MQ-RBF Networks

Uses

How it works

Risks & contraindications

Comparisons

Evidence quality

Highlighted claims

• MQ-RBF networks with data-dependent centers also satisfy efficient conditioning. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
• The MQ-RBF power function is a deterministic worst-case bound rather than a probabilistic posterior variance. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
• MQ-RBF produces diverse batches in the Structural Diversity Diagnostic. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
• MQ-RBF uncertainty is useful for structural diversity but is not a calibrated probabilistic uncertainty quantifier. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not
• Efficient conditioning for MQ-RBF opens a path toward non-probabilistic surrogate alternatives for principled batch selection. — Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not