Research map / Probabilistic and scientific modeling / Molecules and proteins
Variational Learning of Disentangled Representations
Yuli Slavutsky, Ozgur Beker, David Blei, Bianca Dumitrascu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 adversarial classifier (ablation) | CSVAE · Parametric model (synthetic) | KL(q(z|x) || p(z|x)): 6.65 -> 2.85 | Improved |
| add adversarial classifier (ablation) | CSVAE · Parametric model (synthetic) | Delta-Bayes: 24.83 -> 24.33 | Improved |
| add adversarial classifier (ablation) | HCSVAE · Parametric model (synthetic) | KL(q(z|x) || p(z|x)): 4.01 -> 3.99 | No clear effect |
| add adversarial classifier (ablation) | HCSVAE · Noisy Swiss roll (rho=0.3) | I(z;w): 0.024 -> 0.002 | Improved |
| add adversarial classifier (ablation) | CSVAE · Noisy Swiss roll (rho=0.3) | I(z;w): 0.047 -> 0.031 | Improved |
| add adversarial classifier (ablation) | CSVAE · Colored MNIST (label noise p=0.3) | marginal RMSE: 0.142 -> 0.181 | Hurt |
| add adversarial classifier (ablation) | HCSVAE · Colored MNIST (label noise p=0.3) | marginal RMSE: 0.151 -> 0.141 | Improved |
| use DisCoVR instead of CCVAE | CCVAE · Parametric model (synthetic) | Delta-Bayes: 3.57 -> 0.1 | Improved |
| use DisCoVR instead of CCVAE | CCVAE · Noisy Swiss roll (rho=0.3) | I(z;w): 0.502 -> 0.005 | Improved |
| use DisCoVR instead of CCVAE | CCVAE · Colored MNIST (label noise p=0.3) | marginal RMSE: 0.099 -> 0.087 | Improved |
| use DisCoVR instead of CCVAE | CCVAE · CelebA-Glasses | I(z;w): 0.083 -> 0.030 | Improved |
| use DisCoVR instead of CCVAE | CCVAE · CelebA-Glasses | NLL: 131.764 -> 135.677 | Hurt |
Closest papers by meaning
- A Bayesian Nonparametric Framework For Learning Disentangled Representations (ICLR 2026 · similarity 0.89)
- Disentangling Factors of Variation with Cycle-Consistent Variational Auto-Encoders (ECCV 2018 · similarity 0.89)
- Disentangling Disentanglement in Variational Autoencoders (ICML 2019 · similarity 0.88)
- $\alpha$TC-VAE: On the relationship between Disentanglement and Diversity (ICLR 2024 · similarity 0.88)
- An Information Criterion for Controlled Disentanglement of Multimodal Data (ICLR 2025 · similarity 0.88)
- Synergy Between Sufficient Changes and Sparse Mixing Procedure for Disentangled Representation Learning (ICLR 2025 · similarity 0.87)
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.
