chainerx.gaussian_kl_divergence¶

chainerx.gaussian_kl_divergence()¶

Element-wise KL-divergence of Gaussian variables from the standard one.

Given two variable mean representing \(\mu\) and ln_var representing \(\log(\sigma^2)\), this function calculates the element-wise KL-divergence between the given multi-dimensional Gaussian \(N(\mu, S)\) and the standard Gaussian \(N(0, I)\)

\[D_{\mathbf{KL}}(N(\mu, S) \| N(0, I)),\]

where \(S\) is a diagonal matrix such that \(S_{ii} = \sigma_i^2\) and \(I\) is an identity matrix.

Parameters

mean (ndarray) – A variable representing mean of given gaussian distribution, \(\mu\).
ln_var (ndarray) – A variable representing logarithm of variance of given gaussian distribution, \(\log(\sigma^2)\).

Returns

A variable representing KL-divergence between given gaussian distribution and the standard gaussian.

Return type

ndarray