In recent years, the field of automated machine learning (autoML) has quickly attracted significant attention both in academia and industry. The driving force is to reduce the amount of human intervention required to process data and create models for classification and prediction, a tedious and arbitrary process for data scientists that may not often result in achieving a global optimum with respect to multiple objectives. Moreover, existing autoML techniques rely on extremely large collections of relatively clean training data, which is not typical of Multi-Domain Battle (MDB) applications. In this paper, we describe a methodology to optimize underwater seafloor detection for airborne bathymetric lidar, an application domain with sparse truth data, leveraging evolutionary algorithms and genetic programming. Our methodology uses the Evolutionary Multi-objective Algorithm Design Engine (EMADE) and a radiometric waveform simulator generating millions of waveforms from which genetic programming techniques select optimal signal processing techniques and their parameters given the goal of reducing Total Propagated Uncertainty (TPU). The EMADE affords several benefits not found in other autoML solutions, including the ability to stack machine learning models, process time-series data using dozens of signal-processing techniques, and efficiently evaluate algorithms on multiple objectives. Given the lack of truth data, we tune EMADE to produce detection algorithms that improve accuracy and reduce relevant measurement uncertainties for a wide variety of operational and environmental scenarios. Preliminary testing indicates successfully reducing TPU and reducing over- and under-prediction errors by 13.8% and 68.2% respectively, foreshadowing using EMADE to assist in future MDB-application algorithm development.
We are interested in data fusion strategies for Intelligence, Surveillance, and Reconnaissance (ISR) missions. Advances
in theory, algorithms, and computational power have made it possible to extract rich semantic information from a wide
variety of sensors, but these advances have raised new challenges in fusing the data. For example, in developing fusion
algorithms for moving target identification (MTI) applications, what is the best way to combine image data having
different temporal frequencies, and how should we introduce contextual information acquired from monitoring cell
phones or from human intelligence? In addressing these questions we have found that existing data fusion models do not
readily facilitate comparison of fusion algorithms performing such complex information extraction, so we developed a
new model that does. Here, we present the Spatial, Temporal, Algorithm, and Cognition (STAC) model. STAC allows
for describing the progression of multi-sensor raw data through increasing levels of abstraction, and provides a way to
easily compare fusion strategies. It provides for unambiguous description of how multi-sensor data are combined, the
computational algorithms being used, and how scene understanding is ultimately achieved. In this paper, we describe
and illustrate the STAC model, and compare it to other existing models.