Research map / Probabilistic and scientific modeling / Variational and generative sampling
Weakly Supervised Disentanglement by Pairwise Similarities
Junxiang Chen, Kayhan Batmanghelich
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 pairwise similarity supervision to the VAE | VAE · MNIST held-out | MIG: 0.68 vs 0.33 | Improved |
| add pairwise similarity supervision to the VAE | VAE · Fashion-MNIST held-out | MIG: 0.52 vs 0.36 | Improved |
| add pairwise similarity supervision to the VAE | VAE · Yale Faces held-out | MIG: 0.42 vs 0.29 | Improved |
| add pairwise similarity supervision to the VAE | VAE · 3D chairs held-out | MIG: 0.37 vs 0.15 | Improved |
| add pairwise similarity supervision to the VAE | VAE · 3D cars held-out | MIG: 0.41 vs 0.22 | Improved |
| add pairwise similarity supervision to the VAE | VAE · MNIST | Cohen's κ: .969 vs .704 | Improved |
| add pairwise similarity supervision to the VAE | VAE · Fashion-MNIST | Cohen's κ: .857 vs .613 | Improved |
| add pairwise similarity supervision to the VAE | VAE · Yale Faces | R²: .968 vs .760 | Improved |
| add pairwise similarity supervision to the VAE | VAE · 3D chairs | R²: .912 vs .357 | Improved |
| add pairwise similarity supervision to the VAE | VAE · 3D cars | R²: .584 vs .418 | Improved |
| add pairwise similarity supervision to the VAE | VAE · COPD | R² (FEV₁pp): .431 vs .040 | Improved |
| add pairwise similarity supervision to the VAE | VAE · COPD | Cohen's κ (GOLD): .236 vs .089 | Improved |
Closest papers by meaning
- Weakly Supervised Disentanglement with Guarantees (ICLR 2020 · similarity 0.92)
- Weakly-Supervised Disentanglement Without Compromises (ICML 2020 · similarity 0.90)
- Disentangling Disentanglement in Variational Autoencoders (ICML 2019 · similarity 0.90)
- Isolating Sources of Disentanglement in Variational Autoencoders (NeurIPS 2018 · similarity 0.90)
- Disentangling Factors of Variation with Cycle-Consistent Variational Auto-Encoders (ECCV 2018 · similarity 0.89)
- Orthogonality-Enforced Latent Space in Autoencoders: An Approach to Learning Disentangled Representations (ICML 2023 · 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.
