I-X Seminar: Logically-Consistent Deep Learning

Join I-X for the next event in its popular seminar series. Dr Antonio Vergari (University of Edinburgh) will be delivering a talk entitled “Logically-Consistent Deep Learning”.

Hybrid location:  I-X LRT608A level 6 (I-HUB) & via Teams

Seminar Summary

Guaranteeing the safety and reliability of deep learning models is of crucial importance, especially in many high-stake application scenarios. In this lecture, I will focus on the key challenge of designing probabilistic deep learning models that are reliable and yet efficient by design. I will do so within the framework of probabilistic circuits: overparametrized and computational graphs that are just neural networks with lots of structure, enough to guarantee the tractable computation of the probabilistic reasoning scenarios of interest, while not compromising their expressiveness. Second, I will discuss how we can use circuits to build a reliable foundation for neuro-symbolic AI. That is, for example, to provably satisfy certain constraints we can express in propositional logic in neural networks as to increase their performance and robustness. These models can be thought of being “verified by design” and I will showcase some recent applications of this constraint satisfaction by design e.g., scaling link prediction in graphs with millions of nodes and when constraints are both on continuous and discrete domains.