Scientific Meeting 17 April 2020

Speakers and titles (full abstracts are copied below):


0. Jos, Dick & Léon: "Update on COVID-19 measures at CWI", with time for Q&A
1. Jana Sotakova (QuSoft) "Breaking the Decisional Diffie-Hellman problem for class group actions"  Slides Video
2. Esteban Landerreche (Cryptology): "From Hard to Moderately-hard"  Slides Video
3. Peter Grunwald (Machine Learning): "Corona - why the measures were really needed, and why exiting is difficult"  Slides Video
3+ Short contributions from Peter Boncz (Database Architectures) on COVID-19 exit strategies and Marten van Dijk (Cryptology) on DP-3T: Decentralized Privacy-Preserving Proximity Tracing  Video


Abstracts:

1. Jana Sotakova: "Breaking the Decisional Diffie-Hellman problem for class group actions"

Isogeny based cryptography is a recent direction in post-quantum cryptography. One family of such cryptographic schemes roughly generalizes the Diffie-Hellman protocol to group actions by a class group of an imaginary quadratic field on a suitable set of elliptic curves. It was generally
believed that the structure of the group is sufficiently "hidden" by the group action - one only sees the elliptic curves, not the group elements. In many interesting cases, however, we are able to compute characters of the class group very efficiently and directly from the elliptic curves, which allows us to break the Decisional Diffie-Hellman assumption.

2. Esteban Landerreche: "From Hard to Moderately-hard"

Cryptography has generally dealt with making things impossible for an adversary to interfere. An new setting emerged with the emergence of
blockchains, where we can relax our goals and make things hard rather than impossible to achieve. In this talk we will talk about this new paradigm and give an example where we can achieve cryptographic timestamping through moderately-hard functions.

3. Peter Grunwald: "Corona - why the measures were really needed, and why exiting is difficult"

While there is surprisingly widespread acceptance of the anti-corona measures, one still hears dissenting voices like 'in 2018 6000 people died from the flu in NL and nobody cared, and most people who die had a very short life expectancy anyway, so why are we doing all this - the long-term economic damage will cause even more health damage'. I will explain why such reasoning is misguided, and, relatedly, while aiming for herd immunity is highly unlikely to work.