This paper presents stable reinforcement learning through adversarial imitation learning (SRAIL), a two-phase learning framework for automatic proportional integral derivative (PID) tuning in the absence of high-quality demonstration data. First, adversarial imitation learning (AIL) is used to learn an initial control policy using imperfect demonstration data. Subsequently, reinforcement learning (RL) refines this initial policy to derive the optimal PID control policy. SRAIL is designed to ensure stable and safe exploration of the RL agent, even with imperfect demonstration data, thereby enabling efficient and automatic tuning in industrial environments. Conventional PID tuning methods are time-consuming owing to system complexity and lack of prior knowledge. Furthermore, the random exploratory nature of RL presents significant risks during the early training phase, limiting its practical application. SRAIL addresses these challenges by ensuring stable control performance despite the absence of high-quality demonstration data and by enhancing the convergence speed of RL. The experimental results demonstrate that SRAIL minimizes random exploration in the initial learning phase, ensuring safety, and highlighting its potential for implementation of RL in real-world industrial environments.

In this study, we propose a real-time contrasts control chart based on reinforcement learning (RL-RTC). Effective process monitoring, which directly influences productivity and yield, has become increasingly important. The traditional RTC control chart offers a promising approach to process monitoring by transforming the monitoring problem into a real-time classification task, enabling more timely and accurate detection of issues. However, the parameters of the traditional RTC control chart are typically set based on empirical methods, which limits the ability to fine-tune them effectively. To address this limitation and improve the performance of the RTC method, we propose the RL-RTC control chart, which leverages reinforcement learning for more adaptive parameter control. The proposed method takes the data itself as a state of reinforcement learning (RL) and adaptively decides the parameter of the RTC control chart in real-time. In this paper, the control of the moving window size is defined as an action taken by the RL agent. By automatically learning an optimal policy, the RL-RTC method eliminates the need for manual parameter tuning and enhances adaptability in dynamic process environments. The RL-RTC approach can detect shifts more quickly while maintaining the ability to identify the causes of faults. Compared to a conventional RTC control chart, experimental results demonstrate that the RL-RTC method offers improved performance by fine-tuning the window size in response to changing process conditions. Therefore, a wide scope of applications is expected for adaptively controlling the other RTC control chart parameters.

This paper proposes a framework that improves classification performance for multivariate time series data while providing an objective assessment of each variable’s influence, including interaction effects. While convolutional neural networks (CNNs) offer significant advantages in analyzing multivariate time series data, the structural limitations of CNNs have restricted their ability to detect statistical interactions. Our approach creatively modifies convolutional filters and layer structures, enabling feature extraction that captures the influence of pairwise interactions. These extracted features are processed by interpretable models to calculate feature importance, enabling in-depth causal analysis by quantifying both individual and pairwise variable effects. In addition, the proposed method enhances the overall classification performance of multivariate time series data. Synthetic data experiments verified that the proposed method effectively extracted relevant features that explain pairwise interactions. In addition, in the multivariate time series classification experiments using real data, the proposed method demonstrated superior performance compared to baseline methods. These results suggest that the proposed approach is a practical and interpretable solution for multivariate time series classification tasks in domains where variable interactions play a decisive role, such as healthcare, finance, and manufacturing.

Access the original poster in full on F1000Research: Phenotype–genotype integration Using jedae-unshiu reference SNPs for citrus line characterization