, and he is CMU’s acting director of a multi-year, multi-institutional research project on the relationship between science and proven experience. His research focuses on the behavioral foundations of policy, applied to innovation and entrepreneurship, energy, the environment, health, and information and communication technologies. He teaches a graduate course in applied data analysis (19-704).
Alex earned his B.S. from Northern Arizona University in Psychology (2007) and his M.S. (2009) and Ph.D. (2012) from Carnegie Mellon University in Behavioral Decision Research. He worked as a postdoctoral fellow and research scientist at Carnegie Mellon University prior to joining the faculty at Carnegie Mellon.
