Congratulations to Dr. Changzhe Jiao for successfully defending his Ph.D. thesis yesterday! His dissertation is titled “Target Concept Learning from Ambiguously Labeled Data.” His research focused on developing methods that can learn target concepts, a signature or feature vector that describes and represents a target class of interest, from ambiguously and imprecisely labeled training data. The goal of his work is to optimize target concepts for use in target detection while minimizing the effort needed to precisely label data.

Read about his research:

C. Jiao, B. Su, P. Lyons, A. Zare, K. C. Ho and M. Skubic, “Multiple Instance Dictionary Learning for Beat-to-Beat Heart Rate Monitoring from Ballistocardiograms,” Under Review.

A. Zare, C. Jiao, and T. Glenn, “Multiple Instance Hyperspectral Target Characterization,” Under Review.

C. Jiao and A. Zare, “Multiple Instance Hybrid Estimator for Learning Target Signatures” Proc. IEEE Intl. Geosci. Remote Sens. Symp. (IGARSS), Fort Worth, TX, 2017.

C. Jiao, P. Lyons, A. Zare, L. Rosales, and M. Skubic, “Heart Beat Characterization from Ballistocardiogram Signals using Extended Functions of Multiple Instances,” in Proc. EMBC, 2016.

C. Jiao and A. Zare, “Multiple Instance Dictionary Learning using Functions of Multiple Instances,” in Int. Conf. Pattern Recognition (ICPR), 2016.

C. Jiao and A. Zare, “Functions of Multiple Instances for Learning Target Signatures,” IEEE Trans. Geosci. Remote Sens., vol. 53, iss. 8, pp. 4670-4686, 2015.

A. Zare and C. Jiao, “Functions of multiple instances for sub-pixel target characterization in hyperspectral imagery,” in Proc. SPIE 9472, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXI, 2015.

A. Zare and C. Jiao, “Extended functions of multiple instances for target characterization,” in 6th IEEE Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), 2014.