訓練データの生成モデル
\(
\begin{align}
y_{data} &= \boldsymbol{w}' \cdot \boldsymbol{\phi}' (x_{data}) + {\cal N}(0, \sigma) \\
&= \sum_{i = 0}^{M' - 1} w'_i \phi' _i (x_{data}) + {\cal N}(0, \sigma)
\end{align}
\)
\(
\begin{align}
\boldsymbol{w}' &= (w_0, w_1, \cdots, w_{M'-1}) \\
\boldsymbol{\phi}'(x) &= (1, x, \cdots, x^{M'-1}) \\
{\cal N}(0, \sigma) &= \frac{1}{\sqrt{2\pi \sigma^2}} \exp\left(-\frac{x^2}{2\sigma^2} \right)
\end{align}
\)