Research map / Probabilistic and scientific modeling / Bayesian optimization
Bayesian Functional Optimization
Ngo Anh Vien, Heiko Zimmermann, Marc Toussaint
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 |
|---|---|---|---|
| optimize directly in RKHS function space (BFO EI) instead of parametric standard BayesOpt EI | BayesOpt · Synthetic 1D | MSE: 0.0197 ±0.0070 -> 0.0027 ±0.0014 | Improved |
| BFO EI vs standard BayesOpt EI | BayesOpt · Synthetic 2D | MSE: 0.0121 ±0.002 -> 0.0056 ±0.003 | Improved |
| BFO instead of random-embedding RKHS-REMBO (UCB) | RKHS-REMBO · Synthetic 1D | MSE: 0.0780 ±4.58e-6 -> 0.0086 ±0.0036 | Improved |
| BFO acquisition choice: PI/EI instead of UCB | BFO · Synthetic 1D | MSE: 0.0086 -> 0.0026 (PI), 0.0027 (EI) | Improved |
| tune activation function nonparametrically with BFO EI instead of fixed sigmoid | MLP on MNIST · MNIST | validation cross-entropy: 0.112 ± 0.006 -> 0.053± 0.002 | Improved |
| BFO EI instead of standard GP-UCB BayesOpt for activation tuning | MLP on MNIST · MNIST | test accuracy (%): 97.59% -> 98.26% | Improved |
| BFO EI instead of joint training of tunable activation | MLP on MNIST · MNIST | test accuracy (%): 97.77% -> 98.26% | Improved |
| iGP-UCB policy search vs standard parametric BayesOpt | inverted pendulum policy · Inverted Pendulum | discounted cumulative reward | No clear effect |
Closest papers by meaning
- Learning search spaces for Bayesian optimization: Another view of hyperparameter transfer learning (NeurIPS 2019 · similarity 0.89)
- Bayesian Optimization on Function Spaces via Sparse RKHS Manifolds (NeurIPS 2026 · similarity 0.89)
- BayeSQP: Bayesian Optimization through Sequential Quadratic Programming (NeurIPS 2025 · similarity 0.89)
- FunBO: Discovering Acquisition Functions for Bayesian Optimization with FunSearch (ICML 2025 · similarity 0.88)
- A General Framework for User-Guided Bayesian Optimization (ICLR 2024 · similarity 0.88)
- Thompson Sampling in Function Spaces via Neural Operators (NeurIPS 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.
