US 11,747,740 B2
Self-supervised representation learning for interpretation of OCD data
Ran Yacoby, Rehovot (IL); and Boaz Sturlesi, Rehovot (IL)
Assigned to NOVA LTD, Rehovot (IL)
Appl. No. 17/790,765
Filed by NOVA LTD, Rehovot (IL)
PCT Filed Jan. 6, 2021, PCT No. PCT/IL2021/050018
§ 371(c)(1), (2) Date Jul. 5, 2022,
PCT Pub. No. WO2021/140508, PCT Pub. Date Jul. 15, 2021.
Claims priority of provisional application 62/957,339, filed on Jan. 6, 2020.
Prior Publication US 2023/0014976 A1, Jan. 19, 2023
Int. Cl. G01N 21/47 (2006.01); G03F 7/20 (2006.01); G06N 3/08 (2023.01); G01B 11/02 (2006.01); G01N 21/956 (2006.01); G06T 1/40 (2006.01); G03F 7/00 (2006.01)
CPC G03F 7/70625 (2013.01) [G03F 7/70508 (2013.01); G06N 3/08 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A method for OCD metrology, comprising:
receiving multiple first sets of scatterometric data;
dividing each of the multiple first sets of scatterometric data into k sub-vectors;
training, in a self-supervised manner, k2 auto-encoder neural networks, mapping each of the k sub-vectors to each other, wherein the k auto-encoder neural networks include k2 respective encoder neural networks each having at least one internal bottleneck layer;
receiving multiple respective sets of reference parameters and multiple corresponding second sets of scatterometric data, measured from multiple respective wafer patterns; and
training a transfer neural network (NN) having initial layers including a parallel arrangement of the k2 encoder neural networks, wherein the transfer NN training comprises training one or more final layers that follow the bottleneck layers of the encoder neural networks, and wherein target output of the transfer NN training is set to the multiple sets of reference parameters and feature input is set to the multiple corresponding second sets of scatterometric data, such that the transfer NN is trained to estimate new wafer pattern parameters from subsequently measured sets of scatterometric data.