A Case Study Applying Mesoscience to Deep Learning
Li Guo,
Fanyong Meng,
Pengfei Qin,
Zhaojie Xia,
Qi Chang,
Jianhua Chen,
Jinghai Li
Affiliations
Li Guo
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China; School of Chemical Engineering, University of Chinese Academy of Sciences, Beijing 100049, China; Corresponding author.
Fanyong Meng
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
Pengfei Qin
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China; School of Chemical Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
Zhaojie Xia
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China; School of Chemical Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
Qi Chang
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
Jianhua Chen
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
Jinghai Li
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
In this paper, we propose mesoscience-guided deep learning (MGDL), a deep learning modeling approach guided by mesoscience, to study complex systems. When establishing sample dataset based on the same system evolution data, different from the operation of conventional deep learning method, MGDL introduces the treatment of the dominant mechanisms of complex system and interactions between them according to the principle of compromise in competition (CIC) in mesoscience. Mesoscience constraints are then integrated into the loss function to guide the deep learning training. Two methods are proposed for the addition of mesoscience constraints. The physical interpretability of the model-training process is improved by MGDL because guidance and constraints based on physical principles are provided. MGDL was evaluated using a bubbling bed modeling case and compared with traditional techniques. With a much smaller training dataset, the results indicate that mesoscience-constraint-based model training has distinct advantages in terms of convergence stability and prediction accuracy, and it can be widely applied to various neural network configurations. The MGDL approach proposed in this paper is a novel method for utilizing the physical background information during deep learning model training. Further exploration of MGDL will be continued in the future.
