1: \begin{abstract}

2:   Evaluating the channel capacity is one of many key problems in

3:   information theory. In this work we derive rather-mild sufficient

4:   conditions under which the capacity of continuous channels is {\em

5:     finite\/} and {\em achievable\/}.

6:  

7:   These conditions are derived for generic, memoryless and possibly

8:   non-linear additive noise channels. The results are based on a novel

9:   sufficient condition that guarantees the convergence of differential

10:   entropies under point-wise convergence of probability density

11:   functions.

12:  

13:   Perhaps surprisingly, the finiteness of channel capacity holds for

14:   the majority of setups, including those where inputs and outputs

15:   have possibly infinite second-moments.

16: \end{abstract}

17: