AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

Maximizing acquisition functions for Bayesian optimization

James Wilson, Frank Hutter, Marc Deisenroth

NeurIPS 2018 · Probabilistic and scientific modeling › Bayesian optimization · approach: Acquisition function design · 65 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use gradient-based stochastic optimization (Adam) instead of gradient-free or random-search maximizationParallel Monte Carlo EI on synthetic and black-box tasks · Synthetic and black-box tasksBO performanceImproved
Use greedy parallel maximization instead of joint maximizationMyopic maximal acquisition functions · Synthetic and black-box tasksBO performanceImproved
Increase acquisition dimensionalityGradient-based and submodular acquisition maximizers · Synthetic and black-box tasksBO performance gainImproved
Combine gradient-based optimization with greedy submodular maximizationParallel acquisition maximization · Lower-dimensional casesBO performanceNo clear effect
Use parallel UCB instead of parallel EIBayesian optimization · All tasks except LevyBO performanceImproved
Use parallel UCB instead of parallel EIBayesian optimization · LevyBO performanceMixed
Use incremental rather than joint parallel EIGreedy parallel EI · Higher-dimensional acquisition surfacesBO performanceImproved
Initialize query sets by sampling from the marginal acquisition (ablation)Query optimizers · Experiments across tasksBO performanceImproved
Fantasize pending-query outcomes at their predictive mean (ablation)Greedy parallel acquisition maximization · Experiments across tasksBO performanceNo clear effect
Use stochastic rather than deterministic minibatches (ablation)Adam and CMA-ES acquisition maximization · Acquisition optimization experimentsAcquisition maximization performanceImproved
Use Adam rather than L-BFGS-B or Polyak averaging (ablation)Gradient-based acquisition maximization · Experiments across tasksAcquisition maximization performanceImproved
Use CMA-ES rather than DIRECT (ablation)Black-box acquisition maximization · Experiments across tasksAcquisition maximization performanceImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy