Research map / Probabilistic and scientific modeling / Variational and generative sampling
Learning Latent Subspaces in Variational Autoencoders
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 |
|---|---|---|---|
| Use CSVAE's label-associated latent subspaces instead of CondVAE | attribute transfer · TFD | classifier accuracy: 62.97% -> 76.23% | Helped |
| Use CSVAE's label-associated latent subspaces instead of CondVAE | attribute transfer · CelebA-Glasses | classifier accuracy: 96.04% -> 99.59% | Helped |
| Use CSVAE's label-associated latent subspaces instead of CondVAE | attribute transfer · CelebA-FacialHair | classifier accuracy: 88.93% -> 97.75% | Helped |
| Add the CSVAE latent-subspace approach rather than CondVAE | TFD attribute transfer · TFD | target - changed MSE: 74.3354 -> 71.0858 | Helped |
| Add the CSVAE latent-subspace approach rather than CondVAE | TFD attribute transfer · TFD | original - changed MSE: 18.3365 -> 28.1997 | Mixed |
| Add information-factorization loss to CondVAE | CondVAE-info · TFD | classifier accuracy: 62.97% -> 62.27% | No clear effect |
| Add information-factorization loss to CondVAE | CondVAE-info · CelebA-Glasses | classifier accuracy: 96.04% -> 95.16% | No clear effect |
| Add information-factorization loss to CondVAE | CondVAE-info · CelebA-FacialHair | classifier accuracy: 88.93% -> 88.03% | No clear effect |
Closest papers by meaning
- Trading Information between Latents in Hierarchical Variational Autoencoders (ICLR 2023 · similarity 0.90)
- Information-Theoretic Generalization Bounds for VAEs: A Role of Encoder and Latent Variable (ICML 2026 · similarity 0.90)
- Capturing Label Characteristics in VAEs (ICLR 2021 · similarity 0.90)
- Variational Autoencoders Pursue PCA Directions (by Accident) (CVPR 2019 · similarity 0.90)
- An Identifiable Double VAE For Disentangled Representations (ICML 2021 · similarity 0.90)
- A Statistical Analysis of Wasserstein Autoencoders for Intrinsically Low-dimensional Data (ICLR 2024 · 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.