Generation of hotspot-containing physical design layout patterns

A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by at least one computing device to cause the at least one computing device to perform steps of:

• selecting as training data a set of physical design layout patterns of patterned structures;
• training, utilizing a first subset of the training data comprising physical design layout patterns containing hotspots, a first neural network model configured to generate synthetic physical design layout patterns;
• training, utilizing a second subset of the training data comprising at least one physical design layout pattern containing a hotspot and at least one physical design layout pattern not containing a hotspot, a second neural network model configured to classify whether physical design layout patterns contain hotspots;
• generating one or more synthetic physical design layout patterns containing hotspots, wherein generating the one or more synthetic physical design layout patterns containing hotspots comprises utilizing the trained first neural network model to generate a set of synthetic physical design layout patterns and utilizing the trained second neural network model to select one or more of the set of synthetic physical design layout patterns as the one or more synthetic physical design layout patterns containing hotspots;
• and
• evaluating manufacturability of a given patterned structure comprising at least one of the generated synthetic physical design layout patterns containing at least one hotspot.

An apparatus comprising:

• a memory;
• and
• at least one processor coupled to the memory and configured for:
• selecting as training data a set of physical design layout patterns of patterned structures;
• training, utilizing a first subset of the training data comprising physical design layout patterns containing hotspots, a first neural network model configured to generate synthetic physical design layout patterns;
• training, utilizing a second subset of the training data comprising at least one physical design layout pattern containing a hotspot and at least one physical design layout pattern not containing a hotspot, a second neural network model configured to classify whether physical design layout patterns contain hotspots;
• generating one or more synthetic physical design layout patterns containing hotspots, wherein generating the one or more synthetic physical design layout patterns containing hotspots comprises utilizing the trained first neural network model to generate a set of synthetic physical design layout patterns and utilizing the trained second neural network model to select one or more of the set of synthetic physical design layout patterns as the one or more synthetic physical design layout patterns containing hotspots;
• and
• evaluating manufacturability of a given patterned structure comprising at least one of the generated synthetic physical design layout patterns containing at least one hotspot.