A minimal surface is the surface of minimal area between any given boundaries. In nature such shapes result from an equilibrium of homogeneous tension, e.g. in a soap film. Minimal surfaces have a constant mean curvature of zero, i.e. the sum of the principal curvatures at each point is zero. Particularly fascinating are minimal surfaces…
The program focused on the following four themes:
- Optimization: How and why can deep models be fit to observed (training) data?
- Generalization: Why do these trained models work well on similar but unobserved (test) data?
- Robustness: How can we analyze and improve the performance of these models when applied outside their intended conditions?
- Generative methods: How can deep learning be used to model probability distributions?