Lance M. Kaplan

Information fusion is basically the weighted averaging of data from different sources where the weights are inversely proportional to the uncertainty for the data sources. Generally, the uncertainty is aggregated from likelihood models to characterize the probability of the unknown states in light of the observations.  In many fusion systems, the likelihood functions are presumed to be known, but in practice they must be machine learned via a calibration process. In Army applications, there can be little training data to accurately learn these likelihoods. This talk will address the epistemic uncertainty as a second-order uncertainty about the likelihoods in cases where very little training exists.  Specifically, the talk will highlight new methods to compute error bars around probabilistic outputs of Bayesian and neural networks.  Furthermore, it enables new paradigms for establishing prediction sets of feasible hypotheses rather than the most likely hypothesis, which can be very misleading in cases of imbalance of epistemic uncertainty.


Lance M. Kaplan received his undergraduate degree at Duke University in 1989 and a PhD degree from the University of Southern California in 1994, all in Electrical Engineering. He held a National Science Foundation Graduate Fellowship and a USC Dean’s Merit Fellowship from 1990–1993.  Dr. Kaplan previously worked at the Georgia Tech Research Institute (1987-1990) and the Hughes Aircraft Company (1994-1996).  He was a faculty member in the Department of Engineering at Clark Atlanta University from 1996-2004.  Currently, he is a team leader in the Context Aware Processing branch of the DEVCOM Army Research Laboratory (ARL). Dr. Kaplan serves as VP Publications for the IEEE Aerospace and Electronic Systems (AES) Society (2021-Present) and as VP Conferences for the International Society of Information Fusion (ISIF) (2014-Present). Previously, he served as Editor-In-Chief for the IEEE Transactions on AES (2012-2017), on the Board of Governors for the IEEE AES Society (2008-2013, 2018-2020) and on the Board of Directors of ISIF (2012-2014). He is a Fellow of IEEE and of ARL. His current research interests include information/data fusion, reasoning under uncertainty, network science, resource management and signal and image processing.