Neurosymbolic Programming for Science

TL;DR

Neurosymbolic Programming integrates neural and symbolic methods to enhance scientific discovery by leveraging domain knowledge, enabling interpretable models, and addressing current challenges in scientific workflows.

Contribution

This paper identifies opportunities and challenges for applying neurosymbolic programming in scientific workflows, with real-world examples from behavior analysis.

Findings

NP models can incorporate domain knowledge for interpretability

Challenges include integrating NP into existing scientific workflows

Potential to accelerate discovery across natural and social sciences

Abstract

Neurosymbolic Programming (NP) techniques have the potential to accelerate scientific discovery. These models combine neural and symbolic components to learn complex patterns and representations from data, using high-level concepts or known constraints. NP techniques can interface with symbolic domain knowledge from scientists, such as prior knowledge and experimental context, to produce interpretable outputs. We identify opportunities and challenges between current NP models and scientific workflows, with real-world examples from behavior analysis in science: to enable the use of NP broadly for workflows across the natural and social sciences.