# Learning Qualitatively Diverse and Interpretable Rules for Classification

arXiv: https://arxiv.org/abs/1806.08716 · HTML: https://mlsanity.com/paper/1806.08716/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Jun 2018.

## Built on by (most cited first, 10 of 10)

- [Evading the Simplicity Bias: Training a Diverse Set of Models Discovers Solutions with Superior OOD Generalization](https://mlsanity.com/paper/2105.05612/) (2021)
- [GLocalX -- From Local to Global Explanations of Black Box AI Models](https://mlsanity.com/paper/2101.07685/) (2021)
- [ID and OOD Performance Are Sometimes Inversely Correlated on Real-world Datasets](https://mlsanity.com/paper/2209.00613/) (2022)
- [Predicting is not Understanding: Recognizing and Addressing Underspecification in Machine Learning](https://mlsanity.com/paper/2207.02598/) (2022)
- [Controlling Directions Orthogonal to a Classifier](https://mlsanity.com/paper/2201.11259/) (2022)
- [Input-gradient space particle inference for neural network ensembles](https://mlsanity.com/paper/2306.02775/) (2023)
- [CHARM3R: Towards Unseen Camera Height Robust Monocular 3D Detector](https://mlsanity.com/paper/2508.11185/) (2025)
- [Learning Diverse Features in Vision Transformers for Improved Generalization](https://mlsanity.com/paper/2308.16274/) (2023)
- [Understanding Uncertainty in Bayesian Deep Learning](https://mlsanity.com/paper/2106.13055/) (2021)
- [Diverse Concept Proposals for Concept Bottleneck Models](https://mlsanity.com/paper/2412.18059/) (2024)

## Abstract (arXiv)

There has been growing interest in developing accurate models that can also be explained to humans. Unfortunately, if there exist multiple distinct but accurate models for some dataset, current machine learning methods are unlikely to find them: standard techniques will likely recover a complex model that combines them. In this work, we introduce a way to identify a maximal set of distinct but accurate models for a dataset. We demonstrate empirically that, in situations where the data supports multiple accurate classifiers, we tend to recover simpler, more interpretable classifiers rather than more complex ones.