Research map / Probabilistic and scientific modeling / Variational and generative sampling
Functional Variational Inference based on Stochastic Process Generators
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 FVI instead of Bayes-by-Backprop | BNN prior · UCI Boston | test NLL: 2.76±0.04 -> 2.33±0.04 | Helped |
| use FVI instead of VIP-BNN | BNN prior · UCI Naval | test NLL: -5.62±0.04 -> -7.21±0.06 | Helped |
| use FVI instead of VIP-BNN | BNN prior · UCI Yacht | test NLL: -0.02±0.07 -> 0.59±0.11 | Hurt |
| use FVI instead of VIP-BNN | BNN prior · UCI Energy | test NLL: 0.56±0.04 -> 0.58±0.05 | No clear effect |
| use FVI instead of VIP-BNN | BNN prior · UCI regression mean rank | mean rank: 2.11 -> 1.33 | Helped |
| use FVI instead of f-BNN | BNN prior · UCI regression | test NLL | Helped |
| use FVI instead of f-BNN | 2x50 network · Wheel bandit | relative cumulative regret: 77.70 ± 21.2 -> 41.44 ± 9.28 | Helped |
| use FVI instead of f-BNN | 2x50 network · Covertype bandit | relative cumulative regret: 50.10 ± 5.70 -> 49.59 ± 1.61 | No clear effect |
| use FVI instead of f-BNN | 2x50 network · Census bandit | relative cumulative regret: 51.22 ± 2.55 -> 51.77 ± 3.06 | Hurt |
| use FVI instead of BBB 2x50 | BNN · Mushroom bandit | relative cumulative regret: 23.18 ± 5.90 -> 16.46 ± 2.04 | Helped |
| use FVI instead of MFVI | Bayesian CNN · CIFAR10 | OOD-AUC: 0.589±0.01 -> 0.883±0.04 | Helped |
| use FVI instead of MFVI | Bayesian CNN · FMNIST | NLL: 0.343±0.01 -> 0.254±0.05 | Helped |
Closest papers by meaning
- Bridging the Gap between Variational Inference and Stochastic Gradient MCMC in Function Space (ICLR 2025 · similarity 0.90)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.87)
- Sparse Variational Student-t Processes (AAAI 2024 · similarity 0.86)
- Bayesian Optimization through Gaussian Cox Process Models for Spatio-temporal Data (ICLR 2024 · similarity 0.86)
- Functional Gradient Flows for Constrained Sampling (NeurIPS 2024 · similarity 0.86)
- Least squares variational inference (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.