Research map / Probabilistic and scientific modeling / Variational and generative sampling
Information Constraints on Auto-Encoding Variational Bayes
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 |
|---|---|---|---|
| Penalize dependence between the biological latent representation and quality-control nuisance metrics with dHSIC | scVI · PBMC | IDR reproducible correlation: 0.26~$\pm$~0.07 -> 0.43~$\pm$~0.02 | Helped |
| Penalize dependence between the biological latent representation and quality-control nuisance metrics with dHSIC | scVI · PBMC | quality control metrics coefficient of determination (linear regression): 0.195 -> 0.176 | Helped |
| Penalize dependence between the biological latent representation and quality-control nuisance metrics with dHSIC | scVI · PBMC | quality control metrics coefficient of determination (random forest): 0.129 -> 0.123 | Helped |
| Penalize dependence between the biological latent representation and quality-control nuisance metrics with dHSIC | scVI · PBMC | IDR mixture weight: 0.213~$\pm$~0.001 -> 0.217~$\pm$~0.003 | No clear effect |
| Use HSIC-constrained representation learning with lighting direction as a continuous nuisance variable | VAE · Extended Yale B | person identity accuracy: 0.72 -> 0.75 | Helped |
| Use HSIC-constrained representation learning with lighting direction as a continuous nuisance variable | VAE · Extended Yale B | lighting direction average error in degrees (linear regression): 9.40 -> 28.04 | Mixed |
| Use HSIC-constrained representation learning with lighting direction as a continuous nuisance variable | VAE · Extended Yale B | lighting group average classification error (logistic regression): 0.11 -> 0.29 | Mixed |
| Use HSIC-constrained representation learning with lighting direction as a continuous nuisance variable | VAE · Extended Yale B | lighting group average classification error (random forest): 0.26 -> 0.52 | Mixed |
| Use HSIC-constrained representation learning with lighting direction as a continuous nuisance variable | VAE · Extended Yale B | lighting direction average error in degrees (random forest): 14.07 -> 36.15 | Mixed |
Closest papers by meaning
- Decision-Making with Auto-Encoding Variational Bayes (NeurIPS 2020 · similarity 0.87)
- Robust Learning with the Hilbert-Schmidt Independence Criterion (ICML 2020 · similarity 0.87)
- Variational Learning of Fractional Posteriors (ICML 2025 · similarity 0.86)
- Coupled Variational Bayes via Optimization Embedding (NeurIPS 2018 · similarity 0.86)
- Variational Inference for Sparse and Undirected Models (ICML 2017 · similarity 0.86)
- Kernel Semi-Implicit Variational Inference (ICML 2024 · similarity 0.85)
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.