# InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANs

arXiv: https://arxiv.org/abs/1906.06034 · HTML: https://mlsanity.com/paper/1906.06034/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2019.

## Built on by (most cited first, 16 of 16)

- [Disentangled Representation Learning](https://mlsanity.com/paper/2211.11695/) (2022)
- [Progressive Disentangled Representation Learning for Fine-Grained Controllable Talking Head Synthesis](https://mlsanity.com/paper/2211.14506/) (2022)
- [Diagonal Attention and Style-based GAN for Content-Style Disentanglement in Image Generation and Translation](https://mlsanity.com/paper/2103.16146/) (2021)
- [Where and What? Examining Interpretable Disentangled Representations](https://mlsanity.com/paper/2104.05622/) (2021)
- [Commutative Lie Group VAE for Disentanglement Learning](https://mlsanity.com/paper/2106.03375/) (2021)
- [FaceController: Controllable Attribute Editing for Face in the Wild](https://mlsanity.com/paper/2102.11464/) (2021)
- [Closed-Loop Unsupervised Representation Disentanglement with β-VAE Distillation and Diffusion Probabilistic Feedback](https://mlsanity.com/paper/2402.02346/) (2024)
- [Unsupervised Disentanglement of Linear-Encoded Facial Semantics](https://mlsanity.com/paper/2103.16605/) (2021)
- [Bundle Networks: Fiber Bundles, Local Trivializations, and a Generative Approach to Exploring Many-to-one Maps](https://mlsanity.com/paper/2110.06983/) (2021)
- [Semantic and Geometric Unfolding of StyleGAN Latent Space](https://mlsanity.com/paper/2107.04481/) (2021)
- [CoDeGAN: Contrastive Disentanglement for Generative Adversarial Network](https://mlsanity.com/paper/2103.03636/) (2021)
- [Symbolic Disentangled Representations for Images](https://mlsanity.com/paper/2412.19847/) (2024)
- [Learning Sparse Latent Representations for Generator Model](https://mlsanity.com/paper/2209.09949/) (2022)
- [How Video Meetings Change Your Expression](https://mlsanity.com/paper/2406.00955/) (2024)
- [ContraFeat: Contrasting Deep Features for Semantic Discovery](https://mlsanity.com/paper/2212.07277/) (2022)
- [URECA: The Chain of Two Minimum Set Cover Problems exists behind Adaptation to Shifts in Semantic Code Search](https://mlsanity.com/paper/2502.07494/) (2025)

## Abstract (arXiv)

Disentangled generative models map a latent code vector to a target space, while enforcing that a subset of the learned latent codes are interpretable and associated with distinct properties of the target distribution. Recent advances have been dominated by Variational AutoEncoder (VAE)-based methods, while training disentangled generative adversarial networks (GANs) remains challenging. In this work, we show that the dominant challenges facing disentangled GANs can be mitigated through the use of self-supervision. We make two main contributions: first, we design a novel approach for training disentangled GANs with self-supervision. We propose contrastive regularizer, which is inspired by a natural notion of disentanglement: latent traversal. This achieves higher disentanglement scores than state-of-the-art VAE- and GAN-based approaches. Second, we propose an unsupervised model selection scheme called ModelCentrality, which uses generated synthetic samples to compute the medoid (multi-dimensional generalization of median) of a collection of models. The current common practice of hyper-parameter tuning requires using ground-truths samples, each labelled with known perfect disentangled latent codes. As real datasets are not equipped with such labels, we propose an unsupervised model selection scheme and show that it finds a model close to the best one, for both VAEs and GANs. Combining contrastive regularization with ModelCentrality, we improve upon the state-of-the-art disentanglement scores significantly, without accessing the supervised data.