Data-Augmented Regression with Generative Convolutional Network

Publisher:
Springer International Publishing
Publication Type:
Conference Proceeding
Citation:
Web Information Systems Engineering – WISE 2018, 2018, pp. 301 - 311
Issue Date:
2018
Metrics:
Full metadata record
Files in This Item:
Filename Description Size
Ning2018_Chapter_Data-AugmentedRegressionWithGe.pdfPublished version1.18 MB
Adobe PDF
Generative adversarial networks (GAN)-based approaches have been extensively investigated whereas GAN-inspired regression (i.e., numeric prediction) has rarely been studied in image and video processing domains. The lack of sufficient labeled data in many real-world cases poses great challenges to regression methods, which generally require sufficient labeled samples for their training. In this regard, we propose a unified framework that combines a robust autoencoder and a generative convolutional neural network (GCNN)-based regression model to address the regression problem. Our model is able to generate high-quality artificial samples via augmenting the size of a small number of training samples for better training effects. Extensive experiments are conducted on two real-world datasets and the results show that our proposed model consistently outperforms a set of advanced techniques under various evaluation metrics.
Please use this identifier to cite or link to this item: