Deep Adaptive Semantic Logic (DASL): Compiling Declarative Knowledge into Deep Neural Networks

TL;DR

DASL is a framework that integrates formal knowledge into deep neural networks, improving learning efficiency and performance by leveraging logical structures and knowledge representations.

Contribution

It introduces a formal semantics for knowledge integration in neural networks, enabling deeper logical reasoning and reducing data needs in complex tasks.

Findings

Knowledge representation captures all of first order logic.

Reduces data requirements by a factor of 1000 in a toy problem.

Improves visual relationship detection performance by 10.7% with commonsense knowledge.

Abstract

We introduce Deep Adaptive Semantic Logic (DASL), a novel framework for automating the generation of deep neural networks that incorporates user-provided formal knowledge to improve learning from data. We provide formal semantics that demonstrate that our knowledge representation captures all of first order logic and that finite sampling from infinite domains converges to correct truth values. DASL's representation improves on prior neural-symbolic work by avoiding vanishing gradients, allowing deeper logical structure, and enabling richer interactions between the knowledge and learning components. We illustrate DASL through a toy problem in which we add structure to an image classification problem and demonstrate that knowledge of that structure reduces data requirements by a factor of $1000$. We then evaluate DASL on a visual relationship detection task and demonstrate that the addition of commonsense knowledge improves performance by $10.7\%$ in a data scarce setting.