# I Am Going MAD: Maximum Discrepancy Competition for Comparing Classifiers Adaptively

arXiv: https://arxiv.org/abs/2002.10648 · HTML: https://mlsanity.com/paper/2002.10648/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Feb 2020.

## Built on by (most cited first, 13 of 13)

- [RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content](https://mlsanity.com/paper/2101.10955/) (2021)
- [AugMax: Adversarial Composition of Random Augmentations for Robust Training](https://mlsanity.com/paper/2110.13771/) (2021)
- [Self-Damaging Contrastive Learning](https://mlsanity.com/paper/2106.02990/) (2021)
- [Adaptive Testing of Computer Vision Models](https://mlsanity.com/paper/2212.02774/) (2022)
- [Taxonomy of Machine Learning Safety: A Survey and Primer](https://mlsanity.com/paper/2106.04823/) (2021)
- [Sample-Efficient Human Evaluation of Large Language Models via Maximum Discrepancy Competition](https://mlsanity.com/paper/2404.08008/) (2024)
- [Exposing Semantic Segmentation Failures via Maximum Discrepancy Competition](https://mlsanity.com/paper/2103.00259/) (2021)
- [Debiased Subjective Assessment of Real-World Image Enhancement](https://mlsanity.com/paper/2106.10080/) (2021)
- [Image Quality Assessment: Integrating Model-Centric and Data-Centric Approaches](https://mlsanity.com/paper/2207.14769/) (2022)
- [Semi-Supervised Domain Adaptation with Auto-Encoder via Simultaneous Learning](https://mlsanity.com/paper/2210.09486/) (2022)
- [Learning Where to Edit Vision Transformers](https://mlsanity.com/paper/2411.01948/) (2024)
- [Computational Analysis of Degradation Modeling in Blind Panoramic Image Quality Assessment](https://mlsanity.com/paper/2503.03255/) (2025)
- [Steerable Pyramid Transform Enables Robust Left Ventricle Quantification](https://mlsanity.com/paper/2201.08388/) (2022)

## Abstract (arXiv)

The learning of hierarchical representations for image classification has experienced an impressive series of successes due in part to the availability of large-scale labeled data for training. On the other hand, the trained classifiers have traditionally been evaluated on small and fixed sets of test images, which are deemed to be extremely sparsely distributed in the space of all natural images. It is thus questionable whether recent performance improvements on the excessively re-used test sets generalize to real-world natural images with much richer content variations. Inspired by efficient stimulus selection for testing perceptual models in psychophysical and physiological studies, we present an alternative framework for comparing image classifiers, which we name the MAximum Discrepancy (MAD) competition. Rather than comparing image classifiers using fixed test images, we adaptively sample a small test set from an arbitrarily large corpus of unlabeled images so as to maximize the discrepancies between the classifiers, measured by the distance over WordNet hierarchy. Human labeling on the resulting model-dependent image sets reveals the relative performance of the competing classifiers, and provides useful insights on potential ways to improve them. We report the MAD competition results of eleven ImageNet classifiers while noting that the framework is readily extensible and cost-effective to add future classifiers into the competition. Codes can be found at https://github.com/TAMU-VITA/MAD.