I will trace my career in translational research bringing modern data analytic techniques to various different discipline areas. The focus will be on dispelling various myths about the capabilities of so-called "Artificial Intelligence" and "Machine Learning". In particular, I will highlight the importance of understanding what modern analytic techniques can and cannot do, and how we must maintain our own agency when attempting to assess the value, validity and correctness of machine-based output. Emphasis will be on graphical diagnostics to ensure proper understanding.
Biography:
Steven Stern is currently Professor of Data Science at Bond University's Centre for Data Analytics. Previously, he was ABS Chair of Statistics at Queensland University of Technology from 2014 to 2016 and prior to that was on the Faculty of the Australian National University for two decades. He received his Bachelors Degree in Mathematics in 1986, his Masters degree in Applied Statistics in 1987 and his PhD in Mathematical Statistics in 1994 all from Stanford University. He is an award-winning teacher, receiving the Vice Chancellor's Award for Teaching Excellence from ANU in 2002 and has been recognized by the Australian Teaching and Learning Council for his work in statistical education. In addition, he is a highly sought-after data analyst and consultant, having undertaken work for numerous government departments and industry bodies. His research interests cover the full range of data analytics and statistical methodology, but he is probably best known for his work in sports statistics and resampling-based statistical learning techniques.