Computational Mathematics and Scientific Computing Seminar

Time and Location:

Speaker:

Link:

Is it possible to guess what a scene in an image would look like if the picture was taken from a different angle? Would it sound like you if an AI generated a deepfake of your voice? Can we find the solution of a PDE we have never seen, if we collect enough solutions of nearby equations? These questions seem to fit in a common mathematical framework of estimation of low-dimensional latent processes under maps of controlled complexity. Deep networks with sufficient symmetries can provide a compelling tool for redatuming, backed by theoretical recovery guarantees not unlike matrix completion. I will discuss one example in particular: the disentangling of path effects vs source effects in seismograms. Joint work with Pawan Bharadwaj, Matt Li, and Borjan Geshkovski.