Research map / Label-efficient and robust learning / Recognition and long-tail
InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add variational mutual information regularization (InfoGAN) to GAN objective (ablation) | GAN on MNIST · MNIST | lower bound LI (mutual information) | Helped |
| add information regularization on categorical code c1 (InfoGAN) vs. regular GAN (ablation) | GAN on MNIST · MNIST | qualitative interpretability of c1 variation | Helped |
| unsupervised InfoGAN categorical code used as digit classifier | InfoGAN · MNIST | classification error rate | No clear effect |
| add information regularization with a categorical latent code (lambda=1) | InfoGAN · 3D Faces / Chairs / SVHN / CelebA | qualitative disentanglement of factors (pose, lighting, width, glasses, hair, em | Helped |
| InfoGAN objective with Q sharing conv layers with D (negligible extra cost) | InfoGAN · training cost | computation cost relative to GAN | No clear effect |
Closest papers by meaning
- Unsupervised Disentanglement of Linear-Encoded Facial Semantics (CVPR 2021 · similarity 0.89)
- IB-GAN: Disentangled Representation Learning with Information Bottleneck Generative Adversarial Networks (AAAI 2021 · similarity 0.88)
- Interpreting the Latent Space of GANs for Semantic Face Editing (CVPR 2020 · similarity 0.87)
- Interpretable Generative Adversarial Networks (AAAI 2022 · similarity 0.87)
- Adversarial Latent Autoencoders (CVPR 2020 · similarity 0.87)
- Discovering Interpretable Latent Space Directions of GANs Beyond Binary Attributes (CVPR 2021 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.