ECE Seminar: David Carlson

ECE Seminar: David CarlsonPostdoctoral Research ScientistDuke UniversityLight refreshments will be available at 2:45 pm outside of PHO 339. Interpretable Machine Learning to Deconstruct the Neural Basis of Psychiatric DisordersAbstract:There is an extensive literature in machine learning demonstrating extraordinary ability to predict labels based on an abundance of data, such as object and voice recognition. Multiple scientific domains are poised to experience a data revolution, in which the quantity and quality of data will increase dramatically over the next several years. One such area is neuroscience, where novel devices will collect data orders of magnitude larger than current measurement technologies. In addition to being a “big data” problem, this data is incredibly complex. Machine learning approaches can adapt to this complexity to give state-of-the-art predictions. In addition to predictive performance, for many neurological disorders we are most interested in methods that are also interpretable such that they can be used to design interventions. Towards this end, I will discuss my work using machine learning to analyze local field potentials recorded from electrodes implanted at many sites of the brain concurrently. The techniques I developed learn features that are predictive, are interpretable, and generate data-driven hypotheses. Specifically, I first use ideas from dimensionality reduction and factor analysis to map the collected high-dimensional signals to a low-dimensional feature space. Each feature is designed as a Gaussian Process with a novel kernel to capture whole-brain spectral power and phase coherence, which have known neural correlates. These features are interpretable and estimate directionality of information flow. By correlating these features with outcomes, we determine which features or brain networks are associated with behavior outcomes. We can then generate a data-driven hypothesis about how the networks should be modulated. Collaborators have developed optogenetic techniques to test these theories in a mouse model of depression, validating the machine learning approach. I will also discuss current efforts to incorporate additional information sources and apply these ideas to other measurement techniques. Bio: David Carlson is currently a Postdoctoral Research Scientist at Duke University in the Department of Electrical and Computer Engineering and the Department of Psychiatry and Behavioral Studies within the Laboratory for Psychiatric Neuroengineering. From August 2015 to July 2016, he completed postdoctoral training in the Data Science Institute and the Department of Statistics at Columbia University focused on neural data science. He received his Ph.D., M.S., and B.S.E. in Electrical and Computer Engineering from Duke University in 2015, 2014, and 2010, respectively. He received the Charles R. Vail Memorial Outstanding Scholarship Award in 2013 and the Charles R. Vail Memorial Outstanding Graduate Teaching Award in 2014.

When 3:00 pm on 2/6/2017 to 4:00 pm on 3/6/2017
Location Photonics Center, 8 Saint Mary's Street, Room 339