Optimizing a cellular network using a machine learning model

A method performed by a network node, comprising:

• generating models of a plurality of cellular networks;
• simulating radio propagation for a number of user equipment (UEs) of a model of a cellular network, of the models of the plurality of cellular networks;
• defining one or more channel interference (CI) indicators based at least in part on the simulated radio propagation;
• and
• generating a training set including one or more training samples informative of association between:
• data characterizing a given cell and one or more neighbor cells of the given cell, and
• one or more values of the one or more CI indicators for one or more cell pairs including the given cell and the one or more neighbor cells of the given cell.