Trained DC-GANs to predict the evolution of convective weather systems for use in the FAA NextGen program
Wrote white paper defining current state-of-the-art for sequence-to-sequence prediction within the ML community
Designed convolutional neural network architecture for radar images utilizing both amplitude and phase by generalizing model weights and activations to the complex plane
Developed data augmentation and simulation pipeline to improve model robustness
Led data collection and ground testing for an experimental automatic target recognition (ATR) radar system