Browsing by Author "Boler, William"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Asset Allocation with Swarm/Human Blended Intelligence
    (IEEE, 2016-10) Christopher, Lauren; Boler, William; Wieczorek, Calvin; Crespo, Jonah; Witcher, Paul; Hawkins, Scot A.; Stewart, James; Department of Electrical and Computer Engineering, School of Engineering and Technology
    PSO has been used to demonstrate the near-real-time optimization of frequency allocations and spatial positions for receiver assets in highly complex Electronic Warfare (EW) environments. The PSO algorithm computes optimal or near-optimal solutions so rapidly that multiple assets can be exploited in real-time and re-optimized on the fly as the situation changes. The allocation of assets in 3D space requires a blend of human intelligence and computational optimization. This paper advances the research on the tough problem of how humans interface to the swarm for directing the solution. The human intelligence places new pheromone-inspired spheres of influence to direct the final solution. The swarm can then react to the new input from the human intelligence. Our results indicate that this method can maintain the speed goal of less than 1 second, even with multiple spheres of pheromone influence in the solution space.
  • Thumbnail Image
    Item
    Trusted Data Anomaly Detection (TaDA) in Ground Truth Image Data
    (IEEE, 2022-10) Boler, William; Dale, Ashley; Christopher, Lauren; Electrical and Computer Engineering, School of Engineering and Technology
    Current state-of-the-art Artificial Intelligence (AI) anomaly detection from images is primarily used for defect detection and relies on relatively homogeneous datasets of images with similar foregrounds and backgrounds. This type of anomaly detection uses human labelled ground-truth data. In our research, we have extremely heterogeneous datasets and want to identify outliers. We use self-supervised Variational Autoencoders (VAEs) to identify anomalies in the latent vector feature space. Understanding the outliers in a large training data set is important for establishing trustworthiness of the AI models learned from these data, a strong requirement for military AI applications. Our study uses 8984 examples from Kaggle military planes and 4300 examples from Kaggle landscape data. We present the results of the combined heterogeneous dataset on the localized methods, with one such result exhibiting inliers as landscapes/backgrounds and outliers as all aircraft, detecting aircraft as anomalies with a 0.87 AUC. Results also include the inter-class AUC across the different aircraft classes. Our contribution to the state-of-the-art is to apply isolation forests to the latent space data after UMAP embeddings in a strongly heterogeneous image dataset for military applications to identify anomalies.