# Latent reweighting, an almost free improvement for GANs

arXiv: https://arxiv.org/abs/2110.09803 · HTML: https://mlsanity.com/paper/2110.09803/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 38 references that are in the map.
- First posted on arXiv 19 Oct 2021.

## Builds on (most cited first, 22 of 22)

- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Spectral Normalization for Generative Adversarial Networks](https://mlsanity.com/paper/1802.05957/) (2018)
- [LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop](https://mlsanity.com/paper/1506.03365/) (2015)
- [Improved Precision and Recall Metric for Assessing Generative Models](https://mlsanity.com/paper/1904.06991/) (2019)
- [Towards Principled Methods for Training Generative Adversarial Networks](https://mlsanity.com/paper/1701.04862/) (2017)
- [Assessing Generative Models via Precision and Recall](https://mlsanity.com/paper/1806.00035/) (2018)
- [Geometric GAN](https://mlsanity.com/paper/1705.02894/) (2017)
- [Robust Optimal Transport with Applications in Generative Modeling and Domain Adaptation](https://mlsanity.com/paper/2010.05862/) (2020)
- [DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data](https://mlsanity.com/paper/1706.02071/) (2017)
- [Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Sampling](https://mlsanity.com/paper/2003.06060/) (2020)
- [Bias Correction of Learned Generative Models using Likelihood-Free Importance Weighting](https://mlsanity.com/paper/1906.09531/) (2019)
- [Discriminator Rejection Sampling](https://mlsanity.com/paper/1810.06758/) (2018)
- [Metropolis-Hastings Generative Adversarial Networks](https://mlsanity.com/paper/1811.11357/) (2018)
- [Disconnected Manifold Learning for Generative Adversarial Networks](https://mlsanity.com/paper/1806.00880/) (2018)
- [Distributionally Robust Counterfactual Risk Minimization](https://mlsanity.com/paper/1906.06211/) (2019)
- [Discriminator optimal transport](https://mlsanity.com/paper/1910.06832/) (2019)
- [Learning disconnected manifolds: a no GANs land](https://mlsanity.com/paper/2006.04596/) (2020)

## Built on by (most cited first, 6 of 6)

- [The Role of ImageNet Classes in Fr'echet Inception Distance](https://mlsanity.com/paper/2203.06026/) (2022)
- [Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values](https://mlsanity.com/paper/2203.01993/) (2022)
- [Precision-Recall Divergence Optimization for Generative Modeling with GANs and Normalizing Flows](https://mlsanity.com/paper/2305.18910/) (2023)
- [Optimal Budgeted Rejection Sampling for Generative Models](https://mlsanity.com/paper/2311.00460/) (2023)
- [LSRS: Latent Scale Rejection Sampling for Visual Autoregressive Modeling](https://mlsanity.com/paper/2512.03796/) (2025)
- [Training Normalizing Flows with the Precision-Recall Divergence](https://mlsanity.com/paper/2302.00628/) (2023)

## Abstract (arXiv)

Standard formulations of GANs, where a continuous function deforms a connected latent space, have been shown to be misspecified when fitting different classes of images. In particular, the generator will necessarily sample some low-quality images in between the classes. Rather than modifying the architecture, a line of works aims at improving the sampling quality from pre-trained generators at the expense of increased computational cost. Building on this, we introduce an additional network to predict latent importance weights and two associated sampling methods to avoid the poorest samples. This idea has several advantages: 1) it provides a way to inject disconnectedness into any GAN architecture, 2) since the rejection happens in the latent space, it avoids going through both the generator and the discriminator, saving computation time, 3) this importance weights formulation provides a principled way to reduce the Wasserstein's distance to the target distribution. We demonstrate the effectiveness of our method on several datasets, both synthetic and high-dimensional.