As increasingly large nonlinear datasets are acquired, autonomy and adaptivity through deep learning at the intersection of inverse problems have to play a critical role in the interpretation of data from planetary exploration missions. Indeed, approaches and techniques capable of rapidly and intelligently extracting information from these datasets for scientific analysis need to be developed. At the same time, near-future data acquisition on (the surface of) planets and moons of our solar system probing their interiors will be necessarily scarce, which requires fundamentally new developments in the analysis of inverse problems through hierarchical approximations via symmetries, manifold and deep learning including recent advances in generative networks and meta-learning for domain adaptation accommodating, for example, interdisciplinary studies with computational simulations.
The Symposium is bringing planetary scientists, astrophysicists, cosmologists, mathematicians, computer scientists and statisticians together to present the state of the art on the one hand and emerging cross disciplinary directions of research on the other hand, while enabling new collaborations.