Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
FUNCTIONAL VARIATIONAL BAYESIAN NEURAL NETWORKS
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 function-space variational inference with sampling-based measurement sets rather than weight-space BBB | Bayesian neural networks · UCI small-scale regression | Test RMSE | Mixed |
| Use function-space variational inference with sampling-based measurement sets rather than weight-space BBB | Bayesian neural networks · UCI large-scale regression | Test RMSE | Helped |
| Use function-space variational inference with sampling-based measurement sets rather than weight-space BBB | Bayesian neural networks · UCI large-scale regression | Test log-likelihood | Mixed |
| Add a structured periodic GP functional prior | FBNN · Periodic-function extrapolation | Extrapolation quality | Helped |
| Use adversarially optimized measurement sets instead of sampled sets (ablation) | fBNN · Generalization | Extrapolation quality | Hurt |
| Add stochastic activations to the variational posterior (ablation) | fBNN · Experiments | Predictive performance | No clear effect |
| Increase BBB hidden size from 50 to 500 (ablation) | BBB · Contextual bandits | Cumulative regret relative to Uniform | Hurt |
| Increase fBNN depth or width (ablation) | FBNN · Contextual bandits | Cumulative regret relative to Uniform | No clear effect |
Closest papers by meaning
- Deterministic Variational Inference for Robust Bayesian Neural Networks (ICLR 2019 · similarity 0.92)
- Collapsed Variational Bounds for Bayesian Neural Networks (NeurIPS 2021 · similarity 0.91)
- Globally Convergent Variational Inference (NeurIPS 2024 · similarity 0.89)
- Variational Bayesian Last Layers (ICLR 2024 · similarity 0.89)
- Bayesian Online Natural Gradient (BONG) (NeurIPS 2024 · similarity 0.88)
- Bayesian Optimization via Continual Variational Last Layer Training (ICLR 2025 · similarity 0.88)
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.