A Boxology of Design Patterns for Hybrid Learning and Reasoning Systems

Abstract

We propose a set of compositional design patterns to describe a large variety of systems that combine statistical techniques from machine learning with symbolic techniques from knowledge representation. As in other areas of computer science (knowledge engineering, software engineering, ontology engineering, process mining and others), such design patterns help to systematize the literature, clarify which combinations of techniques serve which purposes, and encourage re-use of software components. We have validated our set of compositional design patterns against a large body of recent literature.

arXiv comment12 pages,55 references

References in corpus (10)

• A Review of Relational Machine Learning for Knowledge Graphs
• Deep Learning: A Critical Appraisal
• Differentiable Learning of Logical Rules for Knowledge Base Reasoning
• Towards Deep Symbolic Reinforcement Learning
• Hinge-Loss Markov Random Fields and Probabilistic Soft Logic
• A Semantic Loss Function for Deep Learning with Symbolic Knowledge
• Deep Learning for Ontology Reasoning
• Knowledge-based Transfer Learning Explanation
• Explaining Trained Neural Networks with Semantic Web Technologies: First Steps
• Reasoning over RDF Knowledge Bases using Deep Learning

Cited by in corpus (4)

• Is Neuro-Symbolic AI Meeting its Promise in Natural Language Processing? A Structured Review
• Modular Design Patterns for Hybrid Learning and Reasoning Systems: a taxonomy, patterns and use cases
• Moral Decision-Making in Medical Hybrid Intelligent Systems: A Team Design Patterns Approach to the Bias Mitigation and Data Sharing Design Problems
• Modular Design Patterns for Hybrid Actors