Semi-Supervised Learning with IPM-based GANs: an Empirical Study
Explore this paper's citation graph
Summary
This work investigates how the design of the critic (or discriminator) influences the performance in semi-supervised learning and distill three key take-aways which are important for good SSL performance.
- Type
- preprint
- Published
- 2017-12-07
- Cited by
- 1
- References
- 28
- Access
- Open access
- OpenAlex
- https://openalex.org/W2775549693
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:21245435
Keywords
Discriminator, Computer science, Stability (learning theory), Normalization (sociology), Generative grammar
References
- On integral probability metrics, φ-divergences and binary classification
- On the empirical estimation of integral probability metrics
- Integral Probability Metrics and Their Generating Classes of Functions
- Reading Digits in Natural Images with Unsupervised Feature Learning
- Improved Techniques for Training GANs
- InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
- Normalizing the Normalizers: Comparing and Extending Network Normalization Schemes
- Towards Principled Methods for Training Generative Adversarial Networks
- Least Squares Generative Adversarial Networks
- Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
- Improved Semi-supervised Learning with GANs using Manifold Invariances
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Amortised MAP Inference for Image Super-resolution
- MMD GAN: Towards Deeper Understanding of Moment Matching Network
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Fisher GAN
- Layer Normalization
- Generative Adversarial Networks
- Learning Multiple Layers of Features from Tiny Images
- Fisher GAN
Cited by
Related papers
- Teaching a GAN What Not to Learn
- Understanding GANs: the LQG Setting
- Inductive bias for semi-supervised extreme learning machine
- Measure Based Regularization
- Mind the eigen-gap, or how to accelerate semi-supervised spectral learning algorithms
- Regularized Boost for Semi-Supervised Learning
- Designing GANs: A Likelihood Ratio Approach