Search

One of the key objectives in many fields in machine learning is to discover causal relationships among a set of variables from observational data. In linear non-Gaussian acyclic models (LiNGAM), it can be shown that the true underlying causal structure can be identified uniquely from merely observational data. The DirectLiNGAM algorithm is a well-known solution to learn the true causal structure in a high dimensional setting. DirectLiNGAM algorithm executes in a sequence of iterations and it performs a set of comparisons between pairs of variables in each iteration. Unfortunately, the runtime of this algorithm grows significantly as the number of variables increases. In this paper, we propose a parallel algorithm, called ParaLiNGAM, to learn casual structures based on DirectLiNGAM algorithm. We propose a threshold mechanism that can reduce the number of comparisons remarkably compared with the sequential solution. Moreover, in order to further reduce runtime, we employ a messaging mechanism between workers. We also present an implementation of ParaLiNGAM on GPU, considering hardware constraints. Experimental results on synthetic and real data show that our proposed solution outperforms DirectLiNGAM by a factor up to 4788X, and by a median of 2344X. • A scalable algorithm for learning causal structures in linear non-Gaussian models. • Significant speedup by selecting the necessary computations judiciously. • Speedup ratio of up to 4788 on GPUs in real-world and synthetic datasets. [ABSTRACT FROM AUTHOR]

Online learning is an efficient approach in machine learning and statistics, which iteratively updates models upon the observation of a sequence of training examples. A representative online learning algorithm is the online gradient descent, which has found wide applications due to its low complexity and scalability to large datasets. Kernel-based learning methods have been proven to be quite successful in dealing with nonlinearity in the data and multivariate optimization. In this paper we present a class of kernel-based online gradient descent algorithm for addressing regression problems, which generates sparse estimators in an iterative way to reduce the algorithmic complexity for training streaming datasets and model selection in large-scale learning scenarios. In the setting of support vector regression (SVR), we design the sparse online learning algorithm by introducing a sequence of insensitive distance-based loss functions. We prove consistency and error bounds quantifying the generalization performance of such algorithms under mild conditions. The theoretical results demonstrate the interplay between statistical accuracy and sparsity property during learning processes. We show that the insensitive parameter plays a crucial role in providing sparsity as well as fast convergence rates. The numerical experiments also support our theoretical results. [ABSTRACT FROM AUTHOR]

In recent years there has been increased use of machine learning (ML) techniques within mathematics, including symbolic computation where it may be applied safely to optimise or select algorithms. This paper explores whether using explainable AI (XAI) techniques on such ML models can offer new insight for symbolic computation, inspiring new implementations within computer algebra systems that do not directly call upon AI tools. We present a case study on the use of ML to select the variable ordering for cylindrical algebraic decomposition. It has already been demonstrated that ML can make the choice well, but here we show how the SHAP tool for explainability can be used to inform new heuristics of a size and complexity similar to those human-designed heuristics currently commonly used in symbolic computation. [ABSTRACT FROM AUTHOR]

DNA 4 mC plays a crucial role in the genetic expression process of organisms. However, existing deep learning algorithms have shortcomings in the ability to represent DNA sequence features. In this paper, we propose a 4 mC site identification algorithm, DNABert-4mC, based on a fusion of the pruned pre-training DNABert-Pruning model and artificial feature encoding to identify 4 mC sites. The algorithm prunes and compresses the DNABert model, resulting in the pruned pre-training model DNABert-Pruning. This model reduces the number of parameters and removes redundancy from output features, yielding more precise feature representations while upholding accuracy.Simultaneously, the algorithm constructs an artificial feature encoding module to assist the DNABert-Pruning model in feature representation, effectively supplementing the information that is missing from the pre-trained features. The algorithm also introduces the AFF-4mC fusion strategy, which combines artificial feature encoding with the DNABert-Pruning model, to improve the feature representation capability of DNA sequences in multi-semantic spaces and better extract 4 mC sites and the distribution of nucleotide importance within the sequence. In experiments on six independent test sets, the DNABert-4mC algorithm achieved an average AUC value of 93.81%, outperforming seven other advanced algorithms with improvements of 2.05%, 5.02%, 11.32%, 5.90%, 12.02%, 2.42% and 2.34%, respectively. [Display omitted] • Train a new pruning pretraining model on DNA sequences to extract feature information. • The introduction of manually assisted features has expanded the search space for sequence feature. • Attention fusion strategy better promotes the fusion output of features from different dimensions. • The complementary nature of machine features and manual features better fits the target features. [ABSTRACT FROM AUTHOR]

The topic of algorithmic fairness is of increasing importance to the Human–Computer Interaction research community following accumulating concerns regarding the use and deployment of Artificial Intelligence-based systems. How we conduct research on algorithmic fairness directly influences our inferences and conclusions regarding algorithmic fairness. To better understand the methodological decisions of studies focused on people's perceptions of algorithmic fairness, we systematic analysed relevant papers from the CHI and FAccT conferences. We identified 200 relevant papers published between 1993 and 2022 and assessed their study design, participant sample, and geographical location of participants and authors. Our results highlight that studies are predominantly cross-sectional, cover a wide range of participant roles, and that both authors and participants are primarily from the United States. Based on these findings, we reflect on the potential pitfalls and shortcomings in how the community studies algorithmic fairness. • Mapping of CHI and FAccT papers on algorithmic fairness published 1993–2022. • Over one-third of papers miss details on participant locality and compensation. • Algorithmic fairness research largely restricted to US and other Western countries. • Limited cross-country collaboration, most study samples from a single country. • Remote studies are most common, longitudinal studies are relatively rare. [ABSTRACT FROM AUTHOR]

The correlation filter based online tracking methods usually can achieve high real-time performance due to the leverage of the well-known FFT. However, they are also apt to generate the "corrupted training samples" in scenarios with complex background, which will trigger the model drift and deteriorate the tracking accuracy rapidly. The existing methods usually consider this problem from certain aspect and none of them has mined the potential of combining multiple formulas. In this paper, we propose the Output Constraint Transfer - Discriminative Scale Space Tracking (OCT-DSST) algorithm, which has taken full consideration of multiple channel feature, multiple filters, kernel trick, memory with incremental learning, and the self-supervision mechanism. We re-formulate the online tracking process by combining all formulas above in a unified framework. The so obtained adaptive learning rate can better exploit the feedback information coming from the intermediate tracking results, and effectively mitigate the corrupted sample problem. The experimental results on the OTB-50/100 and the VOT2016 datasets reveal that the proposed method is comparative to most state-of-the-arts algorithms, and can increase the accuracy by 2% and the success rate by 1.7%, compared to the traditional DSST method. [ABSTRACT FROM AUTHOR]

Researchers have been involved for decades in search of an efficient skin detection method. However, current methods have not overcome the significant challenges of skin detection, such as variation of illumination, various skin tones of different ethnic groups, and many others. This research proposed a clustering and region-growing-based skin detection method to overcome these limitations. Together with significant insight, these methods result in a more effective algorithm. The insight concerns the capability to dynamically define the number of clusters in a collection of pixels organized as images. In Clustering for most problem domains, the number of clusters is fixed prior and does not perform effectively over a wide variety of data contents. Therefore, this research paper proposed a skin detection method that validated the above findings. The proposed method assigns the number of clusters based on image properties and ultimately allows freedom from manual thresholds or other manual operations. The dynamic determination of clustering outcomes allows for greater automation of skin detection when dealing with uncertain real-world conditions. • Human skin tone is uniform, and clustering is an efficient method to detect human skin. • CIELAB color space is better for separating illumination from the image. • A small number of significant objects can occupy most of the pixel in an image. • Pixel connectivity of an image can automatically decide the number of clusters. • Clustering performs efficiently in Human skin detection. [ABSTRACT FROM AUTHOR]

Most deep learning-based image enhancement algorithms have been developed based on the image-to-image translation approach, in which enhancement processes are difficult to interpret. In this paper, we propose a novel interpretable image enhancement algorithm that estimates multiple transformation functions to describe complex color mapping. First, we develop a histogram-based multiple transformation function estimation network (HMTF-Net) to estimate multiple transformation functions by exploiting both the spatial and statistical information of the input images. Second, we estimate pixel-wise weight maps, which indicate the contribution of each transformation function at each pixel, based on the local structures of the input image and the transformed images obtained by each transformation function. Finally, we obtain the enhanced image as the weighted sum of the transformed images using the estimated weight maps. Extensive experiments confirm the effectiveness of the proposed approach and demonstrate that the proposed algorithm outperforms state-of-the-art image enhancement algorithms for different image enhancement tasks. [ABSTRACT FROM AUTHOR]

The algorithms for mapping carbonates from Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) data use the depths of the diagnostic carbonate absorption features at ∼2.3 μm and ∼2.5 μm. However, because the band depths are estimated using fixed shoulder wavelengths, subtle shifts in band centres caused by different cations in the carbonates could result in false negatives for carbonates or false positives for other minerals that have absorption features in a similar wavelength range (eg. phyllosilicates, zeolites). This paper proposes a new algorithm that is based on the following attributes of carbonate spectra in the 2.0 to 3.0 μm range: (1) presence of two diagnostic overtones features around ∼2.3 μm and ∼2.5 μm; however, these features may show red shift or blue shift depending on the nature of cation(s); (2) the inter band gap between ∼2.3 μm and ∼2.5 μm carbonate absorption features, which remains relatively constant at ∼0.2 μm, even if there is a shift in the absorption features; (3) the contiguity of these two features, that is, carbonate spectra do not show any absorption features in between the above two features. The algorithm also includes a novel geometric continuum removal technique for locating the absorption features. The effectiveness of the algorithm is demonstrated using laboratory spectra, CRISM machine learning toolkit's mineral dataset, as well as CRISM images. The true positive rate (TPR), true negative rate (TNR) and overall accuracy for the method over the CRISM machine learning toolkit's mineral dataset are 29%, 87% and 83%, respectively. • A novel spectral parameter, inter-band-gap, for detecting carbonates is proposed. • A novel rule-based method for detecting carbonates from CRISM data is proposed. • The geometric Hull-based continuum removal method is optimized for CRISM. [ABSTRACT FROM AUTHOR]

• Rethinking the PASCAL-VOC and MS-COCO dataset for small object detection. • There are different labeling errors in these two benchmarks by visual analysis. • Building several specific datasets to analyze small object detection. • Data quality is another factor that affects the performance of small object detection. The data and the algorithm are critical to deep learning-based small object detectors. In this paper, we rethink the PASCAL-VOC and MS-COCO dataset for small object detection. By visual analysis of the original annotations, we find that there are different labeling errors in these two datasets. To solve these problems, we build specific datasets, including SDOD, Mini6K, Mini2022 and Mini6KClean. The experimental results of several typical algorithms (e.g. SSD, YOLOv5, Faster RCNN and Deformable DETR) on the datasets show that data labeling errors (such as missing labels, category label errors, inappropriate labels) are another factor that affects the detection performance of small objects. [ABSTRACT FROM AUTHOR]

This article presents a new method to identify the main patterns describing the flow motion in complex flows. The algorithm is an extension of the higher-order dynamic mode decomposition (HODMD), which compresses the snapshots from the analysed database and progressively updates new compressed snapshots on-the-fly , so it is denoted as HODMD on-the-fly (HODMD-of). This algorithm can be applied in parallel to the numerical simulations (or experiments), and it exhibits two main advantages over offline algorithms: (i) it automatically selects on-the-fly the number of necessary snapshots from the database to identify the relevant dynamics; and (ii) it can be used from the beginning of a numerical simulation (or experiment), since it uses a sliding-window to automatically select, also on-the-fly, the suitable interval to perform the data analysis, i.e. it automatically identifies and discards the transient dynamics. The HODMD-of algorithm is suitable to build reduced order models, which have a much lower computational cost than the original simulation. The performance of the method has been tested in three different cases: the axi-symmetric synthetic jet, the three-dimensional wake of a circular cylinder and the turbulent wake behind a wall-mounted square cylinder. The obtained speed-up factors are around 7 with respect to HODMD; this value depends on the simulation and the configuration of the hyperparameters. HODMD-of also provides a significant reduction of the memory requirements, between 40 − 80 % amongst the two- and three-dimensional cases studied in this paper. • Novel algorithm based on HODMD that combines rank reduction strategies with the on-the-fly methodology. • The algorithm automatically selects the optimal time interval to extract the main flow patterns. • Significant reduction of the memory requirements and speed-up related with standard HODMD. • Suitable to build reduced order models. [ABSTRACT FROM AUTHOR]

This paper presents an algorithm for automated detection and classification of multi-cell phytoliths, one of the major components of many archaeological and paleoenvironmental deposits. This identification, based on phytolith wave pattern, is made using a pretrained VGG19 deep learning model. This approach has been tested in three key phytolith genera for the study of agricultural origins in Near East archaeology: Avena , Hordeum and Triticum. Also, this classification has been validated at species-level using Triticum boeoticum and dicoccoides images. Due to the diversity of microscopes, cameras and chemical treatments that can influence images of phytolith slides, three types of data augmentation techniques have been implemented: rotation of the images at 45-degree angles, random colour and brightness jittering, and random blur/sharpen. The implemented workflow has resulted in an overall accuracy of 93.68% for phytolith genera, improving previous attempts. The algorithm has also demonstrated its potential to automatize the classification of phytoliths species with an overall accuracy of 100%. The open code and platforms employed to develop the algorithm assure the method's accessibility, reproducibility and reusability. • Deep learning automates identification and classification of multi-cell phytoliths. • Can be applied to digitalised slides and integrated into digital microscopes. • Can differentiate between phytolith genera but also species. • Presents an overall accuracy of 93% and analyses the whole slide. • Provides complementary metrics with new relevant information on phytoliths. [ABSTRACT FROM AUTHOR]

In this paper, we propose forward and backward stochastic differential equations (FBSDEs) based deep neural network (DNN) learning algorithms for the solution of high dimensional quasi-linear parabolic partial differential equations (PDEs), which is related to the FBSDEs from the Pardoux-Peng theory. The algorithms rely on a learning process by minimizing the path-wise difference between two discrete stochastic processes, which are defined by the time discretization of the FBSDEs and the DNN representation of the PDE solution, respectively. The proposed algorithms are shown to generate DNN solution for a 100-dimensional Black–Scholes–Barenblatt equation, which is accurate in a finite region in the solution space, and has a convergence rate close to that of the Euler–Maruyama scheme used for discretizing the FBSDEs. • FBSDE deep neural network (DNN) with loss comparing two stochastic processes for the PDEs from Pardoux-Peng theory. • Strong 1/2 Convergence as the Euler-Maruyama method in solving a 100-dimensional Black-Scholes-Barenblatt equation. • Use of Richardson extrapolation for accuracy enhancement. • Multiscale DNN captures temporal high frequency content of the PDEs more accurately. [ABSTRACT FROM AUTHOR]

The progress in artificial intelligence and machine learning algorithms over the past decade has enabled the development of new methods for the objective measurement of eating, including both the measurement of eating episodes as well as the measurement of in-meal eating behavior. These allow the study of eating behavior outside the laboratory in free-living conditions, without the need for video recordings and laborious manual annotations. In this paper, we present a high-level overview of our recent work on intake monitoring using a smartwatch, as well as methods using an in-ear microphone. We also present evaluation results of these methods in challenging, real-world datasets. Furthermore, we discuss use-cases of such intake monitoring tools for advancing research in eating behavior, for improving dietary monitoring, as well as for developing evidence-based health policies. Our goal is to inform researchers and users of intake monitoring methods regarding (i) the development of new methods based on commercially available devices, (ii) what to expect in terms of effectiveness, and (iii) how these methods can be used in research as well as in practical applications. [ABSTRACT FROM AUTHOR]

Functional linear model is a fruitfully applied general framework for regression problems, including those with intrinsically infinite-dimensional data. Online gradient descent methods, despite their evidenced power of processing online or large-sized data, are not well studied for learning with functional data. In this paper, we study reproducing kernel-based online learning algorithms for functional data, and derive convergence rates for the expected excess prediction risk under both online and finite-horizon settings of step-sizes respectively. It is well understood that nontrivial uniform convergence rates for the estimation task depend on the regularity of the slope function. Surprisingly, the convergence rates we derive for the prediction task can assume no regularity from slope. Our analysis reveals the intrinsic difference between the estimation task and the prediction task in functional data learning. [ABSTRACT FROM AUTHOR]

• A hybrid detection method is proposed for assessing delamination in composites. • Two-step method has combined the global frequency and local FBGs for damage detection. • FBGs were distributed near the damage area predicted by frequency index to further predict damage with better accuracy. • Support vector machine and the extreme learning machine were used as inverse algorithms. In this paper, a combined global-local strategy was developed for structural health monitoring (SHM) of fiber reinforced polymer (FRP) composites to assess the internal delamination damage, which has combined vibrational frequency shifts as the global index and the fiber Bragg grating (FBG) wavelength shifts as the local index. The delamination detection was carried out in two steps. The first step was based on the global damage index, which is the changes in multiple modes of frequencies measured by a non-contact scanning laser Doppler vibrometer (SLDV) system. Machine learning (ML) algorithms including support vector machine (SVM) and extreme learning machine (ELM) algorithms were used to initially predict the delamination interface, location, and size through frequency shifts. Numerical and experimental verification results show that SVM has a higher prediction accuracy than ELM when only a small number of samples exist, and the classification of SVM is better than its regression function to be used in the prediction of discrete delamination interfaces. To verify the damaged area predicted by frequency changes and further update the delamination edges with better accuracy, the second step has taken the changes in the wavelength of multiple FBG sensors as the local damage index. Multistage loads were applied at equidistant positions of the FRP specimens to induce deformations and the wavelength shifts in FBGs were used to determine the boundary of delamination. The results showed that delamination in FRP composites can be assessed more precisely using the global-local two-step SHM strategy, which can compensate for the shortcomings of only using vibration-based or FBG-based monitoring techniques. Besides, the SVM algorithm showed excellent predictive performance in both steps and had great potential in delamination damage monitoring. [ABSTRACT FROM AUTHOR]

The deployment of systems experiencing high-rate dynamic events, such as hypersonic vehicles, advanced weaponry, and active blast mitigation systems, require high-rate structural health monitoring (HRSHM) capabilities in the sub-millisecond realm to ensure continuous operations and safety. However, the development of high-rate feedback systems is a complex task because these dynamic systems are uniquely characterized by (1) large uncertainties in their external loads, (2) high levels of non-stationarity and heavy disturbance, and (3) unmodeled dynamics from changes in system configuration. In this paper, we present a deep learning algorithm specifically engineered for HRSHM applications. It consists of an ensemble of recurrent neural networks (RNNs) constructed with long short-term memory (LSTM) cells with transfer learning capabilities to cope with the highly limited availability of training data as it is typical for high-rate systems. The use of an ensemble of RNNs empowers multi-rate sampling capability to capture multi-temporal features of the time series, thus enabling modeling of non-stationarities. Also, because the RNNs use short-sequence LSTMs and are arranged in parallel, the computation time is substantially reduced to the sub-millisecond range. The performance of the algorithm is investigated on experimental high-rate dynamic data produced from accelerated drop tower tests and benchmarked against results from a prior investigation using a purely on-the-edge algorithm termed variable input space observer (VIO) and its hybrid architecture (hybrid VIO) that incorporated physical knowledge and is considered as an upper bound on the algorithm performance. Numerical results show that the ensemble of RNNs, composed of five RNNs, significantly outperformed the VIO, with performance close to that of the hybrid VIO when it comes to the estimation error metrics, with an average computation time of 25 μ s per prediction step. Yet, the ensemble of RNNs exhibits chattering for low-amplitude excitations, likely attributable to the behavior of RNNs sampling at small time delays. An examination of the RNN weights and hidden states confirms that the algorithm can capture multi-temporal features, and an investigation with respect to noise in the training dataset showed that the algorithm is robust for signal-to-noise ratios up to 20 dB. • Machine learning-based architecture for non-stationary time-series analysis. • Transfer learning for accelerated training with limited data. • Ensemble-based system estimation using parallel recurrent neural networks integrated with attention mechanism is proposed. • Sib-millisecond estimation of a high-rate system is achieved though the proposed architecture. [ABSTRACT FROM AUTHOR]

• Machine learning framework for the reconstruction of unsteady fluid flows fields. • Variational approach for a robust reduction of the high dimensional field. • Investigations on three canonical cases. • Combination of linear or nonlinear reduction with regressive or direct reconstruction. This paper investigates the use of data-driven methods for the reconstruction of unsteady fluid flow fields. The proposed framework is based on the combination of machine learning tools: dimensionality reduction to extract dominant spatial directions from data, reconstruction algorithm to recover encoded data by limited measurements and cross-validation for hyperparameter optimization. For the encoding part, linear and nonlinear extraction of patterns are considered: proper orthogonal decomposition (POD), linear autoencoder (LAE) and variational autoencoder (VAE). For the reconstruction part, regressive reconstruction (neural network, linear, support vector, gradient boosting) and library-based reconstruction are compared, each method being cross-validated to ensure good generalization on testing data. The position of sensors is optimized using an enhanced clustering algorithm. The robustness of regressive reconstructions to noise measurements is also addressed, showing the benefits of variational approaches in the reduction phase. The strategy is tested for three increasing complexity flows: 2D vortex shedding (R e = 200), 2D spatial mixing layer and 3D vortex shedding (R e = 20000). The results suggest that proper machine learning approaches to fluid flow data can lead to effective reconstruction models that can be used for the rapid estimation of complex flows. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel algorithm to learn a Büchi automaton from a teacher who knows an ω -regular language. The learned Büchi automaton can be a nondeterministic Büchi automaton or a limit deterministic Büchi automaton. The learning algorithm is based on learning a formalism called family of DFAs (FDFAs) recently proposed by Angluin and Fisman. The main catch is that we use a classification tree structure instead of the standard observation table structure. The worst case storage space required by our algorithm is quadratically better than that required by the table-based algorithm proposed by Angluin and Fisman. We implement the proposed learning algorithms in the learning library ROLL (Regular Omega Language Learning), which also consists of other complete ω -regular learning algorithms available in the literature. Experimental results show that our tree-based learning algorithms have the best performance among others regarding the number of solved learning tasks. [ABSTRACT FROM AUTHOR]

Multimedia event detection is the task of detecting a specific event of interest in an user-generated video on websites. The most fundamental challenge facing this task lies in the enormously varying quality of the video as well as the high-level semantic abstraction of event inherently. In this paper, we decompose the video into several segments and intuitively model the task of complex event detection as a multiple instance learning problem by representing each video as a "bag" of segments in which each segment is referred to as an instance. Instead of treating the instances equally, we associate each instance with a reliability variable to indicate its importance and then select reliable instances for training. To measure the reliability of the varying instances precisely, we propose a visual-semantic guided loss by exploiting low-level feature from visual information together with instance-event similarity based high-level semantic feature. Motivated by curriculum learning, we introduce a negative elastic-net regularization term to start training the classifier with instances of high reliability and gradually taking the instances with relatively low reliability into consideration. An alternative optimization algorithm is developed to solve the proposed challenging non-convex non-smooth problem. Experimental results on standard datasets, i.e., TRECVID MEDTest 2013 and TRECVID MEDTest 2014, demonstrate the effectiveness and superiority of the proposed method to the baseline algorithms. • A visual-semantic guided loss is proposed to measure reliability of instance for event detection. • Training begins with high-reliability instances and gradually added instances of low reliability. • Promising experimental results show the effectiveness and superiority of the proposed method. [ABSTRACT FROM AUTHOR]

• Method (modBIP) presented to pre-train random coefficients of extreme learning machines (ELM). • modBIP can work with any activation function and is more accurate than the BIP method. • ELM/modBIP is highly accurate and accuracy is insensitive to initial random coefficients. In extreme learning machines (ELM) the hidden-layer coefficients are randomly set and fixed, while the output-layer coefficients of the neural network are computed by a least squares method. The randomly-assigned coefficients in ELM are known to influence its performance and accuracy significantly. In this paper we present a modified batch intrinsic plasticity (modBIP) method for pre-training the random coefficients in the ELM neural networks. The current method is devised based on the same principle as the batch intrinsic plasticity (BIP) method, namely, by enhancing the information transmission in every node of the neural network. It differs from BIP in two prominent aspects. First, modBIP does not involve the activation function in its algorithm, and it can be applied with any activation function in the neural network. In contrast, BIP employs the inverse of the activation function in its construction, and requires the activation function to be invertible (or monotonic). The modBIP method can work with the often-used non-monotonic activation functions (e.g. Gaussian, swish, Gaussian error linear unit, and radial-basis type functions), with which BIP breaks down. Second, modBIP generates target samples on random intervals with a minimum size, which leads to highly accurate computation results when combined with ELM. The combined ELM/modBIP method is markedly more accurate than ELM/BIP in numerical simulations. Ample numerical experiments are presented with shallow and deep neural networks for function approximation and boundary/initial value problems with partial differential equations. They demonstrate that the combined ELM/modBIP method produces highly accurate simulation results, and that its accuracy is insensitive to the random-coefficient initializations in the neural network. This is in sharp contrast with the ELM results without pre-training of the random coefficients. [ABSTRACT FROM AUTHOR]

Structural repairs alter the physical properties of a structure, changing its responses, both in terms of its normal condition and of its different damage states. This difference in responses manifests itself as a shift between the pre- and post-repair data distributions, which can be problematic for conventional data-driven approaches to structural health monitoring (SHM), and limits their effectiveness in industrial applications. This limitation occurs typically because approaches assume that the data distribution is the same in training as appears in testing; with an algorithm failing to generalise when this assumption is not true; that is, pre-repair labels no longer apply to the post-repair data. Transfer learning, in the form of domain adaptation, proposes a solution to this issue, by mapping the pre- and post-repair data distributions onto a shared latent space where their distributions are approximately equal, allowing pre-repair label knowledge to be used to classify the post-repair data. This paper demonstrates the applicability of domain adaptation as a method for overcoming the problem of repair on a dataset from a Gnat trainer aircraft. In addition, a novel modification to an existing domain adaptation technique – joint distribution adaptation – is proposed, which seeks to improve the semi-supervised learning phase of the algorithm by considering a metric-informed procedure. The metric-informed joint distribution adaptation algorithm is benchmarked against, and shown to outperform, both conventional data-based approaches and other domain adaptation techniques. • Proposes, for the first time, the application of domain adaptation to overcome the problem of repair in data-based structural health monitoring. • A novel modification to joint domain adaptation, named metric-informed joint domain adaptation , improves initial pseudo-label estimates for the target dataset. • Demonstrates existing and proposed domain adaptation techniques on a repair problem for a Gnat trainer aircraft wing. [ABSTRACT FROM AUTHOR]