Research map / Label-efficient and robust learning / Self-supervised and contrastive
InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANs
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 the contrastive regularizer to modified InfoGAN | InfoGAN · dSprites | FactorVAE: 0.82 \pm 0.01 -> 0.88 \pm 0.01 | Helped |
| add the contrastive regularizer to modified InfoGAN | InfoGAN · dSprites | DCI: 0.60 \pm 0.02 -> 0.71 \pm 0.01 | Helped |
| select an InfoGAN-CR model with ModelCentrality instead of UDR Lasso | InfoGAN-CR · dSprites | FactorVAE: 0.86 \pm .01 -> 0.92 \pm .00 | Helped |
| select an InfoGAN-CR model with ModelCentrality instead of UDR Spearman | InfoGAN-CR · dSprites | FactorVAE: 0.84 \pm 0.01 -> 0.92 \pm .00 | Helped |
| add the InfoGAN loss to contrastive-regularizer training (ablation) | InfoGAN-CR · dSprites | FactorVAE: 0.39 \pm 0.02 -> 0.88 \pm 0.01 | Helped |
| progressively reduce the contrastive gap from 1.9 to 0 (ablation) | InfoGAN-CR · dSprites | FactorVAE | Helped |
| replace the factorizing decoder with a non-factorizing decoder (ablation) | InfoGAN · dSprites | FactorVAE: 0.2 | Hurt |
| use the contrastive regularizer with InfoGAN-CR (ablation) | VAE · dSprites | disentanglement | No clear effect |
| apply total-correlation regularization in GAN training (ablation) | GAN · disentangled GAN training | disentanglement | No clear effect |
Closest papers by meaning
- OOGAN: Disentangling GAN with One-Hot Sampling and Orthogonal Regularization (AAAI 2020 · similarity 0.90)
- IB-GAN: Disentangled Representation Learning with Information Bottleneck Generative Adversarial Networks (AAAI 2021 · similarity 0.90)
- Improving GAN with Neighbors Embedding and Gradient Matching (AAAI 2019 · similarity 0.90)
- Stein Latent Optimization for Generative Adversarial Networks (ICLR 2022 · similarity 0.89)
- Variational Annealing of GANs: A Langevin Perspective (ICML 2019 · similarity 0.89)
- Training GANs with Stronger Augmentations via Contrastive Discriminator (ICLR 2021 · similarity 0.89)
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.