Timezone: »

Semantic Probabilistic Layers for Neuro-Symbolic Learning
Kareem Ahmed · Stefano Teso · Kai-Wei Chang · Guy Van den Broeck · Antonio Vergari

Thu Dec 01 09:00 AM -- 11:00 AM (PST) @ Hall J #916

We design a predictive layer for structured-output prediction (SOP) that can be plugged into any neural network guaranteeing its predictions are consistent with a set of predefined symbolic constraints. Our Semantic Probabilistic Layer (SPL) can model intricate correlations, and hard constraints, over a structured output space all while being amenable to end-to-end learning via maximum likelihood.SPLs combine exact probabilistic inference with logical reasoning in a clean and modular way, learning complex distributions and restricting their support to solutions of the constraint. As such, they can faithfully, and efficiently, model complex SOP tasks beyond the reach of alternative neuro-symbolic approaches. We empirically demonstrate that SPLs outperform these competitors in terms of accuracy on challenging SOP tasks such as hierarchical multi-label classification, pathfinding and preference learning, while retaining perfect constraint satisfaction.

Author Information

Kareem Ahmed (UCLA)
Stefano Teso (University of Trento)
Kai-Wei Chang (UCLA)
Guy Van den Broeck (UCLA)

I am an Assistant Professor and Samueli Fellow at UCLA, in the Computer Science Department, where I direct the Statistical and Relational Artificial Intelligence (StarAI) lab. My research interests are in Machine Learning (Statistical Relational Learning, Tractable Learning), Knowledge Representation and Reasoning (Graphical Models, Lifted Probabilistic Inference, Knowledge Compilation), Applications of Probabilistic Reasoning and Learning (Probabilistic Programming, Probabilistic Databases), and Artificial Intelligence in general.

Antonio Vergari (University of Edinburgh)

More from the Same Authors