This lecture gives a broad overview of the area of neuro-symbolic modeling. We will see how both learning and reasoning are important for intelligent behavior, thus motivating the need for their interplay. We will also look at motivating examples where symbolic knowledge may augment a neural network. We will wrap up this unit with a discussion of the challenges we need to address for neuro-symbolic integration, and a high level overview of the kinds of approaches that exist for this integration.
Neuro-Symbolic Modeling
CS 6964, Fall 2024