# Teaching a GAN What Not to Learn

arXiv: https://arxiv.org/abs/2010.15639 · HTML: https://mlsanity.com/paper/2010.15639/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Oct 2020.

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

- [Understanding Instance-based Interpretability of Variational Auto-Encoders](https://mlsanity.com/paper/2105.14203/) (2021)
- [Data Redaction from Pre-trained GANs](https://mlsanity.com/paper/2206.14389/) (2022)
- [Constraining Generative Models for Engineering Design with Negative Data](https://mlsanity.com/paper/2306.15166/) (2023)
- [RareGAN: Generating Samples for Rare Classes](https://mlsanity.com/paper/2203.10674/) (2022)
- [Data Redaction from Conditional Generative Models](https://mlsanity.com/paper/2305.11351/) (2023)
- [IB-GAN: A Unified Approach for Multivariate Time Series Classification under Class Imbalance](https://mlsanity.com/paper/2110.07460/) (2021)
- [OMASGAN: Out-of-Distribution Minimum Anomaly Score GAN for Sample Generation on the Boundary](https://mlsanity.com/paper/2110.15273/) (2021)
- [FROB: Few-shot ROBust Model for Classification and Out-of-Distribution Detection](https://mlsanity.com/paper/2111.15487/) (2021)
- [Leveraging Contaminated Datasets to Learn Clean-Data Distribution with Purified Generative Adversarial Networks](https://mlsanity.com/paper/2302.01722/) (2023)
- [Mining Multi-Label Samples from Single Positive Labels](https://mlsanity.com/paper/2206.05764/) (2022)
- [Heavy-Tailed Class-Conditional Priors for Long-Tailed Generative Modeling](https://mlsanity.com/paper/2509.02154/) (2025)
- [Label Geometry Aware Discriminator for Conditional Generative Networks](https://mlsanity.com/paper/2105.05501/) (2021)
- [ScarceGAN: Discriminative Classification Framework for Rare Class Identification for Longitudinal Data with Weak Prior](https://mlsanity.com/paper/2505.03811/) (2025)

## Abstract (arXiv)

Generative adversarial networks (GANs) were originally envisioned as unsupervised generative models that learn to follow a target distribution. Variants such as conditional GANs, auxiliary-classifier GANs (ACGANs) project GANs on to supervised and semi-supervised learning frameworks by providing labelled data and using multi-class discriminators. In this paper, we approach the supervised GAN problem from a different perspective, one that is motivated by the philosophy of the famous Persian poet Rumi who said, "The art of knowing is knowing what to ignore." In the GAN framework, we not only provide the GAN positive data that it must learn to model, but also present it with so-called negative samples that it must learn to avoid - we call this "The Rumi Framework." This formulation allows the discriminator to represent the underlying target distribution better by learning to penalize generated samples that are undesirable - we show that this capability accelerates the learning process of the generator. We present a reformulation of the standard GAN (SGAN) and least-squares GAN (LSGAN) within the Rumi setting. The advantage of the reformulation is demonstrated by means of experiments conducted on MNIST, Fashion MNIST, CelebA, and CIFAR-10 datasets. Finally, we consider an application of the proposed formulation to address the important problem of learning an under-represented class in an unbalanced dataset. The Rumi approach results in substantially lower FID scores than the standard GAN frameworks while possessing better generalization capability.