AUTHOR=Junhuai Li , Yunwen Wu , Huaijun Wang , Jiang Xu 

TITLE=Fault detection method based on adversarial reinforcement learning

JOURNAL=Frontiers in Computer Science

VOLUME=Volume 4 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/computer-science/articles/10.3389/fcomp.2022.1007665

DOI=10.3389/fcomp.2022.1007665

ISSN=2624-9898

ABSTRACT=Fault detection is an essential task for large-scale industrial maintenance. However, in practical applications, due to the possible harm caused by the collection of fault data, the fault samples that lead to the labeling are usually very few. Most existing methods consider training unsupervised models with a large amount of unlabeled data, while ignoring the rich knowledge existing in a small amount of labeled data. To make full use of these prior knowledge, this paper proposes an reinforcement learning model—namely Adversarial Reinforcement Learning in Weakly Supervised(WS-ARL), which performs significantly better by jointly learning a small labeled anomaly data and a large unlabeled data. We use an agent of the reinforcement learning model as fault detector and add a new environmental agent as a sample selector, By providing opposite reward for two agents, they learn in an adversarial environment. The feasibility and effectiveness of the model are verified by experimental analysis and compared of the performance of the model with four state-of-the-art weakly/un-supervised methods in the hydraulic press fault detection task.