EVOTER: Evolution of Transparent Explainable Rule-sets

TL;DR

EVOTER is a method that evolves transparent, explainable rule-sets for AI, enabling interpretability, bias detection, and trustworthiness, suitable for real-world applications.

Contribution

It introduces a novel evolutionary approach to generate transparent rule-sets, contrasting with black-box models, and demonstrates their effectiveness across multiple domains.

Findings

Evolved rule-sets perform comparably to black-box models.

Rules provide domain insights and reveal hidden biases.

Method is applicable to prediction, classification, and policy domains.

Abstract

Most AI systems are black boxes generating reasonable outputs for given inputs. Some domains, however, have explainability and trustworthiness requirements that cannot be directly met by these approaches. Various methods have therefore been developed to interpret black-box models after training. This paper advocates an alternative approach where the models are transparent and explainable to begin with. This approach, EVOTER, evolves rule-sets based on simple logical expressions. The approach is evaluated in several prediction/classification and prescription/policy search domains with and without a surrogate. It is shown to discover meaningful rule sets that perform similarly to black-box models. The rules can provide insight into the domain, and make biases hidden in the data explicit. It may also be possible to edit them directly to remove biases and add constraints. EVOTER thus forms a promising foundation for building trustworthy AI systems for real-world applications in the future.