A robust low data solution: Dimension prediction of semiconductor nanorods
Precise control over dimension of nanocrystals is critical to tune the properties for various applications. However, the traditional control through experimental optimization is slow, tedious and time consuming. Herein a robust deep neural network-based regression algorithm has been developed for pr...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier BV
2021
|
| Online Access: | https://hdl.handle.net/1721.1/138488 |
| _version_ | 1811068977394221056 |
|---|---|
| author | Liu, Xiaoli Xu, Yang Li, Jiali Ong, Xuanwei Ali Ibrahim, Salwa Buonassisi, Tonio Wang, Xiaonan |
| author_facet | Liu, Xiaoli Xu, Yang Li, Jiali Ong, Xuanwei Ali Ibrahim, Salwa Buonassisi, Tonio Wang, Xiaonan |
| author_sort | Liu, Xiaoli |
| collection | MIT |
| description | Precise control over dimension of nanocrystals is critical to tune the properties for various applications. However, the traditional control through experimental optimization is slow, tedious and time consuming. Herein a robust deep neural network-based regression algorithm has been developed for precise prediction of length, width, and aspect ratios of semiconductor nanorods (NRs). Given there is limited experimental data available (28 samples), a Synthetic Minority Oversampling Technique for regression (SMOTE-REG) is employed first for data generation. Deep neural network is further applied to develop regression model which demonstrated the well performed prediction on both the original and generated data with a similar distribution. The prediction model is further validated with additional experimental data, showing accurate prediction results. Additionally, Local Interpretable Model-Agnostic Explanations (LIME) is used to interpret the weight for each sample, corresponding to its importance towards the target dimension, which is well validated by experimental observations. |
| first_indexed | 2024-09-23T08:03:44Z |
| format | Article |
| id | mit-1721.1/138488 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T08:03:44Z |
| publishDate | 2021 |
| publisher | Elsevier BV |
| record_format | dspace |
| spelling | mit-1721.1/1384882021-12-16T03:40:34Z A robust low data solution: Dimension prediction of semiconductor nanorods Liu, Xiaoli Xu, Yang Li, Jiali Ong, Xuanwei Ali Ibrahim, Salwa Buonassisi, Tonio Wang, Xiaonan Precise control over dimension of nanocrystals is critical to tune the properties for various applications. However, the traditional control through experimental optimization is slow, tedious and time consuming. Herein a robust deep neural network-based regression algorithm has been developed for precise prediction of length, width, and aspect ratios of semiconductor nanorods (NRs). Given there is limited experimental data available (28 samples), a Synthetic Minority Oversampling Technique for regression (SMOTE-REG) is employed first for data generation. Deep neural network is further applied to develop regression model which demonstrated the well performed prediction on both the original and generated data with a similar distribution. The prediction model is further validated with additional experimental data, showing accurate prediction results. Additionally, Local Interpretable Model-Agnostic Explanations (LIME) is used to interpret the weight for each sample, corresponding to its importance towards the target dimension, which is well validated by experimental observations. 2021-12-15T16:47:03Z 2021-12-15T16:47:03Z 2021 2021-12-15T16:36:46Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/138488 Liu, Xiaoli, Xu, Yang, Li, Jiali, Ong, Xuanwei, Ali Ibrahim, Salwa et al. 2021. "A robust low data solution: Dimension prediction of semiconductor nanorods." Computers and Chemical Engineering, 150. en 10.1016/J.COMPCHEMENG.2021.107315 Computers and Chemical Engineering Creative Commons Attribution-NonCommercial-NoDerivs License http://creativecommons.org/licenses/by-nc-nd/4.0/ application/pdf Elsevier BV arXiv |
| spellingShingle | Liu, Xiaoli Xu, Yang Li, Jiali Ong, Xuanwei Ali Ibrahim, Salwa Buonassisi, Tonio Wang, Xiaonan A robust low data solution: Dimension prediction of semiconductor nanorods |
| title | A robust low data solution: Dimension prediction of semiconductor nanorods |
| title_full | A robust low data solution: Dimension prediction of semiconductor nanorods |
| title_fullStr | A robust low data solution: Dimension prediction of semiconductor nanorods |
| title_full_unstemmed | A robust low data solution: Dimension prediction of semiconductor nanorods |
| title_short | A robust low data solution: Dimension prediction of semiconductor nanorods |
| title_sort | robust low data solution dimension prediction of semiconductor nanorods |
| url | https://hdl.handle.net/1721.1/138488 |
| work_keys_str_mv | AT liuxiaoli arobustlowdatasolutiondimensionpredictionofsemiconductornanorods AT xuyang arobustlowdatasolutiondimensionpredictionofsemiconductornanorods AT lijiali arobustlowdatasolutiondimensionpredictionofsemiconductornanorods AT ongxuanwei arobustlowdatasolutiondimensionpredictionofsemiconductornanorods AT aliibrahimsalwa arobustlowdatasolutiondimensionpredictionofsemiconductornanorods AT buonassisitonio arobustlowdatasolutiondimensionpredictionofsemiconductornanorods AT wangxiaonan arobustlowdatasolutiondimensionpredictionofsemiconductornanorods AT liuxiaoli robustlowdatasolutiondimensionpredictionofsemiconductornanorods AT xuyang robustlowdatasolutiondimensionpredictionofsemiconductornanorods AT lijiali robustlowdatasolutiondimensionpredictionofsemiconductornanorods AT ongxuanwei robustlowdatasolutiondimensionpredictionofsemiconductornanorods AT aliibrahimsalwa robustlowdatasolutiondimensionpredictionofsemiconductornanorods AT buonassisitonio robustlowdatasolutiondimensionpredictionofsemiconductornanorods AT wangxiaonan robustlowdatasolutiondimensionpredictionofsemiconductornanorods |