events

Kick-off workshop: 16-17 July 2024

This workshop brought together international experts on the topic of “Logic and AI” with the goal of exploring the promising interactions of these two fields. We had 8 talks and plenty of discussions divided into 3 subtopics: (a) different forms of neuro-symbolic integration, (b) expressive and computational power of machine learning, and (c) the intersection of causality, logic, and machine learning.

The talks are listed below (you can click on the title to see the slides, if available), and the workshop website is here.

Neuro-symbolic integration 1 (hosted by Sara Magliacane)

• Giuseppe Marra (KU Leuven)

  From Statistical Relational to Neuro-Symbolic AI

• Levin Hornischer (LMU Munich)

  Semantics for Non-symbolic Computation: Including Neural Networks and Analog Computers

Expressive power of machine learning (hosted by Balder ten Cate)

• Lena Strobl (Umeå University)

  Expressivity of Transformers: What Formal Languages Can They Represent?

• Martin Grohe (RWTH Aachen)

  The Logic of Graph Neural Networks

Neuro-symbolic integration 2 (hosted by Levin Hornischer)

• Sebastijan Dumančić (TU Delft)

  Machine learning models that provably satisfy constraints

• Herbert Jaeger (University of Groningen)

  What a mathematical foundation for unconventional computing should deliver and how it might look like

Causality, logic, and machine learning (hosted by Johan van Benthem)

• Atticus Geiger (Pr(Ai)²R)

  Causal Abstraction as a Theoretical Foundation for Mechanistic Interpretability

• Thomas Icard (Stanford University)

  Causal Inference from a Logical Point of View

We very much enjoyed the many inspiring discussions bridging the different areas of the field. We warmly thank the speakers, the on-site participants, and especially also the many online participants for their contributions to the workshop!