Deep learning has significantly improved performance in survival analysis (SA), which plays a crucial role in various fields, including medicine and engineering. A recent approach involves directly modeling the transition probability matrix using neural networks (NNs). However, modeling NNs as a transition probability matrix necessitates satisfying the constraints specific to transition probabilities. Existing methods often struggle to meet these constraints without compromising model performance (e.g., by limiting the model architectures). We propose a novel deep learning method for SA that models the transition probability matrix using NNs and trains them to meet the constraints through a specialized loss function. This loss function is designed to penalize the violation of the constraints using automatic differentiation (AD). Since any model capable of AD can compute this loss function, it enables the use of flexible and potentially higher-performing model architectures while forcing the model to satisfy the constraints of the transition probability. The experiments on real-world datasets demonstrate the superior performance of the proposed method compared to existing approaches. Additionally, ablation studies verify that all the components in the proposed method contribute to the performance.

Fault diagnosis is still a challenging task especially for unseen faults, which could happen in the systems. Zero-sample fault diagnosis alleviates this issue by utilizing information from seen faults and attributes defined by domain knowledge. However, existing methods suffer from irrelevant features in the original space while existing supervised feature selection methods yield unsatisfactory performance due to discrepancy caused by domain difference. In this article, we propose a novel feature selection method for zero-sample fault diagnosis, called concrete partial autoencoder. The concrete partial autoencoder selects features beneficial for both seen and unseen faults through striking a balance between classification accuracy and reconstruction errors of selected features. The concrete partial autoencoder utilizes categorical reparameterization to efficiently solve the feature selection problem. The evaluation results on the Tennessee Eastman Process show that the proposed method improves classification accuracy and robustness against irrelevant features at zero-sample fault diagnosis.

Remaining useful life (RUL) estimation is a crucial enabler of predictive maintenance. Since an adequate amount of labeled data is unavailable due to a low frequency of failures, semisupervised RUL estimation methods have been proposed to utilize unlabeled data collected in maintenance intervals to improve performance. However, those methods have limited capability since they only partially utilize unlabeled data during training. This article proposes a novel semisupervised method for RUL estimation in which both pretext and auxiliary tasks aim to estimate relative RULs. It draws more unlabeled samples from shorter RULs with a geometric progression and applies different strategies for pairing unlabeled samples at different training stages. The proposed method is evaluated on NASA C-MAPSS dataset and Backblaze HDD dataset. The results show that the proposed method significantly improves the performance with scarce labeled data for training.

Unsupervised anomaly detection approaches have been widely accepted in applications for industrial systems. Industrial systems often operate with multiple modes since they work for multiple purposes or under different conditions. In order to deal with the difficulty of anomaly detection due to multiple operating modes, multiple modeling strategies are employed. However, estimating the optimal set of models is a challenging problem due to the lack of supervision and computational burden. In this paper, we propose DeconAnomaly, a deep learning framework to estimate the optimal set of models using transfer learning for unsupervised anomaly detection under a multiple modeling strategy. It reduces computational burden with transfer learning and optimizes the number of models based on a surrogate metric of detection performance. The experimental results show clear advantages of DeconAnomaly.