Research map / Probabilistic and scientific modeling / Variational and generative sampling
General Latent Feature Models for Heterogeneous Datasets
Isabel Valera, Melanie F. Pradier, Maria Lomeli, Zoubin Ghahramani
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 |
|---|---|---|---|
| use GLFM with likelihoods matched to data type instead of BPMF treating all attributes as Gaussian | missing data estimation · Five heterogeneous databases, missing completely at random | average test log-likelihood per missing datum | Improved |
| use GLFM instead of standard linear-Gaussian IBP (SIBP) treating all attributes as Gaussian | latent feature model · Five heterogeneous databases, missing completely at random | average test log-likelihood per missing datum | Mixed |
| vary BPMF latent features among 10, 20 and 50 (ablation) | BPMF · Five heterogeneous databases | average test log-likelihood per missing datum: K=10 for NESARC and wine; K=50 for the remainder | Mixed |
| increase percentage of missing data (ablation) | GLFM, SIBP, BPMF · Five heterogeneous databases | average test log-likelihood per missing datum | Hurt |
| use GLFM instead of Gaussian baselines for each variable type | per-attribute missing data estimation · Five heterogeneous databases, 50% missing | average test log-likelihood per attribute | Improved |
| add the gender attribute as model input | GLFM on NESARC · NESARC psychiatric disorders | number of latent features: three latent features | No clear effect |
| add the age attribute as model input | GLFM on NESARC · NESARC psychiatric disorders | latent features inferred | Mixed |
Closest papers by meaning
- Adaptive Latent Feature Sharing for Piecewise Linear Dimensionality Reduction (JMLR 2024 · similarity 0.86)
- Discriminative Nonparametric Latent Feature Relational Models with Data Augmentation (AAAI 2016 · similarity 0.86)
- Bayesian group factor analysis with structured sparsity (JMLR 2016 · similarity 0.85)
- Latent Discriminant Analysis with Representative Feature Discovery (AAAI 2017 · similarity 0.85)
- Identifying Metric Structures of Deep Latent Variable Models (ICML 2025 · similarity 0.84)
- Identification of Latent Confounders via Investigating the Tensor Ranks of the Nonlinear Observations (ICML 2025 · 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.
