Research map / Image generation / GANs
Adversarial Latent Autoencoders
Check 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 |
|---|---|---|---|
| Use ALAE short features instead of BiGAN short features | MNIST linear SVM · MNIST SW | accuracy (%): 90.52 -> 93.47 | Helped |
| Use ALAE short features instead of BiGAN short features | MNIST linear SVM · MNIST DW | accuracy (%): 91.34 -> 94.23 | Helped |
| Use ALAE short features instead of AE(l2) short features | MNIST 1NN · MNIST SW | accuracy (%): 97.52 -> 93.79 | Hurt |
| Use ALAE long features instead of BiGAN long features | MNIST linear SVM · MNIST SW | accuracy (%): 97.59 -> 98.20 | Helped |
| Use StyleALAE generation instead of StyleGAN generation | FFHQ · FFHQ | FID: 4.40 -> 13.09 | Hurt |
| Represent images in W space rather than Z space (ablation) | StyleALAE · FFHQ | PPL full: 300.5 -> 134.5 | Helped |
| Represent images in W space rather than Z space (ablation) | StyleALAE · FFHQ | PPL end: 292.0 -> 103.4 | Helped |
| Use StyleALAE instead of StyleGAN | CelebA-HQ 256×256 · CelebA-HQ | PPL full: 229.2 -> 33.29 | Helped |
Closest papers by meaning
- Autoencoding beyond pixels using a learned similarity metric (ICML 2016 · similarity 0.89)
- A Primal-Dual link between GANs and Autoencoders (NeurIPS 2019 · similarity 0.89)
- Ensembling With Deep Generative Views (CVPR 2021 · similarity 0.89)
- Interpreting the Latent Space of GANs for Semantic Face Editing (CVPR 2020 · similarity 0.89)
- A Style-Based Generator Architecture for Generative Adversarial Networks (CVPR 2019 · similarity 0.89)
- Deep Automodulators (NeurIPS 2020 · similarity 0.88)
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.