دورية أكاديمية
Probabilistic invertible neural network for inverse design space exploration and reasoning
العنوان: | Probabilistic invertible neural network for inverse design space exploration and reasoning |
---|---|
المؤلفون: | Yiming Zhang, Zhiwei Pan, Shuyou Zhang, Na Qiu |
المصدر: | Electronic Research Archive, Vol 31, Iss 2, Pp 860-881 (2023) |
بيانات النشر: | AIMS Press, 2023. |
سنة النشر: | 2023 |
المجموعة: | LCC:Mathematics LCC:Applied mathematics. Quantitative methods |
مصطلحات موضوعية: | inverse design, invertible neural network, probabilistic machine learning, uncertainty quantification, turbine blade design, Mathematics, QA1-939, Applied mathematics. Quantitative methods, T57-57.97 |
الوصف: | Invertible neural network (INN) is a promising tool for inverse design optimization. While generating forward predictions from given inputs to the system response, INN enables the inverse process without much extra cost. The inverse process of INN predicts the possible input parameters for the specified system response qualitatively. For the purpose of design space exploration and reasoning for critical engineering systems, accurate predictions from the inverse process are required. Moreover, INN predictions lack effective uncertainty quantification for regression tasks, which increases the challenges of decision making. This paper proposes the probabilistic invertible neural network (P-INN): the epistemic uncertainty and aleatoric uncertainty are integrated with INN. A new loss function is formulated to guide the training process with enhancement in the inverse process accuracy. Numerical evaluations have shown that the proposed P-INN has noticeable improvement on the inverse process accuracy and the prediction uncertainty is reliable. |
نوع الوثيقة: | article |
وصف الملف: | electronic resource |
اللغة: | English |
تدمد: | 2688-1594 |
Relation: | https://doaj.org/toc/2688-1594 |
DOI: | 10.3934/era.2023043?viewType=HTML |
DOI: | 10.3934/era.2023043 |
URL الوصول: | https://doaj.org/article/8c899d8899784355a03a8176980e9368 |
رقم الأكسشن: | edsdoj.8c899d8899784355a03a8176980e9368 |
قاعدة البيانات: | Directory of Open Access Journals |
تدمد: | 26881594 |
---|---|
DOI: | 10.3934/era.2023043?viewType=HTML |