Presentation: Networks from Time Series: Deciphering the Brain Connectome
Speaker: Ali Shojaie, Ph.D., Associate Professor of Biostatistics, University of Washington
Abstract: High-dimensional time series have become ubiquitous in many scientific fields. They are particularly instrumental in monitoring the activities of neurons and/or brain regions, and have led to valuable insight into cognitive processes and neurodegenerative diseases.
I will present new methodological, computational and theoretical developments for learning networks from high-dimensional time series. Motivated by emerging calcium florescent imaging data that record neuronal activity, the talk will primarily focus on learning neuronal connectivity networks from high-dimensional spike train data. Time permitting, I will also discuss a new approach for analyzing non-stationary electroencephalography (EEG) data, and detecting changes in the brain connectome associated with seizure.