# On-line Adaptative Curriculum Learning for GANs

arXiv: https://arxiv.org/abs/1808.00020 · HTML: https://mlsanity.com/paper/1808.00020/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2018.

## Built on by (most cited first, 15 of 15)

- [StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesis](https://mlsanity.com/paper/2110.08985/) (2021)
- [Curriculum Learning: A Survey](https://mlsanity.com/paper/2101.10382/) (2021)
- [Projected GANs Converge Faster](https://mlsanity.com/paper/2111.01007/) (2021)
- [Ensembling Off-the-shelf Models for GAN Training](https://mlsanity.com/paper/2112.09130/) (2021)
- [Video Summarization Using Deep Neural Networks: A Survey](https://mlsanity.com/paper/2101.06072/) (2021)
- [HumanGAN: A Generative Model of Humans Images](https://mlsanity.com/paper/2103.06902/) (2021)
- [Curriculum Direct Preference Optimization for Diffusion and Consistency Models](https://mlsanity.com/paper/2405.13637/) (2024)
- [A Quantitative Metric for Privacy Leakage in Federated Learning](https://mlsanity.com/paper/2102.13472/) (2021)
- [MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators](https://mlsanity.com/paper/2107.07260/) (2021)
- [Enhanced Multimodal Representation Learning with Cross-modal KD](https://mlsanity.com/paper/2306.07646/) (2023)
- [Self-Supervised GANs with Label Augmentation](https://mlsanity.com/paper/2106.08601/) (2021)
- [Adversarial Regularization as Stackelberg Game: An Unrolled Optimization Approach](https://mlsanity.com/paper/2104.04886/) (2021)
- [ObjBlur: A Curriculum Learning Approach With Progressive Object-Level Blurring for Improved Layout-to-Image Generation](https://mlsanity.com/paper/2404.07564/) (2024)
- [Curriculum-DPO++: Direct Preference Optimization via Data and Model Curricula for Text-to-Image Generation](https://mlsanity.com/paper/2602.13055/) (2026)
- [Learning to Play: Reinforcement Learning and Games](https://mlsanity.com/paper/2610.17391/) (2026)

## Abstract (arXiv)

Generative Adversarial Networks (GANs) can successfully approximate a probability distribution and produce realistic samples. However, open questions such as sufficient convergence conditions and mode collapse still persist. In this paper, we build on existing work in the area by proposing a novel framework for training the generator against an ensemble of discriminator networks, which can be seen as a one-student/multiple-teachers setting. We formalize this problem within the full-information adversarial bandit framework, where we evaluate the capability of an algorithm to select mixtures of discriminators for providing the generator with feedback during learning. To this end, we propose a reward function which reflects the progress made by the generator and dynamically update the mixture weights allocated to each discriminator. We also draw connections between our algorithm and stochastic optimization methods and then show that existing approaches using multiple discriminators in literature can be recovered from our framework. We argue that less expressive discriminators are smoother and have a general coarse grained view of the modes map, which enforces the generator to cover a wide portion of the data distribution support. On the other hand, highly expressive discriminators ensure samples quality. Finally, experimental results show that our approach improves samples quality and diversity over existing baselines by effectively learning a curriculum. These results also support the claim that weaker discriminators have higher entropy improving modes coverage. Keywords: multiple discriminators, curriculum learning, multiple resolutions discriminators, multi-armed bandits, generative adversarial networks, smooth discriminators, multi-discriminator gan training, multiple experts.