[Defense] ML for Generalized Hyperarousal Prediction
Monday, April 18, 2022
11:00 am - 12:30 pm
will defend her proposal
ML for Generalized Hyperarousal Prediction
Hyperarousal manifests across the spectrum of human activities when subjects are challenged. Hyperarousal is often associated with stress and represents a precarious state where subjects commit operational errors or misbehave. For this reason, the study of hyperarousal is of immense importance in human-machine interactions and human-human interactions. With the advent of ubiquitous unobtrusive physiological sensing, continuous measurement of hyperarousal in the lab and in the wild became feasible. In this work, we investigate the question if there are underlying universal features of hyperarousal across the problem space that would allow a unifying treatment. We define the problem space along two dimensions: the type of subject activity (cognitive and dexterous) vs. the realism of the activity (controlled and naturalistic). We use datasets from major studies in each of the four cells that arise from this problem decomposition. We experiment with ML training configurations and feature architectures. Our ultimate goal is to identify the minimum type of training that leads to maximum universal performance for hyperarousal detection. Our effort comes to fill a gap in the literature, which has been treating hyperarousal in the context of specific study paradigms, rather than generally.
11:00AM - 12:30PM CT
Online via MS Teams
Dr. Ioannis Pavlidis, dissertation advisor
Faculty, students and the general public are invited.