Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development
Herein, novel neural network (NN) methods that improve prediction accuracy and reduce output variance of the optimized input in the gradient method for cross‐sectional data are proposed, and the variability evaluation approach of optimized inputs in the semiconductor process is suggested. Specifical...
| Principais autores: | , , , , , , , , , |
|---|---|
| Formato: | Artigo |
| Idioma: | English |
| Publicado em: |
Wiley
2023-09-01
|
| coleção: | Advanced Intelligent Systems |
| Assuntos: | |
| Acesso em linha: | https://doi.org/10.1002/aisy.202300089 |
| _version_ | 1827808642019098624 |
|---|---|
| author | Hyeok Yun Chang-Hyeon An Hyundong Jang Kyeongrae Cho Jeong-Sik Lee Seungjoon Eom Choong-Ki Kim Min-Soo Yoo Hyun-Chul Choi Rock-Hyun Baek |
| author_facet | Hyeok Yun Chang-Hyeon An Hyundong Jang Kyeongrae Cho Jeong-Sik Lee Seungjoon Eom Choong-Ki Kim Min-Soo Yoo Hyun-Chul Choi Rock-Hyun Baek |
| author_sort | Hyeok Yun |
| collection | DOAJ |
| description | Herein, novel neural network (NN) methods that improve prediction accuracy and reduce output variance of the optimized input in the gradient method for cross‐sectional data are proposed, and the variability evaluation approach of optimized inputs in the semiconductor process is suggested. Specifically, electrical parameter measurements (EPMs) and power‐delay product of industrial high‐k metal gate DRAM peripheral 29‐stage ring oscillator circuits, including NMOS, PMOS, and interconnects, are focused on. The proposed methods find an optimized input to achieve a lower NN output variance in the gradient descent than one multilayer perceptron (MLP) and mean ensemble of MLPs even when considering the variabilities of the devices and interconnects. The local optima problem of one MLP is resolved by utilizing multiple MLPs trained with different train/validation data, their trimmed mean, and an additional learnable layer. Moreover, adding the learnable layer secures versatility for various parametric datasets. The methods improve the prediction accuracy (R2) by 5.6–15.6% in sparse data space compared to one MLP and the mean ensemble, decrease the NN output variance of the optimized input by 73.0–81.6% compared to one MLP and the mean ensemble, and are successfully verified by implementing it on EPMs of 3977 test patterns of 314 wafers and 16 lots. |
| first_indexed | 2024-03-11T22:30:09Z |
| format | Article |
| id | doaj.art-ab5cb2a7ed1c40f0b5d17fbb27fbd5d2 |
| institution | Directory Open Access Journal |
| issn | 2640-4567 |
| language | English |
| last_indexed | 2024-03-11T22:30:09Z |
| publishDate | 2023-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advanced Intelligent Systems |
| spelling | doaj.art-ab5cb2a7ed1c40f0b5d17fbb27fbd5d22023-09-23T07:09:23ZengWileyAdvanced Intelligent Systems2640-45672023-09-0159n/an/a10.1002/aisy.202300089Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology DevelopmentHyeok Yun0Chang-Hyeon An1Hyundong Jang2Kyeongrae Cho3Jeong-Sik Lee4Seungjoon Eom5Choong-Ki Kim6Min-Soo Yoo7Hyun-Chul Choi8Rock-Hyun Baek9Department of Electrical Engineering Pohang University of Science and Technology Pohang 37673 Republic of KoreaDepartment of Electronic Engineering Yeungnam University Gyeongsan 38541 Republic of KoreaDepartment of Electrical Engineering Pohang University of Science and Technology Pohang 37673 Republic of KoreaDepartment of Electrical Engineering Pohang University of Science and Technology Pohang 37673 Republic of KoreaDepartment of Electronic Engineering Yeungnam University Gyeongsan 38541 Republic of KoreaDepartment of Electrical Engineering Pohang University of Science and Technology Pohang 37673 Republic of KoreaDepartment of DRAM Development SK hynix Inc. Icheon 17336 Republic of KoreaDepartment of DRAM Development SK hynix Inc. Icheon 17336 Republic of KoreaDepartment of Electronic Engineering Yeungnam University Gyeongsan 38541 Republic of KoreaDepartment of Electrical Engineering Pohang University of Science and Technology Pohang 37673 Republic of KoreaHerein, novel neural network (NN) methods that improve prediction accuracy and reduce output variance of the optimized input in the gradient method for cross‐sectional data are proposed, and the variability evaluation approach of optimized inputs in the semiconductor process is suggested. Specifically, electrical parameter measurements (EPMs) and power‐delay product of industrial high‐k metal gate DRAM peripheral 29‐stage ring oscillator circuits, including NMOS, PMOS, and interconnects, are focused on. The proposed methods find an optimized input to achieve a lower NN output variance in the gradient descent than one multilayer perceptron (MLP) and mean ensemble of MLPs even when considering the variabilities of the devices and interconnects. The local optima problem of one MLP is resolved by utilizing multiple MLPs trained with different train/validation data, their trimmed mean, and an additional learnable layer. Moreover, adding the learnable layer secures versatility for various parametric datasets. The methods improve the prediction accuracy (R2) by 5.6–15.6% in sparse data space compared to one MLP and the mean ensemble, decrease the NN output variance of the optimized input by 73.0–81.6% compared to one MLP and the mean ensemble, and are successfully verified by implementing it on EPMs of 3977 test patterns of 314 wafers and 16 lots.https://doi.org/10.1002/aisy.202300089gradient descent methodholistic parameter optimizationmachine learningnetwork ensembleneural networkssemiconductor manufacturing |
| spellingShingle | Hyeok Yun Chang-Hyeon An Hyundong Jang Kyeongrae Cho Jeong-Sik Lee Seungjoon Eom Choong-Ki Kim Min-Soo Yoo Hyun-Chul Choi Rock-Hyun Baek Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development Advanced Intelligent Systems gradient descent method holistic parameter optimization machine learning network ensemble neural networks semiconductor manufacturing |
| title | Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development |
| title_full | Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development |
| title_fullStr | Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development |
| title_full_unstemmed | Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development |
| title_short | Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development |
| title_sort | accurate prediction and reliable parameter optimization of neural network for semiconductor process monitoring and technology development |
| topic | gradient descent method holistic parameter optimization machine learning network ensemble neural networks semiconductor manufacturing |
| url | https://doi.org/10.1002/aisy.202300089 |
| work_keys_str_mv | AT hyeokyun accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT changhyeonan accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT hyundongjang accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT kyeongraecho accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT jeongsiklee accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT seungjooneom accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT choongkikim accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT minsooyoo accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT hyunchulchoi accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment AT rockhyunbaek accuratepredictionandreliableparameteroptimizationofneuralnetworkforsemiconductorprocessmonitoringandtechnologydevelopment |