AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

For high-dimensional hierarchical models, consider exchangeability of effects across covariates instead of across datasets

Brian L. Trippe, Hilary Finucane, Tamara Broderick

NeurIPS 2021 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Nonparametric and PAC-Bayes Bayesian learning · 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
model effects as exchangeable across covariates (ECov) instead of across datasets (EData)hierarchical Bayesian regression · simulation, Q=10, D > Qparameter estimation errorImproved
model effects as exchangeable across datasets (EData) when datasets outnumber covariateshierarchical Bayesian regression · simulation, Q=10, D < Qparameter estimation errorImproved
estimate Sigma by maximum marginal likelihood (PSD-constrained) instead of moment matchingECov and EData · simulationparameter estimation errorImproved
share strength across datasets via ECov rather than fitting each dataset independently (ID)ECov · simulation with independent effectsparameter estimation errorNo clear effect
ECov vs least squares and EData when effects are independent across datasetsECov · simulation, high Dparameter estimation errorImproved
use ECov instead of EData, independent Bayes and least squaresECov · law enforcement, blog posts, CIFAR10held-out prediction errorImproved
model dataset heterogeneity instead of pooling all datasetsECov · three real applicationsheld-out prediction errorImproved
ECov moment-matching estimator replaces least squares when D > 2Q+2least squares · theory, orthogonal designsquared error riskImproved
EData moment-matching estimator versus least squares when D is large relative to Qleast squares · theory, orthogonal designsquared error riskHurt
replace moment-matching with maximum marginal likelihood (positive-part shrinkage) for ECovECov · theory, D > Q+1squared error riskImproved
use a Sigma estimate with factor 1/(D-Q-1) instead of the maximum marginal likelihood oneECov positive-part estimator · unreported experimentsestimation errorHurt

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