AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

Black-Box Optimization with Local Generative Surrogates

Sergey Shirobokov, Vladislav Belavin, Michael Kagan, Andrei Ustyuzhanin, Atilim Gunes Baydin

NeurIPS 2020 · Probabilistic and scientific modeling › Bayesian optimization · approach: Offline black-box optimization with surrogates · 24 citations (OpenAlex)

Read the paper

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 L-GSO instead of BOMagnet geometry optimization · Physics experimentobjective value: ~3000 -> ~2200; approximately 25% smallerImproved
Use L-GSO instead of BOMagnet geometry optimization · Physics experimentMagnet length (m): 35.44 -> 33.39Improved
Use L-GSO instead of BOMagnet geometry optimization · Physics experimentMagnet weight (kt): 1.27 -> 1.05Improved
Use L-GSO rather than score function-based optimizationBlack-box simulator optimization · Toy experimentssimulator calls to converge: approximately an order of magnitudeImproved
Use local generative surrogates on parameters lying on a low-dimensional submanifoldL-GSO · Submanifold optimization problemssimulator calls to convergeImproved
Train each local surrogate with fewer parameter points than the full parameter-space dimension (ablation)L-GSO · Submanifold problemssimulator calls to convergeImproved
Reuse previously sampled simulator data in local surrogate training (ablation)L-GSO · Optimization experimentssimulator calls to attain the minimumImproved
Use BO rather than L-GSOLow-dimensional Three Hump problem · Three Hump problemsimulator calls to convergeImproved
Use BO on a high-curvature objectiveBayesian optimization · Rosenbrock problemobjective valueHurt

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