How Simple Can It Get? From Interpretable Equations to Readable Rules for Financial Decision Making

2026-08-10

AI Summary

This study addresses the critical need in highly regulated domains—such as finance—for intrinsically interpretable models rather than post hoc explanations. The authors propose constructing classifiers directly as single analytic equations, which are then systematically simplified into pruned monomials, directional if-then rules, or integer-based scorecards. Throughout this simplification process, information loss is quantified and accompanied by theoretical bounds. The methodology integrates feature pruning, rule extraction, and readability assessment, and is validated on four financial datasets. Results demonstrate that the simplified models retain nearly identical predictive performance while substantially improving interpretability. Human evaluations reveal varying preferences across users with different professional backgrounds, yet all affirm the practical utility of the approach. The core contribution lies in treating the model itself as an interpretable equation and proving that simplified rules need not replicate the original model to remain effective.

Abstract

In regulated domains such as finance, a model that cannot be explained cannot be deployed, yet many interpretable classifiers defeat their own purpose by producing formulas with dozens of features that no regulator could read. We take the reverse direction. Starting from an interpretable classifier expressed as a single equation over the input features, we progressively simplify it into more readable forms, including a pruned monomial, a directional if--then rule, and the integer scorecards and tallies that finance already deploys. Because the equation is itself the predictive model rather than a post-hoc explanation we can directly quantify what is lost under each simplification. Across four financial datasets, we find that pruning is nearly free and that fidelity can erode faster than predictive performance, allowing simpler rules to remain effective classifiers without faithfully reproducing the original model. A human assessment shows that simplification improves perceived readability, while preferences for different representations vary by professional background. Beyond measuring these losses empirically, we show that some can be anticipated from the original model: we derive a bound on the change caused by pruning and predict how faithfully a rule retaining only the direction of each feature's effect preserves the original ranking.

Problem

• interpretable models
• financial decision making
• model simplification
• readability
• regulatory compliance

Innovation

• interpretable machine learning
• model simplification
• financial decision making
• readable rules
• fidelity bounds