Deep learning based prediction and correction of fabrication-process-induced structural variations in nanophotonic devices

A computer-implemented method comprising the steps of:

• with an imaging device, acquiring a plurality of images of at least one fabricated structure associated with a device, wherein the at least one fabricated structure comprises structural features;
• preprocessing the plurality of images;
• creating at least one image dataset from the preprocessed plurality of images;
• generating a corrector model;
• training the corrector model with the at least one image dataset to automatically correct the device design to minimize the impact of fabrication deviations.

A neural network unit comprising:

• at least one processing unit;
• and
• a non-transitory memory communicatively coupled to the at least one processing unit and comprising computer-readable program instructions that when executed by the at least one processing unit, cause the neural network unit to perform operations including:
• training the neural network unit associated with at least one of a predictive model and a corrective model using a plurality of datasets associated with pairs of input layout images and their corresponding acquired images, the neural network unit comprising at least one fully connected layer comprising a plurality of input nodes, a plurality of output nodes, and a plurality of connections for connecting each one of the plurality of input nodes to each one of the plurality of output nodes;
• inputting the plurality of datasets into the neural network using the plurality of input nodes;
• extracting at least one feature of associated with the input layout images and the acquired images and in accordance with one or more pre-programmed functions to learn a relationship between the input layout images and the corresponding acquired images,
• wherein the predictive model is useful for completing tasks comprising of at least:
• identifying structural features of the fabricated device with a propensity for fabrication deviations;
• predicting planar fabrication deviations in silicon photonic devices;
• predicting fabrication deviations validating pre-fabrication correction;
• minimizing the impact of fabrication deviations;
• and
• determining manufacturing feasibility of the device based on the predicted fabrication deviations;
• wherein the corrector model is useful for completing tasks comprising of at least:
• receiving a desired layout as input and generating a layout output, and automatically correcting the device design to minimize fabrication deviations.