Modification of neural network topology

A system comprising:

• one or more processors configured to perform or control performance of operations comprising:
• obtaining data representative of a neural network topology (NNT) of a graph-based model, the NNT comprising multiple components that include multiple neural nodes and at least one connection that associates two or more of the neural nodes;
• displaying, in a graphical user interface (GUI) via a display screen, the NNT including the multiple components;
• receiving, via the GUI, user input effective to indicate that at least one of the components of the NNT is to be modified;
• modifying the at least one of the components of the NNT based on the user input;
• displaying, via the GUI, the NNT that comprises the modification of the at least one of the components;
• obtaining a validation dataset;
• generating validation machine learning results based on machine learning performed using the NNT that comprises the modification of the at least one of the components and the validation dataset;
• generating comparison machine learning results based on machine learning performed using a pre-selected NNT and the validation dataset;
• and
• validating the NNT that comprises the modification of the at least one of the components by comparing the validation machine learning results with the comparison machine learning results.

A non-transitory computer-readable storage medium having computer-executable instructions stored thereon that are executable by a processor device to perform or control performance of operations comprising:

• obtaining data representative of a neural network topology (NNT) of a graph-based model, the NNT comprising multiple components that include multiple neural nodes and at least one connection that associates two or more of the neural nodes;
• displaying, in a graphical user interface (GUI) via a display screen, the NNT including the multiple components;
• receiving, via the GUI, user input effective to indicate that at least one of the components of the NNT is to be modified;
• modifying the at least one of the components of the NNT based on the user input;
• determining whether the modification of the at least one of the components is against one or more source rules of a source network, wherein:
• responsive to the modification of the at least one of the components not being against one or more of the source rules, automatically proceeding with the modification of the at least one of the components;
• or
• responsive to the modification of the at least one of the components being against one or more of the source rules, providing a notice via the GUI that includes information related to the one or more source rules that the modification of the at least one of the components will be against;
• and
• displaying, via the GUI, the NNT that comprises the modification of the at least one of the components.