Multi-output regression using polygon generation and conditional generative adversarial networks
作者：
Highlights：
• The paper proposes accurate and representative multi-output regression method.
• The method is based on polygon generation and deep generative modeling.
• It learns the true distribution of complex industrial data.
• The method expresses all interrelationships between data inputs and outputs (KPIs).
• It is tested for KPIs’ prediction using data collected from an industrial plant.
摘要
•The paper proposes accurate and representative multi-output regression method.•The method is based on polygon generation and deep generative modeling.•It learns the true distribution of complex industrial data.•The method expresses all interrelationships between data inputs and outputs (KPIs).•It is tested for KPIs’ prediction using data collected from an industrial plant.
论文关键词：Generative modeling,Conditional generative adversarial network (cGAN),Polygon generation,Multi-output regression,Deep learning (DL),Hamiltonian cycles,Image processing
论文评审过程：Received 1 April 2021, Revised 25 October 2021, Accepted 22 April 2022, Available online 28 April 2022, Version of Record 19 May 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.117288
