In this final short lecture, we wrap-up the course by presenting a few interesting directions for future research at the intersection of AI and cryptography. Specifically, we consider the evolutionary design of algebraic constructions for Boolean functions, differential distinguishers based and genetic programming, and the analysis of the fitness landscapes of adversarial examples with local optima networks.
- Evolving Constructions of Boolean Functions
- Neural- and Evolutionary-based distinguishers
- Local Optima Networks for loss landscapes
See also the references reported in the last slide.