US 12,254,392 B2
Apparatus and method for property joint interpolation and prediction
Faegheh Hasibi, Eindhoven (NL); Leon Paul Van Dijk, Eindhoven (NL); Maialen Larranaga, Eindhoven (NL); Alexander Ypma, Eindhoven (NL); and Richard Johannes Franciscus Van Haren, Waalre (NL)
Assigned to ASML NETHERLANDS B.V., Veldhoven (NL)
Appl. No. 17/423,658
Filed by ASML NETHERLANDS B.V., Veldhoven (NL)
PCT Filed Dec. 12, 2019, PCT No. PCT/EP2019/084923
§ 371(c)(1), (2) Date Jul. 16, 2021,
PCT Pub. No. WO2020/156724, PCT Pub. Date Aug. 6, 2020.
Claims priority of application No. 19154587 (EP), filed on Jan. 30, 2019; and application No. 19164072 (EP), filed on Mar. 20, 2019.
Prior Publication US 2022/0083834 A1, Mar. 17, 2022
Int. Cl. G06F 30/398 (2020.01); G03F 7/00 (2006.01); G06F 30/27 (2020.01); G06N 3/04 (2023.01); G06F 119/18 (2020.01); G06N 3/0464 (2023.01)
CPC G06N 3/04 (2013.01) [G03F 7/705 (2013.01); G03F 7/70616 (2013.01); G06F 30/27 (2020.01); G06F 30/398 (2020.01); G06F 2119/18 (2020.01); G06N 3/0464 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining a plurality of data sets, wherein each of the plurality of data sets comprises data associated with a spatial distribution of a parameter across a substrate and wherein one of the plurality of data sets has a different resolution and/or different spatial positions of data points compared to at least one other of the plurality of data sets;
representing each of the plurality of data sets as a multidimensional object;
obtaining a convolutional neural network model trained with previously obtained multidimensional objects and properties of previous substrates; and
applying the convolutional neural network model to the plurality of multidimensional objects representing the plurality of data sets, to predict a property associated with the substrate.