The Network & System Wide Attack Assessment (N&SWAA), uses deep learning with an Associated Random Neural Network (ARNN). It assesses the overall security of a networked system by identifying the nodes that have been compromised by a cyberattack and the system software components that are vulnerable to attacks. The original ARNN architecture, is comprised of competing neuron pairs, each pair being associated to each network node and software component. Each pair of neurons computes the probability of being compromised for the corresponding node or software component. The computation is based on the local traffic of the nodes, their interconnections, the software components and their inter-relations, and the links between nodes and software components. The decisions are computed collectively by all competing neuron pairs, either for network nodes and software components taken separately, by all competing neuron pairs, or together for network nodes and software components, including all interconnections and all neurons in the ARNN architecture.