Machine Learning Methods for Discrete Event Detection in Wearable Sensor Data Streams

09 Jan
Monday, 01/09/2017 9:30am to 11:30am
Computer Science Building, Room 142
Ph.D. Dissertation Proposal Defense
Speaker: Roy Adams

"Machine Learning Methods for Discrete Event Detection in Wearable Sensor Data Streams"

Wearable wireless sensors have the potential for transformative impact on the fields of health and behavioral science. Recent advances in wearable sensor technology have made it possible to simultaneously collect multiple streams of physiological and context data from individuals as they go about their daily activities in natural environments; however, extracting reliable higher-level inferences from these raw data streams remains a key data analysis challenge. In this work, we focus on the problem of discrete event detection from wearable sensor streams. First, we improve detection accuracy by adapting probabilistic context free grammar models to encode common structures in health and behavioral science. Second, we present a latent variable model that enables weakly supervised learning of discrete event detection models and propose extensions of this algorithm to structured prediction. Finally, we propose work investigating approximate inference algorithms based on cutset conditioning and model compression for structured models so that these models may be used in real-time settings.

Advisor: Benjamin Marlin