Logic rules overview improves neurosymbolic AI across fields
Rule-Based Languages for Neurosymbolic AI
Artificial IntelligenceLogic in Computer Science
Summary
Logic programming helps computers combine math-like rules with neural networks to solve problems more effectively. The authors reviewed many kinds of rule-based languages used in this combination, comparing how they work, how powerful they are, how they interact with neural networks, and how they process information. They also looked at popular systems in databases, machine learning, vision, and robotics to see which languages fit best in each area. Finally, they created a guide to help decide which logic rules to use for different tasks and pointed out challenges that remain open.
What this means in practice
- •For database developers: Select the best logic rule languages for managing complex queries combined with neural networks in databases.
- •For robotics engineers: Choose rule-based systems that effectively integrate symbolic logic with neural models to enhance robot decision-making.
A survey. It maps existing work.
Authors
Stefania Dumbrava, Efthymia Tsamoura
Abstract
Logic programming is increasingly used as the symbolic component of neurosymbolic AI systems. We survey the main rule-based languages in this setting, namely Datalog, answer set, and probabilistic logic programs, along four axes: semantics, expressiveness, neural integration, and evaluation mechanism. We analyse over 50 recent systems and applications, comparing formalism usage across four research areas: databases and programming languages, machine learning, vision, and robotics. We provide a decision matrix mapping application scenarios to required features and close by outlining open problems.