Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
On the Expressiveness of Approximate Inference in 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 active learning instead of random selection | Infinite-width GP · Naval, 1 HL | RMSE: $0.12 \pm 0.01$ -> $0.04 \pm 0.00$ | Helped |
| Use active learning instead of random selection | Infinite-width GP · Naval, 2 HL | RMSE: $0.13 \pm 0.01$ -> $0.04 \pm 0.00$ | Helped |
| Use active learning instead of random selection | Infinite-width GP · Naval, 3 HL | RMSE: $0.15 \pm 0.01$ -> $0.04 \pm 0.00$ | Helped |
| Use active learning instead of random selection | Infinite-width GP · Naval, 4 HL | RMSE: $0.16 \pm 0.01$ -> $0.05 \pm 0.00$ | Helped |
| Use active learning instead of random selection | MFVI BNN · Naval, 1 HL | RMSE: $0.15 \pm 0.01$ -> $0.94 \pm 0.11$ | Hurt |
| Use active learning instead of random selection | MFVI BNN · Naval, 2 HL | RMSE: $0.23 \pm 0.01$ -> $0.46 \pm 0.04$ | Hurt |
| Use active learning instead of random selection | MFVI BNN · Naval, 3 HL | RMSE: $0.28 \pm 0.01$ -> $0.35 \pm 0.03$ | Hurt |
| Use active learning instead of random selection | MFVI BNN · Naval, 4 HL | RMSE: $0.32 \pm 0.01$ -> $0.31 \pm 0.02$ | No clear effect |
| Use active learning instead of random selection | MCDO BNN · Naval, 1 HL | RMSE: $0.22 \pm 0.01$ -> $0.69 \pm 0.04$ | Hurt |
| Use active learning instead of random selection | MCDO BNN · Naval, 2 HL | RMSE: $0.35 \pm 0.01$ -> $0.36 \pm 0.02$ | No clear effect |
| Use active learning instead of random selection | MCDO BNN · Naval, 3 HL | RMSE: $0.43 \pm 0.01$ -> $0.38 \pm 0.02$ | Helped |
| Use active learning instead of random selection | MCDO BNN · Naval, 4 HL | RMSE: $0.47 \pm 0.02$ -> $0.45 \pm 0.02$ | No clear effect |
Closest papers by meaning
- Deterministic Variational Inference for Robust Bayesian Neural Networks (ICLR 2019 · similarity 0.92)
- Liberty or Depth: Deep Bayesian Neural Nets Do Not Need Complex Weight Posterior Approximations (NeurIPS 2020 · similarity 0.92)
- Do Bayesian Neural Networks Actually Behave Like Bayesian Models? (ICML 2025 · similarity 0.92)
- VIKING: Deep variational inference with stochastic projections (NeurIPS 2025 · similarity 0.91)
- A Study of Bayesian Neural Network Surrogates for Bayesian Optimization (ICLR 2024 · similarity 0.89)
- Reparameterized Importance Sampling for Robust Variational Bayesian Neural Networks (ICML 2024 · 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.