Learning Interpretable Differentiable Logic Networks for Tabular Regression

TL;DR

This paper extends Differentiable Logic Networks (DLNs) to tabular regression, demonstrating they achieve comparable or better accuracy than neural networks while maintaining interpretability and computational efficiency.

Contribution

The paper introduces a unified, differentiable training framework for regression DLNs and validates their effectiveness on multiple benchmarks.

Findings

Regression DLNs match or outperform neural networks in accuracy.

DLNs provide interpretable models with fast inference.

They are a cost-effective alternative for transparent regression tasks.

Abstract

Neural networks (NNs) achieve outstanding performance in many domains; however, their decision processes are often opaque and their inference can be computationally expensive in resource-constrained environments. We recently proposed Differentiable Logic Networks (DLNs) to address these issues for tabular classification based on relaxing discrete logic into a differentiable form, thereby enabling gradient-based learning of networks built from binary logic operations. DLNs offer interpretable reasoning and substantially lower inference cost. We extend the DLN framework to supervised tabular regression. Specifically, we redesign the final output layer to support continuous targets and unify the original two-phase training procedure into a single differentiable stage. We evaluate the resulting model on 15 public regression benchmarks, comparing it with modern neural networks and classical regression baselines. Regression DLNs match or exceed baseline accuracy while preserving interpretability and fast inference. Our results show that DLNs are a viable, cost-effective alternative for regression tasks, especially where model transparency and computational efficiency are important.