Vector-borne disease outbreaks are closely tied to vector abundance, which makes knowledge of population dynamics useful in preventing future outbreaks. The Aedes-AI models are collection of neural network models of Aedes aegypti mosquito abundance. The models are trained on synthetic data generated from a mechanistic model, in contrast to other models of mosquito abundance that rely on noisy, real world trap data for training. This framework presents an opportunity to study and advance interpretability methods because the models are trained on data generated from known dynamics. Here, we discuss efforts to extract feature importance and represent the learned latent features of the hidden layers.

Place: Hybrid: Math, 402
Zoom:   Link  https://arizona.zoom.us/j/83541348598 
Password:  BB2022