# Good Semi-supervised Learning that Requires a Bad GAN

arXiv: https://arxiv.org/abs/1705.09783 · HTML: https://mlsanity.com/paper/1705.09783/

## Key facts

- Cited by 86 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 May 2017.

## Built on by (most cited first, 30 of 86)

- [Diffusion Models: A Comprehensive Survey of Methods and Applications](https://mlsanity.com/paper/2209.00796/) (2022)
- [Maximum Likelihood Training of Score-Based Diffusion Models](https://mlsanity.com/paper/2101.09258/) (2021)
- [A Survey on Deep Semi-supervised Learning](https://mlsanity.com/paper/2103.00550/) (2021)
- [A Cookbook of Self-Supervised Learning](https://mlsanity.com/paper/2304.12210/) (2023)
- [FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning](https://mlsanity.com/paper/2205.07246/) (2022)
- [A Comprehensive Survey on Graph Anomaly Detection with Deep Learning](https://mlsanity.com/paper/2106.07178/) (2021)
- [Adversarial Reciprocal Points Learning for Open Set Recognition](https://mlsanity.com/paper/2103.00953/) (2021)
- [Semi-supervised Semantic Segmentation with Directional Context-aware Consistency](https://mlsanity.com/paper/2106.14133/) (2021)
- [Recent advances and clinical applications of deep learning in medical image analysis](https://mlsanity.com/paper/2105.13381/) (2021)
- [USB: A Unified Semi-supervised Learning Benchmark for Classification](https://mlsanity.com/paper/2208.07204/) (2022)
- [Synthetic Data -- what, why and how?](https://mlsanity.com/paper/2205.03257/) (2022)
- [Diffusion Models, Image Super-Resolution And Everything: A Survey](https://mlsanity.com/paper/2401.00736/) (2024)
- [Debiased Self-Training for Semi-Supervised Learning](https://mlsanity.com/paper/2202.07136/) (2022)
- [BoostMIS: Boosting Medical Image Semi-supervised Learning with Adaptive Pseudo Labeling and Informative Active Annotation](https://mlsanity.com/paper/2203.02533/) (2022)
- [Semi-Supervised and Unsupervised Deep Visual Learning: A Survey](https://mlsanity.com/paper/2208.11296/) (2022)
- [CLDA: Contrastive Learning for Semi-Supervised Domain Adaptation](https://mlsanity.com/paper/2107.00085/) (2021)
- [Diffusion Models With Learned Adaptive Noise](https://mlsanity.com/paper/2312.13236/) (2023)
- [SemiFL: Semi-Supervised Federated Learning for Unlabeled Clients with Alternate Training](https://mlsanity.com/paper/2106.01432/) (2021)
- [Memory Consolidation Enables Long-Context Video Understanding](https://mlsanity.com/paper/2402.05861/) (2024)
- [Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels](https://mlsanity.com/paper/2302.10586/) (2023)
- [ECACL: A Holistic Framework for Semi-Supervised Domain Adaptation](https://mlsanity.com/paper/2104.09136/) (2021)
- [Towards Label-efficient Automatic Diagnosis and Analysis: A Comprehensive Survey of Advanced Deep Learning-based Weakly-supervised, Semi-supervised and Self-supervised Techniques in Histopathological Image Analysis](https://mlsanity.com/paper/2208.08789/) (2022)
- [Pseudo Label-Guided Model Inversion Attack via Conditional Generative Adversarial Network](https://mlsanity.com/paper/2302.09814/) (2023)
- [Removing Undesirable Feature Contributions Using Out-of-Distribution Data](https://mlsanity.com/paper/2101.06639/) (2021)
- [Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models](https://mlsanity.com/paper/2103.00429/) (2021)
- [MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency Regularization](https://mlsanity.com/paper/2203.14316/) (2022)
- [Neural Diffusion Models](https://mlsanity.com/paper/2310.08337/) (2023)
- [Real Risks of Fake Data: Synthetic Data, Diversity-Washing and Consent Circumvention](https://mlsanity.com/paper/2405.01820/) (2024)
- [Pseudo-label Alignment for Semi-supervised Instance Segmentation](https://mlsanity.com/paper/2308.05359/) (2023)
- [Generative Adversarial Networks: A Survey Towards Private and Secure Applications](https://mlsanity.com/paper/2106.03785/) (2021)

## Abstract (arXiv)

Semi-supervised learning methods based on generative adversarial networks (GANs) obtained strong empirical results, but it is not clear 1) how the discriminator benefits from joint training with a generator, and 2) why good semi-supervised classification performance and a good generator cannot be obtained at the same time. Theoretically, we show that given the discriminator objective, good semisupervised learning indeed requires a bad generator, and propose the definition of a preferred generator. Empirically, we derive a novel formulation based on our analysis that substantially improves over feature matching GANs, obtaining state-of-the-art results on multiple benchmark datasets.