1: \begin{definition}

2: \label{def:evalrelax}

3: Suppose a processing chain as in Definition~\ref{def:processing-chain} is given and let $f_j\colon M_{j - 1}^{c_j}\to N_j$ be the unique functions with $T_j = \Slide_{f_j}$ for all $j\inint{1}{L}$ due to Theorem~\ref{thm:sliding-subsignal}.

4: Then for $j\inint{0}{L}$, the operator $\EvalRelax_j\colon\cup_\ROI(M_0)\to\cup_1(M_j)$,

5: \begin{displaymath}

6:   \xi\mapsto

7:   \begin{cases}

8:     \xi\text{,} & \text{if } j = 0\text{,}\\

9:     \Stride_{g_j}(\Slide_{f_j}(\EvalRelax_{j - 1}(\xi)))\text{, } & \text{if } j > 0\text{,}

10:   \end{cases}

11: \end{displaymath}

12: applies the processing chain in a \emph{relaxed} fashion.

13: \end{definition}

14: