AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

Constrained Causal Bayesian Optimization

Virginia Aglietti, Alan Malek, Ira Ktena, Silvia Chiappa

ICML 2023 · Probabilistic and scientific modeling › Bayesian optimization · approach: Acquisition function design · 1 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 observational data to construct GP prior parameterscCBO · Synthetic-1 and Synthetic-2, N_O=500convergence speedImproved
Capture covariance between target and constraint effectscCBO with observational-data priors · Synthetic-1, N_O=500convergence speedImproved
Use the SCM structure to model effect covariancecCBO with G-mtgp · Synthetic-2, N_O=500feasible interventions collectedImproved
Use observational-data priors when the observational sample is smallcCBO with G-mtgp · Health, N_O=10convergenceHurt
Use observational-data priors when the observational sample is smallcCBO with mtgp+ · Synthetic-2, N_O=100 and N_O=10convergence speedHurt
Use mtgp+ rather than G-mtgp when the observational SCM fit is inaccuratecCBO · Synthetic-1, N_O=100 and N_O=10convergence speedImproved
Use mtgp+ rather than observational-data-based priors with a very small observational samplecCBO · Synthetic-2, N_O=100 and N_O=10convergence speedImproved
Use G-mtgp rather than cBOconstrained optimization · Health, N_O=100convergence speedImproved
Use G-mtgp rather than cBOconstrained optimization · Health, N_O=100feasible interventions collectedImproved
Intervene on all variables instead of selecting causally relevant intervention setscBO · Synthetic-1target effectHurt
Model target and constraint effects independently rather than capturing their correlation (ablation)cCBO with stgp+ · Synthetic-2, N_O=500feasible interventions collectedHurt
Use observational-data priors rather than zero-mean GP priors (ablation)cCBO · Synthetic-1 and Synthetic-2, N_O=500convergence speedImproved

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