Research map / Probabilistic and scientific modeling / Variational and generative sampling
Scalable Random Feature Latent Variable Models
Ying Li, Zhidi Lin, Yuhao Liu, Michael Minyi Zhang, Pablo M. Olmos, Petar M. Djurić
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 |
|---|---|---|---|
| replace MCMC inference with block coordinate descent variational inference | G-RFLVM · cifar-10 wall-time | wall-time (s): 9011.504 -> 130.710 | Improved |
| replace MCMC inference with block coordinate descent variational inference | G-RFLVM · mnist wall-time | wall-time (s): 9151.876 -> 111.560 | Improved |
| replace MCMC with variational inference on accuracy | G-RFLVM · mnist KNN accuracy | KNN accuracy: 0.602 -> 0.584 | Hurt |
| replace MCMC with variational inference on accuracy | G-RFLVM · cifar-10 KNN accuracy | KNN accuracy: 0.284 -> 0.279 | No clear effect |
| use negative binomial instead of Gaussian likelihood | SRFLVM · montreal KNN accuracy | KNN accuracy: 0.744 -> 0.781 | Improved |
| use SRFLVM instead of RFLVM for imputation with 60% missing | latent variable model · CIFAR imputation | test MSE: 0.127 -> 0.048 | Improved |
| use SRFLVM instead of BGPLVM for imputation with 60% missing | latent variable model · CIFAR imputation | test MSE: 0.119 -> 0.048 | Improved |
| use GP-based SRFLVM instead of VAE on f-Mnist | G-OURS · f-Mnist larger dataset | KNN accuracy: 0.783 -> 0.740 | Hurt |
| use G-OURS instead of VAE on f-Cifar | VAE · f-Cifar larger dataset | KNN accuracy: 0.190 -> 0.199 | Improved |
| use G-OURS instead of VAE on mnist small dataset | VAE · mnist KNN accuracy | KNN accuracy: 0.643 -> 0.584 | Hurt |
Closest papers by meaning
- Preventing Model Collapse in Gaussian Process Latent Variable Models (ICML 2024 · similarity 0.88)
- Diversity-Promoting Bayesian Learning of Latent Variable Models (ICML 2016 · similarity 0.87)
- Variational Inference for Sequential Data with Future Likelihood Estimates (ICML 2020 · similarity 0.87)
- Quasi-Monte Carlo Methods Enable Extremely Low-Dimensional Deep Generative Models (ICLR 2026 · similarity 0.87)
- Multi-View Oriented GPLVM: Expressiveness and Efficiency (NeurIPS 2025 · similarity 0.86)
- Learning Latent Variable Models via Jarzynski-adjusted Langevin Algorithm (NeurIPS 2025 · 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.
