Research map / Probabilistic and scientific modeling / Variational and generative sampling
Variational Inference for Latent Variables and Uncertain Inputs in Gaussian Processes
Andreas C. Damianou, Michalis K. Titsias, Neil D. Lawrence
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 temporal dynamics prior (exp. quadratic) to variational GP-LVM (ablation) | VGP-LVM · Motion capture body (L) | angle-space RMSE: 2.79 -> 1.90 | Improved |
| add temporal dynamics prior (exp. quadratic) to variational GP-LVM (ablation) | VGP-LVM · Motion capture legs (L) | angle-space RMSE: 5.09 -> 3.28 | Improved |
| use Matern 3/2 instead of exponentiated quadratic dynamics prior (ablation) | Dyn. VGP-LVM · Motion capture legs (L) | scaled-space error: 7.76 -> 6.84 | Improved |
| use exponentiated quadratic instead of Matern 3/2 dynamics prior (ablation) | Dyn. VGP-LVM · Motion capture body (B) | scaled-space error: 13.93 -> 11.95 | Improved |
| variational non-dynamical GP-LVM versus nearest neighbour | VGP-LVM · Motion capture legs (L) | scaled-space error: 14.22 -> 14.0 | No clear effect |
| variational (marginalised latents) versus MAP GP-LVM with dynamics-free q=4 | VGP-LVM · Motion capture legs (L) | scaled-space error: 9.7 -> 14.22 | Hurt |
| add variational marginalisation of latents with ARD, 10 latent dims, versus sparse GP-LVM with q=2 | Variational GP-LVM · Oil flow | nearest neighbour errors: 26 -> 1 | Improved |
| variational GP-LVM versus full GP-LVM with ARD kernel | Variational GP-LVM · Oil flow | nearest neighbour errors: 8 -> 1 | Improved |
| variational GP-LVM versus full GP-LVM with EQ kernel | Variational GP-LVM · Oil flow | nearest neighbour errors: 2 -> 1 | Improved |
Closest papers by meaning
- Generic Inference in Latent Gaussian Process Models (JMLR 2019 · similarity 0.94)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.92)
- Preventing Model Collapse in Gaussian Process Latent Variable Models (ICML 2024 · similarity 0.91)
- Neighbour-Driven Gaussian Process Variational Autoencoders for Scalable Structured Latent Modelling (ICML 2025 · similarity 0.91)
- Bayesian Basis Function Approximation for Scalable Gaussian Process Priors in Deep Generative Models (ICML 2025 · similarity 0.90)
- New Bounds for Sparse Variational Gaussian Processes (ICML 2025 · 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.
