ExplainabilityTransparencyAccountabilitycs.AI
Explainable AI through the Learning of Arguments
Jonas Bei, David Pomerenke, Lukas Schreiner, Sepideh Sharbaf, Pieter Collins, Nico Roos
February 1, 2022
arXiv.org
1 citations
Key Takeaways
-
Arguments as interpretable rules: The paper demonstrates that learning arguments—small rules with exceptions that can be chained together—offers a human-interpretable alternative to black-box AI models, making it easier for non-experts to understand how AI systems reach decisions.
-
Scalability beyond legal cases: While argument learning was originally developed for legal case analysis with small datasets, this research tests whether the approach can work with much larger datasets, potentially expanding its applicability to diverse real-world domains beyond the courtroom.
-
Comparison with established methods: The study benchmarks argument learning against HeRO algorithms and decision trees, suggesting that researchers are systematically evaluating whether this symbolic approach is competitive with other interpretable machine learning techniques.
-
Symbolic AI for explainability: By using symbolic representations (arguments with exceptions) rather than numerical weights or neural networks, this approach aligns with efforts to make AI systems more transparent and auditable—key concerns for responsible AI deployment in high-stakes decisions.
Abstract
Learning arguments is highly relevant to the field of explainable artificial intelligence. It is a family of symbolic machine learning techniques that is particularly human-interpretable. These techniques learn a set of arguments as an intermediate representation. Arguments are small rules with exceptions that can be chained to larger arguments for making predictions or decisions. We investigate the learning of arguments, specifically the learning of arguments from a 'case model' proposed by Verheij [34]. The case model in Verheij's approach are cases or scenarios in a legal setting. The number of cases in a case model are relatively low. Here, we investigate whether Verheij's approach can be used for learning arguments from other types of data sets with a much larger number of instances. We compare the learning of arguments from a case model with the HeRO algorithm [15] and learning a decision tree.