AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

Trading Convergence Rate with Computational Budget in High Dimensional Bayesian Optimization

Hung Tran-The, Sunil Gupta, Santu Rana, Svetha Venkatesh

AAAI 2020 · Probabilistic and scientific modeling › Bayesian optimization · approach: Acquisition function design · 12 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
Optimize the acquisition function over multiple sampled low-dimensional subspaces instead of the full spaceMS-UCB · High-dimensional benchmark functionslog distance to true optimumImproved
Use MS-UCB rather than the compared methodsNeural network parameter search on MNIST · MNIST validation loss, 100 and 500 dimensionsvalidation lossImproved
Use MS-UCB rather than the compared methodsRamp-loss classification on Gisette · Gisette, varying ClossImproved
Increase the number of sampled subspaces by increasing N0 or alpha (ablation)MS-UCB · Ackley and Hyper-Ellipsoid, D=100log regretImproved
Set alpha to 2 instead of 0 with N0=1 (ablation)MS-UCB · Ackley and Hyper-Ellipsoid, D=100log regretImproved
Increase subspace dimension d (ablation)MS-UCB · Ackley and Hyper-Ellipsoid, D=100log regretImproved
Increase subspace dimension d (ablation)MS-UCB · Ackley and Hyper-Ellipsoid, D=100computational costHurt
Use d=5-dimensional subspaces rather than full-space acquisition optimizationMS-UCB · Ackley, Levy, Hyper-Ellipsoid, and Camelbacklog distance to true optimumImproved
Use MS-UCB rather than SREHigh-dimensional benchmark optimization · Camelback, D=50accuracyImproved
Use MS-UCB rather than SREHigh-dimensional benchmark optimization · Camelback, D=100accuracyImproved
Increase the number of subspaces as alpha grows (ablation)MS-UCB · Cumulative regret boundcumulative regret: R_T ≤ O*(√(dTγ_T) + D) if α ≥ D-d-1Improved

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