A monotonically increasing function is a function that as x gets bigger so does its output. So, if plotted, it will never go down. Although the outputs might stay constant.

Logically this seems like an easier function to learn when compared to something that can, when plotted, go up or down.

Wikipedia has some example diagrams on monotonic functions.

If I were to say that it is easier for a neural network to learn a monotonic function compared to a non-monotonic function would the statement be correct? If so, is there any reason to it other than 'it only goes one way'?

  • 2
    $\begingroup$ This paper Monotonic Networks (1997), by Joseph Sill, may be of interest to you. $\endgroup$
    – nbro
    Jan 14, 2021 at 20:13


You must log in to answer this question.