A Novel Ensemble Adaptive Sparse Bayesian Transfer Learning Machine for Nonlinear Large-Scale Process Monitoring

Scalable and memory-efficient sparse learning for classification with approximate Bayesian regularization priors

In this article we study the sparse signal recovery problem in a Bayesian framework using a novel Bootstrapped Sparse Bayesian Learning method. Sparse Bayesian Learning (SBL) framework is an effective tool for pruning out the irrelevant features and ending up with a sparse representation. In SBL the choice of prior over the variances of the Gaussian Scale mixture has been an interesting area of research for some time now. This motivates us to use a more generalized maximum entropy density as the prior which results in a new variant of SBL. It has been shown to perform better than traditional SBL empirically and it also accelerates the pruning procedure. Because of this advantage, this variant of SBL can be claimed as more robust choice as it is less sensitive to the threshold for pruning. Theoretical justifications have also been provided to show that the proposed model actually promotes sparse point estimates.

This study aimed to find suitable source domain data in cross-domain transfer learning to extract robust image features. Then, a model was built to preoperatively distinguish lung granulomatous nodules (LGNs) from lung adenocarcinoma (LAC) in solitary pulmonary solid nodules (SPSNs). Data from 841 patients with SPSNs from five centres were collected retrospectively. First, adaptive cross-domain transfer learning was used to construct transfer learning signatures (TLS) under different source domain data and conduct a comparative analysis. The Wasserstein distance was used to assess the similarity between the source domain and target domain data in cross-domain transfer learning. Second, a cross-domain transfer learning radiomics model (TLRM) combining the best performing TLS, clinical factors and subjective CT findings was constructed. Finally, the performance of the model was validated through multicentre validation cohorts. Relative to other source domain data, TLS based on lung whole slide images as source domain data (TLS-LW) had the best performance in all validation cohorts (AUC range: 0.8228-0.8984). Meanwhile, the Wasserstein distance of TLS-LW was 1.7108, which was minimal. Finally, TLS-LW, age, spiculated sign and lobulated shape were used to build the TLRM. In all validation cohorts, The AUC ranges were 0.9074-0.9442. Compared with other models, decision curve analysis and integrated discrimination improvement showed that TLRM had better performance. The TLRM could assist physicians in preoperatively differentiating LGN from LAC in SPSNs. Furthermore, compared with other images, cross-domain transfer learning can extract robust image features when using lung whole slide images as source domain data and has a better effect.

In this paper, we develop a new sparse Bayesian learning method for recovery of block-sparse signals with un- known cluster patterns. A pattern-coupled hierarchical Gaussian prior model is introduced to characterize the statistical depen- dencies among coefficients, where a set of hyperparameters are employed to control the sparsity of signal coefficients. Unlike the conventional sparse Bayesian learning framework in which each individual hyperparameter is associated independently with each coefficient, in this paper, the prior for each coefficient not only involves its own hyperparameter, but also the hyperparameters of its immediate neighbors. In doing this way, the sparsity patterns of neighboring coefficients are related to each other and the hierarchical model has the potential to encourage structured- sparse solutions. The hyperparameters, along with the sparse signal, are learned by maximizing their posterior probability via an expectation-maximization (EM) algorithm. Index Terms— Sparse Bayesian learning, pattern-coupled hi- erarchical model, block-sparse signal recovery.

An efficient off-grid direction-of-arrival estimation method based on inverse-free sparse Bayesian learning

Sparse arrays are special geometrical arrangements of sensors which overcome some of the drawbacks associated with dense uniform arrays and require fewer sensors. For direction finding applications, sparse arrays with the same number of sensors can resolve more sources while providing higher resolution than a dense uniform array. This has been verified numerically and with real data for one-dimensional microphone arrays. In this study the use of nested and co-prime arrays is examined with sparse Bayesian learning (SBL), which is a compressive sensing algorithm, for estimating sparse vectors and support. SBL is an iterative parameter estimation method and can process multiple snapshots as well as multiple frequency data within its Bayesian framework. A multi-frequency variant of SBL is proposed, which accounts for non-flat frequency spectra of the sources. Experimental validation of azimuth and elevation [two-dimensional (2D)] direction-of-arrival (DOA)estimation are provided using sparse arrays and real data acquired in an anechoic chamber with a rectangular array. Both co-prime and nested arrays are obtained by sampling this rectangular array. The SBL method is compared with conventional beamforming and multiple signal classification for 2D DOA estimation of experimental data.

Recent progress in machine learning has been mainly due to the availability of large amounts of annotated data used for training complex models with deep architectures. Annotating this training data becomes burdensome and creates a major bottleneck in maintaining machine-learning databases. Moreover, these trained models fail to generalize to new categories or new varieties of the same categories. This is because new categories or new varieties have data distribution different from the training data distribution. To tackle these problems, this thesis proposes to develop a family of transfer-learning techniques that can deal with different training (source) and testing (target) distributions with the assumption that the availability of annotated data is limited in the testing domain. This is done by using the auxiliary data-abundant source domain from which useful knowledge is transferred that can be applied to data-scarce target domain. This transferable knowledge serves as a prior that biases target-domain predictions and prevents the target-domain model from overfitting. Specifically, we explore structural priors that encode relational knowledge between different data entities, which provides more informative bias than traditional priors. The choice of the structural prior depends on the information availability and the similarity between the two domains. Depending on the domain similarity and the information availability, we divide the transfer learning problem into four major categories and propose different structural priors to solve each of these sub-problems. This thesis first focuses on the unsupervised-domain-adaptation problem, where we propose to minimize domain discrepancy by transforming labeled source-domain data to be close to unlabeled target-domain data. For this problem, the categories remain the same across the two domains and hence we assume that the structural relationship between the source-domain samples is carried over to the target domain. Thus, graph or hyper-graph is constructed as the structural prior from both domains and a graph/hyper-graph matching formulation is used to transform samples in the source domain to be closer to samples in the target domain. An efficient optimization scheme is then proposed to tackle the time and memory inefficiencies associated with the matching problem. The few-shot learning problem is studied next, where we propose to transfer knowledge from source-domain categories containing abundantly labeled data to novel categories in the target domain that contains only few labeled data. The knowledge transfer biases the novel category predictions and prevents the model from overfitting. The knowledge is encoded using a neural-network-based prior that transforms a data sample to its corresponding class prototype. This neural network is trained from the source-domain data and applied to the target-domain data, where it transforms the few-shot samples to the novel-class prototypes for better recognition performance. The few-shot learning problem is then extended to the situation, where we do not have access to the source-domain data but only have access to the source-domain class prototypes. In this limited information setting, parametric neural-network-based priors would overfit to the source-class prototypes and hence we seek a non-parametric-based prior using manifolds. A piecewise linear manifold is used as a structural prior to fit the source-domain-class prototypes. This structure is extended to the target domain, where the novel-class prototypes are found by projecting the few-shot samples onto the manifold. Finally, the zero-shot learning problem is addressed, which is an extreme case of the few-shot learning problem where we do not have any labeled data in the target domain. However, we have high-level information for both the source and target domain categories in the form of semantic descriptors. We learn the relation between the sample space and the semantic space, using a regularized neural network so that classification of the novel categories can be carried out in a common representation space. This same neural network is then used in the target domain to relate the two spaces. In case we want to generate data for the novel categories in the target domain, we can use a constrained generative adversarial network instead of a traditional neural network. Thus, we use structural priors like graphs, neural networks and manifolds to relate various data entities like samples, prototypes and semantics for these different transfer learning sub-problems. We explore additional post-processing steps like pseudo-labeling, domain adaptation and calibration and enforce algorithmic and architectural constraints to further improve recognition performance. Experimental results on standard transfer learning image recognition datasets produced competitive results with respect to previous work. Further experimentation and analyses of these methods provided better understanding of machine learning as well.

Refined sparse Bayesian learning configuration for stochastic response analysis

The concept of Industry 4.0 uses cyber-physical systems and the Internet of Things to create "smart factories" that enable automated and connected production. However, the complex and diverse nature of cyber-physical systems makes them vulnerable to physical threats, where any fault in their physical components can lead to serious consequences, from financial losses to human safety. To reduce the risk of such failures, it is essential to have techniques in place to monitor the condition of these physical components closely. In recent years, data-driven predictive maintenance methods have become common due to the growth in data and advancements in machine learning. This thesis focuses on fault diagnosis, a key part of predictive maintenance, which aims to detect and identify faults in physical components. Incorporating advanced techniques, including machine learning algorithms and sensor data analysis, enhanced fault diagnosis systems' accuracy and efficiency. However, several real-world barriers challenge the effectiveness of these techniques. This research is structured around three primary objectives. The first objective aims to establish an efficient fault diagnosis system that is highly sensitive to fault occurrences while being resilient to uncertainties such as variability in fault patterns and data imbalance. To address variability in fault patterns, we developed a multi-step approach integrating clustering techniques and an adaptive neuro-fuzzy inference system, which significantly improved diagnosis robustness even under noisy conditions. For imbalanced datasets, a two-step method was introduced, combining permutation entropy for fault detection and envelope analysis for fault isolation. This approach was enhanced by an automatic frequency band selection method leveraging wavelet transform and statistical criteria, resulting in high fault diagnosis accuracy and reliability, even with limited labeled faulty data. The second objective focuses on developing fault diagnosis systems that can maintain high performance despite domain shifts and limited training data in both source and target domains. To address the challenge of domain shifts, a transfer learning technique was introduced, and a systematic literature review was conducted. This study bridges a technical gap by providing a comprehensive taxonomy, state-of-the-art methods, and insights into the challenges of applying transfer learning to predictive maintenance. Additionally, a conceptual framework was proposed to explore scenarios for transferring diagnostic knowledge from different source domains to target domains with insufficient training datasets. In a subsequent study, a new diagnostic framework was developed using digital twins and partial transfer learning to address the lack of real training datasets in both source and target domains as well as label space inconsistencies between them. The framework includes a general approach for developing digital twins of real machines to generate synthetic diagnostic data as source domain data. Furthermore, by incorporating a deep learning network and a double-layer attention mechanism, a partial transfer learning diagnostic model was developed to transfer diagnostic knowledge from the digital twin to the real target domain despite label space inconsistencies. The proposed framework was validated through an industrial case study involving rolling bearings in rotating equipment. The third objective focuses on developing an efficient fault diagnosis system capable of maintaining reliability under unseen and evolving working conditions. To achieve this, a novel self-adaptive fault diagnosis system was proposed, incorporating the MAPE-K loop, a domain generalization network model, and digital twin technologies. This self-adaptive fault diagnosis system leverages the MAPE-K loop to adapt in real time to evolving working conditions. Digital twins were employed to continuously generate data aligned with the real system’s operational context, enabling accurate fault diagnosis in dynamic environments. The domain generalization network model enhances generalization to unseen target working conditions by addressing overfitting and feature loss through multi-domain data augmentation, adversarial learning, and domain discrepancy modules. Validation using multiple rotating machinery datasets demonstrated that the developed fault diagnosis system outperformed state-of-the-art methods across various scenarios. In summary, this dissertation proposes a comprehensive and practical approach to addressing variability in fault patterns, imbalance datasets, data scarcity, domain shifts, and unseen and evolving working conditions challenges in existing fault diagnosis systems.

In this paper, we focus on a multi-task sparse Bayesian learning (SBL) theory that simultaneously utilizes multiple measurement vectors that are marked by a similar sparseness profile. Joint learning for sparse representations of multiple tasks has been mostly overlooked, although it is a useful tool for exploiting data redundancy by modeling the statistical relationships within groups of measurements. We first present a hierarchical Bayesian model and Bayesian inference framework for the multi-task SBL algorithm. Then we investigate the application of the multi-task SBL for model updating in structural health monitoring. The results verify that exploiting common information among multiple related tasks leads to better performance of the Bayesian inversion

Dynamic source domain selection: An adaptive EEG transfer learning framework for mitigating negative transfer.

A global sensitivity analysis and Bayesian inference framework was developed for improving the parameterization and predictability of a monthly time step process‐based biogeochemistry model. Using a Latin Hypercube sampler and an existing Terrestrial Ecosystem Model (TEM), a set of 500,000 Monte Carlo ensemble simulations was conducted for a black spruce forest ecosystem. A global sensitivity analysis was then conducted to identify the key model parameters and examine the interaction structures among TEM parameters. Bayesian inference analysis was also performed using these ensemble simulations and eddy flux data of carbon, latent heat flux, and MODIS gross primary production (GPP) to reduce the uncertainty of parameter estimation and prediction of TEM. We found that (1) the simulated carbon fluxes are mostly affected by parameters of the maximum rate of photosynthesis (CMAX), the half‐saturation constant for CO2 uptake by plants (kc), the half‐saturation constant for Photosynthetically Active Radiation used by plants (ki), and the change in autotrophic respiration due to 10°C temperature increase (RHQ10); (2) the effect of parameters on seasonal carbon dynamics varies from one parameter to another during a year; (3) to well constrain the uncertainties of TEM predictions and parameters using the Bayesian inference technique, at least two different fluxes of NEP, GPP, and ecosystem respiration (RESP) are required; and (4) different assumptions of the error structures of the flux data used in the Bayesian inference analysis result in different uncertainty bounds of the posterior parameters and model predictions. We further found that, using the Bayesian framework and eddy flux and satellite data, the uncertainty of simulated carbon fluxes has been remarkably reduced. The developed global sensitivity analysis and Bayesian framework could further be used to analyze and improve the predictability and parameterization of relatively coarse time step biogeochemistry models when the eddy flux and satellite data are available for other terrestrial ecosystems.

A Robust Transfer Dictionary Learning Algorithm for Industrial Process Monitoring

The advent of modern data acquisition and computing techniques has enabled high-speed monitoring of high-dimensional processes. The short sampling interval makes the samples temporally correlated, even if there is no underlying autocorrelation among covariates. In this study, we introduce a new process monitoring scheme in a Bayesian framework. The key strategy of this study is to incorporate sequential observations into the estimation procedure for the parameters of interest to update the prior distribution. Based on the updated prior, we obtain the most appropriate estimation of the process parameters at each sampling epoch by maximizing the posterior probability. In addition, conventional statistical process control and monitoring methodologies suffer from the “curse of dimensionality.” The closed form of the estimate developed in this study through Bayesian updates enables the proposed method to be effective for high-dimensional process monitoring. Various simulation studies demonstrate the superiority of the proposed scheme in the high-speed monitoring of high-dimensional processes. Moreover, a few sample paths of the estimated mean in a procedure of the proposed method are illustrated to provide practitioners with insights into the monitoring and control of the process. Finally, we provide a real-life application to illustrate the proposed method.

Domain-transfer learning is a machine learning task to explore a source domain data set to help the learning problem in a target domain. Usually, the source domain has sufficient labeled data, while the target domain does not. In this paper, we propose a novel domain-transfer convolutional model by mapping a target domain data sample to a proxy in the source domain and applying a source domain model to the proxy for the purpose of prediction. In our framework, we firstly represent both source and target domains to feature vectors by two convolutional neural networks and then construct a proxy for each target domain sample in the source domain space. The proxy is supposed to be matched to the corresponding target domain sample convolutional representation vector well. To measure the matching quality, we proposed to maximize their squared-loss mutual information (SMI) between the proxy and target domain samples. We further develop a novel neural SMI estimator based on a parametric density ratio estimation function. Moreover, we also propose to minimize the classification error of both source domain samples and target domain proxies. The classification responses are also smoothened by manifolds of both the source domain and proxy space. By minimizing an objective function of SMI, classification error, and manifold regularization, we learn the convolutional networks of both source and target domains. In this way, the proxy of a target domain sample can be matched to the source domain data and thus benefits from the rich supervision information of the source domain. We design an iterative algorithm to update the parameters alternately and test it over benchmark data sets of abnormal behavior detection in video, Amazon product reviews sentiment analysis, etc.