Improving Online non-destructive Moisture Content Estimation using Data Augmentation by Feature Space Interpolation with Variational Autoencoders

TY - GEN

T1 - Improving Online non-destructive Moisture Content Estimation using Data Augmentation by Feature Space Interpolation with Variational Autoencoders

AU - Wewer, Christian Remi

AU - Iosifidis, Alexandros

PY - 2023/8

Y1 - 2023/8

N2 - Data augmentation techniques have proven to be highly effective for many types of problems. However, the development of data augmentation for continuous input-output mappings in regression problems has not received much attention. Insufficient training data remains a significant challenge in machine learning, especially for industrial applications, as the cost of experimentation on the production line can be prohibitively expensive. This acts as a barrier to adoption of machine learning methods in industrial applications. In this study, we propose a data augmentation method called feature space interpolation for continuous input-output regression problems based on discontinuous data sets with clear gaps in the data. The proposed method is applied to a dataset of industrial drying of bulky filter media products. It is shown, that augmenting the original dataset by generated synthetic data points in the gap of the dataset by interpolation in the latent space of a well-trained variational autoencoder (VAE) can improve the performance of state-of-the-art of bulky filter media product moisture content estimation models, as measured by the mean absolute error and mean squared error by 4.82% and 6.32% respectively, and outperforms baseline generative data augmentation methods such as latent space sampling from VAEs.

AB - Data augmentation techniques have proven to be highly effective for many types of problems. However, the development of data augmentation for continuous input-output mappings in regression problems has not received much attention. Insufficient training data remains a significant challenge in machine learning, especially for industrial applications, as the cost of experimentation on the production line can be prohibitively expensive. This acts as a barrier to adoption of machine learning methods in industrial applications. In this study, we propose a data augmentation method called feature space interpolation for continuous input-output regression problems based on discontinuous data sets with clear gaps in the data. The proposed method is applied to a dataset of industrial drying of bulky filter media products. It is shown, that augmenting the original dataset by generated synthetic data points in the gap of the dataset by interpolation in the latent space of a well-trained variational autoencoder (VAE) can improve the performance of state-of-the-art of bulky filter media product moisture content estimation models, as measured by the mean absolute error and mean squared error by 4.82% and 6.32% respectively, and outperforms baseline generative data augmentation methods such as latent space sampling from VAEs.

KW - artificial neural networks

KW - data augmentation

KW - drying

KW - moisture content

KW - variational autoencoders

UR - http://www.scopus.com/inward/record.url?scp=85171189132&partnerID=8YFLogxK

U2 - 10.1109/INDIN51400.2023.10218063

DO - 10.1109/INDIN51400.2023.10218063

M3 - Article in proceedings

SN - 978-1-6654-9314-7

T3 - IEEE International Conference on Industrial Informatics Proceedings

BT - 2023 IEEE 21st International Conference on Industrial Informatics (INDIN)

PB - IEEE

ER -