Gaussian Processes

Cross-source consensus on Gaussian Processes from 1 sources and 5 claims.

1 sources · 5 claims

How it works

Variance reduction in the Gaussian Process posterior is identified as the main driver of batch diversity. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
A pseudo-observation reduces posterior variance at the selected point and around it through posterior covariance. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
Gaussian Process posterior conditioning identities provide the canonical implementation of efficient conditioning. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
Acquisition functions such as EI, UCB, and PI are depressed near the selected pseudo-observation because they are monotone in uncertainty for fixed mean. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
Gaussian Process based Kriging Believer achieved performance statistically indistinguishable from joint q-EI on Hartmann-6D. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not