AI Paper Atlas

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

Mixtures of Gaussian Process Experts with SMC^2

Teemu Härkönen, Sara Wade, Kody Law, Lassi Roininen

JMLR 2025 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Structured Gaussian process models

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
replace importance sampling with SMC^2 inference (alpha=0.1)mixture of GP experts · Synthetic 2: discontinuous, heteroskedastic noiseL1 distance: 993.265 -> 509.876Improved
replace importance sampling with SMC^2 inference (alpha=0.1)mixture of GP experts · Synthetic 2: discontinuous, heteroskedastic noisepredictive log-likelihood: -258.943 -> -224.955Improved
replace importance sampling with SMC^2 inference (alpha=0.1)mixture of GP experts · Synthetic 1: non-stationary, homoskedastic noiseL1 distance: 588.068 -> 375.731Improved
replace importance sampling with SMC^2 inference (alpha=0.1)mixture of GP experts · Synthetic 1: non-stationary, homoskedastic noisepredictive log-likelihood: -293.514 -> -296.381Hurt
use SMC^2-MoE (alpha=0.1) instead of single GPGP regression · Synthetic 2: discontinuous, heteroskedastic noiseL1 distance: 1078.485 -> 509.876Improved
use SMC^2-MoE (alpha=0.1) instead of single GPGP regression · Synthetic 1: non-stationary, homoskedastic noiseL1 distance: 373.676 -> 375.731No clear effect
use SMC^2-MoE (alpha=0.1) instead of TreedGPmixture of GP experts · Synthetic 1: non-stationary, homoskedastic noiseL1 distance: 363.797 -> 375.731Hurt
use SMC^2-MoE (alpha=0.1) instead of TreedGPmixture of GP experts · Synthetic 2: discontinuous, heteroskedastic noiseL1 distance: 655.436 -> 509.876Improved
increase Dirichlet concentration alpha from 0.1 to K/2 (ablation)SMC^2-MoE · Synthetic 2: discontinuous, heteroskedastic noiseL1 distance: 509.876 -> 558.190Hurt
increase Dirichlet concentration alpha from 0.1 to K/2 (ablation)SMC^2-MoE · Synthetic 1: non-stationary, homoskedastic noiseL1 distance: 375.731 -> 403.566Hurt
increase Dirichlet concentration alpha from 0.1 to K/2 (ablation)SMC^2-MoE · computational costcostImproved
use SMC^2-MoE (alpha=0.1) instead of RBCMGP regression · Synthetic 2: discontinuous, heteroskedastic noiseL1 distance: 1217.425 -> 509.876Improved

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