ABSTRACT In the chemical mechanical polishing (CMP) process in semiconductor manufacturing, run‐to‐run (R2R) process control is critical for consistently producing high‐quality products. R2R process control optimizes equipment parameters to maintain product specifications within target ranges while compensating for process noise. In this paper, we propose a novel method to overcome two limitations of existing R2R process control algorithms in the CMP process. The first limitation is that the relationships between control variables, which can be represented by their ratios, are crucial but not adequately considered. The proposed method overcomes this limitation by preserving control variable ratios derived from historical process data, thus ensuring robust process control under high‐intensity noise conditions. The second limitation is that existing methods cannot prioritize control variables when multiple combinations of these variables can reach the target. Adjusting each control variable has a different associated cost, but current approaches fail to consider these cost differences. Consequently, they cannot consider the priorities of control variables, which may lead to unnecessary costs in the control process. The proposed method introduces a priority‐based approach for control variable adjustments, which can reduce overall process control costs. Furthermore, by incorporating an autoencoder to estimate process environment changes, the proposed method achieves superior performance across various process control metrics compared with traditional R2R algorithms. These improvements enable more robust and cost‐effective process control in the increasingly complex CMP process.

Real-time dynamic scheduling in modern manufacturing is highly complex due to frequent disturbances and intricate operational constraints. While reinforcement learning (RL) has shown promise, existing approaches often rely on extensive dispatching rules and struggle to scale to factory-wide settings. This study introduces a scalable multi-agent RL (MARL) framework with a leader–follower structure, enabling decentralized agents to handle sub-problems while maintaining global coordination through abstract goals. To further enhance robustness, a limited rule-based conversion algorithm is proposed to mitigate performance degradation from poor agent decisions. Our experimental results demonstrate that the proposed model outperforms the state-of-the-art deep RL-based scheduling methods in various aspects. Additionally, the proposed model provides the most robust scheduling performance to demand changes. Overall, the proposed MARL-based scheduling model presents a promising solution to the real-time scheduling problem, with potential applications in various manufacturing industries.

When an unknown example, one that was not seen during training, appears, most recognition systems usually produce overgeneralized results and determine that the example belongs to one of the known classes. To address this problem, teacher-explorer-student (T/E/S) learning, which adopts the concept of open set recognition (OSR) to reject unknown samples while minimizing the loss of classification performance on known samples, is proposed in this study. In this novel learning method, the overgeneralization of deep-learning classifiers is significantly reduced by exploring various possibilities for unknowns. The teacher network extracts hints about unknowns by distilling the pretrained knowledge about knowns and delivers this distilled knowledge to the student network. After learning the distilled knowledge, the student network shares its learned information with the explorer network. Next, the explorer network shares its exploration results by generating unknown-like samples and feeding those samples to the student network. As this alternating learning process is repeated, the student network experiences a variety of synthetic unknowns, reducing overgeneralization. The results of extensive experiments show that each component proposed in this article significantly contributes to improving OSR performance. It is found that the proposed T/E/S learning method outperforms current state-of-the-art methods.

Unknown examples that are unseen during training often appear in real-world pattern recognition tasks, and an intelligent self-learning system should be able to distinguish between known examples and unknown examples. Accordingly, open-set recognition (OSR), which addresses the problem of classifying knowns and identifying unknowns, has recently been highlighted. However, conventional deep neural networks (DNNs) using a softmax layer are vulnerable to overgeneralization, producing high confidence scores for unknowns. In this article, we propose a simple OSR method that is based on the intuition that the OSR performance can be maximized by setting strict and sophisticated decision boundaries that reject unknowns while maintaining satisfactory classification performance for knowns. For this purpose, a novel network structure, in which multiple one-vs-rest networks (OVRNs) follow a convolutional neural network (CNN) feature extractor, is proposed. Here, an OVRN is a simple feedforward neural network that is designed to assign confidence scores that are lower than those in the softmax layer to unknown samples so that unknown samples can be more effectively separated from known classes. Furthermore, the collective decision score is modeled by combining the multiple decisions reached by the OVRNs to alleviate overgeneralization. Extensive experiments were conducted on various datasets, and the experimental results show that the proposed method performs significantly better than the state-of-the-art methods by effectively reducing overgeneralization. The code is available at https://github.com/JaeyeonJang/Openset-collective-decision.

Recent advances in single-cell RNA sequencing (scRNA-seq) technology have enabled the acquisition of RNA at the single-cell level, which showed that the expression level of genes is highly variable across and within the cell types. Even well-known housekeeping genes showed high expression variance in a single condition and within the same cell types. Previous studies made efforts to identify stably expressed genes and use them as a yardstick for robust gene expression normalization. On the other hand, drugs were shown to be less effective on genes with high expression variance. Thus, identifying both stably and variably expressed genes is an important task, especially at the single-cell level. In this study, using the Kullback-Leibler divergence method, we proposed a metric to measure the expression stability of each gene. Using private scRNA-seq data composed of 25 severe COVID-19 patients and 40 healthy individuals, we identified variably expressed genes specific to COVID-19-infected patients and healthy cohorts.

In many real-world prognostics and health management tasks, where the available training samples are insufficient, deep neural networks are highly vulnerable to overfitting. To address this problem, in this article, we propose a novel health representation learning method based on a Siamese network. This method prevents overfitting by utilizing a constraint by which the differences between samples in the embedding space of the Siamese network should follow the differences in the remaining useful life (RUL) values via the introduction of a multitask learning scheme. In addition, since the learned embedding space reflects the dynamics of degradation, each training sample can be used as a reference to estimate the RUL of a test sample. By combining the estimates for all training samples, the proposed method enables robust RUL prediction. Experimental results show that the proposed learning and estimation method contributes to improving not only RUL prediction performance but also robustness to data insufficiency.