Multi-Criteria Comparison as a Method of Advancing Knowledge-Guided Machine Learning
This addresses the challenge of comparing diverse models in fields like Psychology and Decision Science, though it appears incremental as an adaptation of existing evaluation techniques.
The paper tackles the problem of evaluating AI/ML models by proposing a generalizable method that compares models across multiple criteria, such as scientific principles and practical outcomes, using ordinal rankings and voting rules from computational social choice to enable holistic assessment.
This paper describes a generalizable model evaluation method that can be adapted to evaluate AI/ML models across multiple criteria including core scientific principles and more practical outcomes. Emerging from prediction competitions in Psychology and Decision Science, the method evaluates a group of candidate models of varying type and structure across multiple scientific, theoretic, and practical criteria. Ordinal ranking of criteria scores are evaluated using voting rules from the field of computational social choice and allow the comparison of divergent measures and types of models in a holistic evaluation. Additional advantages and applications are discussed.