this is for holding javascript data
Xavier Holt edited Bayesian_Optimisation_over_the_Hyperparameters__.md
over 8 years ago
Commit id: 3137650de90a0644407fecdfa4582a11cc8d7731
deletions | additions
diff --git a/Bayesian_Optimisation_over_the_Hyperparameters__.md b/Bayesian_Optimisation_over_the_Hyperparameters__.md
index b87f5f8..45de7cb 100644
--- a/Bayesian_Optimisation_over_the_Hyperparameters__.md
+++ b/Bayesian_Optimisation_over_the_Hyperparameters__.md
...
\mathbf{x}_{n+1} = \text{argmax}_{\mathbf{x}} \sigma(\mathbf{x}) \left(\gamma(\mathbf{x})\Phi(\gamma(\mathbf{x})) + \mathcal{N}(\gamma(\mathbf{x}), \mathbf{I})\right))
\]
Where \(\gamma( \mathbf{x} ) := \left( f ( \mathbf{x}_\text{best}) - \mu(\mathbf{x})\right) \\ / \\ \sigma(\mathbf{x}) \).
Thanks to the assumption of normality, this value This is solvable
analytically. analytically given Guassian priors \cite{snoek2012practical}.