Back to results
US 11,768,440 B2Granted
Training methods for machine learning assisted optical proximity error correction
AssigneeASML Netherlands B.V.
InventorsJing Su, Yen-Wen Lu, Ya Luo
Life of this patent
Estimating the term…
Abstract
A method including: obtaining data based an optical proximity correction for a spatially shifted version of a training design pattern; and training a machine learning model configured to predict optical proximity corrections for design patterns using data regarding the training design pattern and the data based on the optical proximity correction for the spatially shifted version of the training design pattern.
Claims11
A method comprising:
- obtaining a weighting function or data to assign a different weight to a region of an optical proximity correction of a design pattern than another region thereof;
- and
- training, by a hardware computer system, a machine learning model configured to predict optical proximity corrections for design patterns by evaluating a predicted optical proximity correction of the design pattern by the machine learning model against a benchmark optical proximity correction of the design pattern based on the weighting data.
A method comprising:
- obtaining a benchmark optical proximity correction for a design pattern;
- generating, by a hardware computer system, a predicted optical proximity correction for the design pattern using a machine learning model;
- assigning, by the hardware computer system, a relatively high weight at a portion of the predicted and/or benchmark optical proximity correction;
- and
- training, by the hardware computer system, the machine learning model using an objective function evaluating the weighted predicted and/or benchmark optical proximity corrections.
2 more elements
A method comprising:
- applying, by a hardware computer system, a binarization function to image data of a predicted optical proximity correction of a design pattern by a machine learning model and/or a benchmark optical proximity correction of the design pattern to produce a respective binarized version thereof;
- and
- training, by a hardware computer system, a machine learning model configured to predict optical proximity corrections for design patterns by evaluating the predicted optical proximity correction against the benchmark optical proximity correction based on the binarized version of the predicted and/or benchmark optical proximity correction.
Dependencies are inferred from the claim text.