Research map / Probabilistic and scientific modeling / Variational and generative sampling
Unsupervised Learning with Contrastive Latent Variable Models
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 robust cLVM instead of cPCA with mean imputation | Mice protein expression · Mice protein expression, natural missingness and artificially increased missingness | subgroup discovery | Helped |
| Use robust cLVM instead of standard cLVM (ablation) | Synthetic data with outliers · Synthetic data with outliers | recovery of target latent representation | Helped |
| Add ARD for automatic selection of the shared latent dimension (ablation) | cLVM · Single-cell RNA-Seq | subgroup discovery | Helped |
| Add ARD for automatic selection of the shared latent dimension (ablation) | cLVM · Synthetic data with cluster-specific target latent means | shared-dimension selection | Helped |
| Add group sparsity regularization to the target factor loading (ablation) | cLVM · mHealth sensor data | feature selection | Helped |
| Use a horseshoe prior for the target factor loading (ablation) | cLVM · mHealth sensor data | feature selection | Helped |
| Use cVAE instead of a standard VAE with the same combined latent dimensionality | VAE · MNIST digits on grass | digit clustering in latent space | Helped |
| Use a two-dimensional target latent space and an eight-dimensional shared space to separate target and background structure (ablation) | cVAE · MNIST digits on grass | selective generation of target and background | Helped |
| Use cLVM instead of PCA | cLVM · Single-cell RNA-Seq | subgroup discovery | Helped |
| Use cLVM instead of PCA | cLVM · Mice protein expression | subgroup discovery | Helped |
| Set the shared latent dimension to zero (ablation) | cLVM · Synthetic data | contrastive latent structure | Mixed |
| Use a Student's t likelihood instead of a Gaussian likelihood (ablation) | cLVM · Synthetic data with outliers | recovery of target latent representation | Helped |
Closest papers by meaning
- On the Provable Advantage of Unsupervised Pretraining (ICLR 2024 · similarity 0.84)
- Exploring Diffusion Time-steps for Unsupervised Representation Learning (ICLR 2024 · similarity 0.84)
- Probing the Latent Hierarchical Structure of Data via Diffusion Models (ICLR 2025 · similarity 0.83)
- Latent variable model for high-dimensional point process with structured missingness (ICML 2024 · similarity 0.83)
- Machine Unlearning in Low-Dimensional Feature Subspace (NeurIPS 2026 · similarity 0.83)
- Learning Discrete Latent Variable Structures with Tensor Rank Conditions (NeurIPS 2024 · similarity 0.83)
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.