Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Doubly Stochastic Variational Inference for Neural Processes with Hierarchical Latent Variables
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 hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · Cart-Pole state transition test | MSE: 0.067 -> 0.036 | Helped |
| Use hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · Cart-Pole state transition test | NLL: -1.821 -> -2.145 | Helped |
| Use hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · Sarcos | MSE: 1.215 -> 0.839 | Helped |
| Use hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · WQ | MSE: 0.007 -> 0.006 | Helped |
| Use hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · SCM20d | MSE: 0.015 -> 0.007 | Helped |
| Use hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · CIFAR10 | accuracy: 0.834 -> 0.863 | Helped |
| Use hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · Synthetic stochastic process interpolation | NLL: -1.149 -> -0.975 | Hurt |
| Use hierarchical global and target-specific local latent variables instead of a single global latent | Neural Processes · Synthetic stochastic process extrapolation | NLL: 8.091 -> 4.203 | Helped |
| Add deterministic attention-based local context | Neural Processes · CIFAR10 out-of-distribution detection | entropy: CIFAR10: 0.124 -> 0.010 (AttnNP); Rademacher: 0.078 -> 0.010 | Hurt |
| Increase the local-latent KL penalty (β₂=1000) (ablation) | DSVNP · 1-D stochastic process | prediction accuracy | Hurt |
Closest papers by meaning
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.88)
- Bootstrapping neural processes (NeurIPS 2020 · similarity 0.88)
- Doubly Stochastic Variational Inference for Deep Gaussian Processes (NeurIPS 2017 · similarity 0.88)
- Flow Matching Neural Processes (NeurIPS 2025 · similarity 0.86)
- Rényi Neural Processes (ICML 2025 · similarity 0.86)
- Distance-informed Neural Processes (NeurIPS 2025 · similarity 0.86)
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.