Search

There has always been high interest in predicting the solder joint fatigue life in advanced packaging with high accuracy and efficiency. Artificial Intelligence Plus (AI+) is becoming increasingly popular as computational facilities continue to develop. This study will introduce machine learning (a core component of AI). With machine learning, metamodels that approximate the attributes of systems or functions are created to predict the fatigue life of advanced packaging. However, the prediction ability is highly dependent on the size and distribution of the training data. Increasing the amount of training data is the most intuitive approach to improve prediction performance, but this implies a higher computational cost. In this research, the adaptive sampling methods are applied to build the machine learning model with a small dataset sampled from an existing database. The performance of the model will be visualized using predefined criteria. Moreover, ensemble learning can be used to improve the performance of AI models after they have been fully trained. [ABSTRACT FROM AUTHOR]

Recently, fault diagnosing supervised classifiers have been widely proposed to diagnose both electric and mechanical faults in induction motors (IM). However, many of them require a large amount of data, which implies a great effort required for processing fault-related features and building the training set. Furthermore, in real-world datasets, it is required to deal with highly skewed data distributions, also known as class imbalance, which is a limiting issue and can misguide the tuning of machine learning algorithms. Resampling techniques based on a synthetic generation of minority class observations aim to address this problem. Last but not least is the fact that inverter-fed IM introduces undesired harmonics in the monitoring signal altering the diagnosis patterns. This diagnosis scheme is evaluated on experimental imbalanced data oriented to deal with the diagnosis of a rotor in situations where it is fed with an inverter. The results show how this imbalanced approach determines the actual diagnosis performance on a small amount of data. The experimental results demonstrate that balanced training sets built with class balancing techniques improve the classifier and therefore its performance for diagnosing incipient rotor faults in inverter-fed IM with studied and interpretable features recently proposed in this field of study. [ABSTRACT FROM AUTHOR]

Background: Texture analysis extracts many quantitative image features, offering a valuable, cost-effective, and non-invasive approach for individual medicine. Furthermore, multimodal machine learning could have a large impact for precision medicine, as texture biomarkers can underlie tissue microstructure. This study aims to investigate imaging-based biomarkers of radio-induced neurotoxicity in pediatric patients with metastatic medulloblastoma, using radiomic and dosiomic analysis. Methods: This single-center study retrospectively enrolled children diagnosed with metastatic medulloblastoma (MB) and treated with hyperfractionated craniospinal irradiation (CSI). Histological confirmation of medulloblastoma and baseline follow-up magnetic resonance imaging (MRI) were mandatory. Treatment involved helical tomotherapy (HT) delivering a dose of 39 Gray (Gy) to brain and spinal axis and a posterior fossa boost up to 60 Gy. Clinical outcomes, such as local and distant brain control and neurotoxicity, were recorded. Radiomic and dosiomic features were extracted from tumor regions on T1, T2, FLAIR (fluid-attenuated inversion recovery) MRI-maps, and radiotherapy dose distribution. Different machine learning feature selection and reduction approaches were performed for supervised and unsupervised clustering. Results: Forty-eight metastatic medulloblastoma patients (29 males and 19 females) with a mean age of 12 ± 6 years were enrolled. For each patient, 332 features were extracted. Greater level of abstraction of input data by combining selection of most performing features and dimensionality reduction returns the best performance. The resulting one-component radiomic signature yielded an accuracy of 0.73 with sensitivity, specificity, and precision of 0.83, 0.64, and 0.68, respectively. Conclusions: Machine learning radiomic-dosiomic approach effectively stratified pediatric medulloblastoma patients who experienced radio-induced neurotoxicity. Strategy needs further validation in external dataset for its potential clinical use in ab initio management paradigms of medulloblastoma. [ABSTRACT FROM AUTHOR]

In this work, we propose to tackle the problem of domain generalization in the context of insufficient samples. Instead of extracting latent feature embeddings based on deterministic models, we propose to learn a domain-invariant representation based on the probabilistic framework by mapping each data point into probabilistic embeddings. Specifically, we first extend empirical maximum mean discrepancy (MMD) to a novel probabilistic MMD that can measure the discrepancy between mixture distributions (i.e., source domains) consisting of a series of latent distributions rather than latent points. Moreover, instead of imposing the contrastive semantic alignment (CSA) loss based on pairs of latent points, a novel probabilistic CSA loss encourages positive probabilistic embedding pairs to be closer while pulling other negative ones apart. Benefiting from the learned representation captured by probabilistic models, our proposed method can marriage the measurement on the distribution over distributions (i.e., the global perspective alignment) and the distribution-based contrastive semantic alignment (i.e., the local perspective alignment). Extensive experimental results on three challenging medical datasets show the effectiveness of our proposed method in the context of insufficient data compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Prognostics and health management (PHM) is critical for enhancing equipment reliability and reducing maintenance costs, and research on intelligent PHM has made significant progress driven by big data and deep learning techniques in recent years. However, complex working conditions and high-cost data collection inherent in real-world scenarios pose small-data challenges for the application of these methods. Given the urgent need for data-efficient PHM techniques in academia and industry, this paper aims to explore the fundamental concepts, ongoing research, and future trajectories of small data challenges in the PHM domain. This survey first elucidates the definition, causes, and impacts of small data on PHM tasks, and then analyzes the current mainstream approaches to solving small data problems, including data augmentation, transfer learning, and few-shot learning techniques, each of which has its advantages and disadvantages. In addition, this survey summarizes benchmark datasets and experimental paradigms to facilitate fair evaluations of diverse methodologies under small data conditions. Finally, some promising directions are pointed out to inspire future research. [ABSTRACT FROM AUTHOR]

This study presents a method for estimating construction costs, even when dealing with limited and unreliable data, to enhance decision-making in the early project stages. Owners, particularly in green building projects, often face challenges due to the scarcity of usable data, making cost estimation a complex task. They struggle to differentiate between costs associated with existing buildings and green buildings. To address this issue, we introduce a novel approach that leverages conditional tabular generative adversarial networks (CTGANs) for data augmentation, overcoming the limitations of relying solely on historical data. This involves training an artificial neural network (ANN)–based model using synthetic data, effectively addressing the scarcity and imbalance present in the original small data set. Compared to models trained exclusively on the original data set, our approach yielded a remarkable reduction of approximately 66% in root-mean-square error (RMSE), while increasing the validity from 0% to 15.09%. This study not only improves construction cost estimation but also facilitates more informed decision-making for owners, even in cases with limited and unreliable data, ultimately contributing to the efficiency of the construction project planning process. [ABSTRACT FROM AUTHOR]

In this work, we present a novel methodology for performing the supervised classification of time-ordered noisy data; we call this methodology Entropic Sparse Probabilistic Approximation with Markov regularization (eSPA-Markov). It is an extension of entropic learning methodologies, allowing the simultaneous learning of segmentation patterns, entropy-optimal feature space discretizations, and Bayesian classification rules. We prove the conditions for the existence and uniqueness of the learning problem solution and propose a one-shot numerical learning algorithm that—in the leading order—scales linearly in dimension. We show how this technique can be used for the computationally scalable identification of persistent (metastable) regime affiliations and regime switches from high-dimensional non-stationary and noisy time series, i.e., when the size of the data statistics is small compared to their dimensionality and when the noise variance is larger than the variance in the signal. We demonstrate its performance on a set of toy learning problems, comparing eSPA-Markov to state-of-the-art techniques, including deep learning and random forests. We show how this technique can be used for the analysis of noisy time series from DNA and RNA Nanopore sequencing. [ABSTRACT FROM AUTHOR]

Machine learning (ML) has been harnessed as a promising modelling tool for materials research. However, small data, or data scarcity, is a bottleneck when incorporating ML in studies involving experimentation. Current experiment planning methods show several disadvantages: one‐factor‐at‐a‐time (OFAT) experimentation became impractical due to limited laboratory resources; conventional design of experiments (DoE) failed to incorporate high‐dimensional features in ML; Surrogate‐based or Bayesian optimization (BO) shifted the goal to optimize material properties rather than guiding training data accumulation. The present research proposes leveraging active learning (AL) to strategically select critical data for experimentation. Two AL strategies, query‐by‐Committee (QBC) algorithm and Greedy method, are benchmarked against random query baseline on various materials datasets. AL is shown to efficiently reduce model prediction errors with minimal additional experiment data. Investigation of hyperparameters revealed benefits of applying AL at an early stage of experimental dataset construction. Moreover, AL is implemented and validated for an in‐house materials development task ‐ electrospray modelling. AL exploration as a paradigm is highlighted to guide experiment design for efficient data accumulation purposes, and its potential for further ML modelling. In doing so, the power of ML is expected to be fully unleashed to experimental researchers. [ABSTRACT FROM AUTHOR]

The approaches used in biomedicine to analyze epidemics take into account features such as exponential growth in the early stages, slowdown in dynamics upon saturation, time delays in spread, segmented spread, evolutionary adaptations of the pathogen, and preventive measures based on universal communication protocols. All these characteristics are also present in modern cyber epidemics. Therefore, adapting effective biomedical approaches to epidemic analysis for the investigation of the development of cyber epidemics is a promising scientific research task. The article is dedicated to researching the problem of predicting the development of cyber epidemics at early stages. In such conditions, the available data is scarce, incomplete, and distorted. This situation makes it impossible to use artificial intelligence models for prediction. Therefore, the authors propose an entropy-extreme model, defined within the machine learning paradigm, to address this problem. The model is based on estimating the probability distributions of its controllable parameters from input data, taking into account the variability characteristic of the last ones. The entropy-extreme instance, identified from a set of such distributions, indicates the most uncertain (most negative) trajectory of the investigated process. Numerical methods are used to analyze the generated set of investigated process development trajectories based on the assessments of probability distributions of the controllable parameters and the variability characteristic. The result of the analysis includes characteristic predictive trajectories such as the average and median trajectories from the set, as well as the trajectory corresponding to the standard deviation area of the parameters’ values. Experiments with real data on the infection of Windows-operated devices by various categories of malware showed that the proposed model outperforms the classical competitor (least squares method) in predicting the development of cyber epidemics near the extremum of the time series representing the deployment of such a process over time. Moreover, the proposed model can be applied without any prior hypotheses regarding the probabilistic properties of the available data.

This research develops a data-driven methodology for structural instability problems with highly nonlinear, difficult, noisy, and small data. A fast analysis and prediction (FAP) approach for instability problems of thin shells is first proposed. This approach contains two phases: the fast numerical analysis and the pure prediction utilizing artificial neural networks (ANNs) incorporated with the Bayesian regularization (B-R) algorithm as follows: (1) in Phase 1 (the fast numerical analysis), post-buckling analysis is conducted utilizing a minor amount of load steps. The load–displacement relation achieved from Phase 1 is not exact because of the small number of load steps utilized; (2) in Phase 2 (the prediction), the loads and deflections achieved from Phase 1 were employed as the data for training ANNs. The trained networks, including the load and displacement networks, were employed to fast predict loads and deflections at any step of the post-buckling analysis. After utilizing Phase 2, a smooth, complete and exact load–displacement curve was achieved. In Phase 1, the available formulation for post-buckling analysis of thin shells in the literature was utilized. Five popular types of instabilities chosen to confirm the effectiveness and exactness of the FAP were snap-through, snap-back, softening–hardening, kink instabilities, and delamination buckling and post-buckling of composites. The high exactness and effectiveness of the FAP were confirmed in the numerical verification section. The present approach saves a huge computation compared to the other ones. It was found that ANNs incorporated with the B-R algorithm have notable advantages compared to numerous neural networks. The proposed approach is applicable to simulations or experiments where data are “expensive”, highly nonlinear, difficult, and limited. Utilizing the proposed approach for these fields can dramatically save time and money.

In the era of big data processing, numerous techniques prove valuable for analyzing large-scale datasets. However, the efficient processing of small data is equally crucial, particularly in the context of intelligent mobile systems. This importance stems from resource constraints, real-time requirements, privacy concerns, and the emergence of edge computing paradigms. In this paper, we adapt an existing input doubling method to address classification tasks when facing limited training datasets. The proposed solution is implemented by modifying the procedure for aggregating predicted values, based on the principles of majority voting. This approach differs significantly from the existing method, which relies on averaging prediction results. A flowchart of the adapted method is presented, detailing the procedures for training and application based on the use of a plurality voting scheme. Modeling is conducted on a short, imbalanced dataset for solving express diagnostic tasks in the field of transplantology. The results demonstrate the high effectiveness of employing the adapted input doubling method for classification tasks in transplantology, where acquiring a substantial training dataset for analysis is challenging. [ABSTRACT FROM AUTHOR]

The emergence of SARS-CoV-2 variants during the COVID-19 pandemic caused frequent global outbreaks that confounded public health efforts across many jurisdictions, highlighting the need for better understanding and prediction of viral evolution. Predictive models have been shown to support disease prevention efforts, such as with the seasonal influenza vaccine, but they require abundant data. For emerging viruses of concern, such models should ideally function with relatively sparse data typically encountered at the early stages of a viral outbreak. Conventional discrete approaches have proven difficult to develop due to the spurious and reversible nature of amino acid mutations and the overwhelming number of possible protein sequences adding computational complexity. We hypothesized that these challenges could be addressed by encoding discrete protein sequences into continuous numbers, effectively reducing the data size while enhancing the resolution of evolutionarily relevant differences. To this end, we developed a viral protein evolution prediction model (VPRE), which reduces amino acid sequences into continuous numbers by using an artificial neural network called a variational autoencoder (VAE) and models their most statistically likely evolutionary trajectories over time using Gaussian process (GP) regression. To demonstrate VPRE, we used a small amount of early SARS-CoV-2 spike protein sequences. We show that the VAE can be trained on a synthetic dataset based on this data. To recapitulate evolution along a phylogenetic path, we used only 104 spike protein sequences and trained the GP regression with the numerical variables to project evolution up to 5 months into the future. Our predictions contained novel variants and the most frequent prediction mapped primarily to a sequence that differed by only a single amino acid from the most reported spike protein within the prediction timeframe. Novel variants in the spike receptor binding domain (RBD) were capable of binding human angiotensin-converting enzyme 2 (ACE2) in silico, with comparable or better binding than previously resolved RBD-ACE2 complexes. Together, these results indicate the utility and tractability of combining deep learning and regression to model viral protein evolution with relatively sparse datasets, toward developing more effective medical interventions. [ABSTRACT FROM AUTHOR]

Machine learning (ML) has been harnessed as a promising modelling tool for materials research. However, small data, or data scarcity, is a bottleneck when incorporating ML in studies involving experimentation. Current experiment planning methods show several disadvantages: one‐factor‐at‐a‐time (OFAT) experimentation became impractical due to limited laboratory resources; conventional design of experiments (DoE) failed to incorporate high‐dimensional features in ML; Surrogate‐based or Bayesian optimization (BO) shifted the goal to optimize material properties rather than guiding training data accumulation. The present research proposes leveraging active learning (AL) to strategically select critical data for experimentation. Two AL strategies, query‐by‐Committee (QBC) algorithm and Greedy method, are benchmarked against random query baseline on various materials datasets. AL is shown to efficiently reduce model prediction errors with minimal additional experiment data. Investigation of hyperparameters revealed benefits of applying AL at an early stage of experimental dataset construction. Moreover, AL is implemented and validated for an in‐house materials development task ‐ electrospray modelling. AL exploration as a paradigm is highlighted to guide experiment design for efficient data accumulation purposes, and its potential for further ML modelling. In doing so, the power of ML is expected to be fully unleashed to experimental researchers.

A novel approach has been introduced to estimate the parameters of exponential and DN distributions during the rejection testing of electronic devices, accompanied by a detailed procedure for its implementation. This innovative method enhances noise immunity and minimizes the error associated with the rejection process through the application of a clustering technique involving wavelet transform. The effectiveness of the method has been verified using resistors, employing criteria such as noise level and stability. The substantial improvement in noise immunity and the reduction in rejection procedure errors are achieved by incorporating an adaptive clustering method coupled with wavelet transform. Notably, in clustering with a signal-to-noise ratio by amplitude of 1.17, the relative error in determining the minimum of the test function was reduced to 8.32%. These promising outcomes substantiate the recommendation of the developed method for the automated selection of resistors, particularly those designated for long-term operational equipment with critical applications. The presented method thus contributes significantly to enhancing the reliability and accuracy of electronic device testing and selection processes.

This study proposes a discrete mathematical Susceptible–Exposed–Preinfectious–Infectious–Recovered (SEPIR) states model for seasonal influenza. In a previous study, focusing on infections by preinfectious people using preexisting data, the author showed that the super-spreading of seasonal influenza occurred before the day that the first patients were discovered (D-day). In addition, when people do not take precautionary measures, the infectivity rate (from preinfected people) was determined as 0.041. After D-day in the community, the implementation of countermeasures was observed to reduce the infectivity rate to 0.002 and 0.013 in working and living spaces, respectively. The number of infectious people can be estimated by summing up each group in the community. This study performed a multiagent simulation (MAS) of seasonal influenza from preinfectious people in closed spaces based on decomposability. Then, the basic simulation is validated and the appropriateness of infective rates, changed infectivity rates, and near decomposability is confirmed.

Consider the second order nonlinear partial differential equation: $ \partial_t^2 u = F(u, \partial_x u), \quad (t, x) \in \mathbb{C}\times \mathbb{R}. $ Given small analytic data, Yamane was able to obtain the order of the lifespan of the solution with respect to the smallness parameter $ \varepsilon $. On the other hand, Gourdin and Mechab studied the lifespan of the solution given small Gevrey data, but under the assumption that $ F $ is independent of $ u $. In this paper, we considered non-vanishing Gevrey data and used the method of successive approximations to obtain a solution and constructively estimate its lifespan.

The incompressible Schrödinger flow is a Madelung's hydrodynamical form of quantum mechanics, which can simulate classical fluids with particular advantage in its simplicity and its ability of capturing thin vortex dynamics. This model enables robust simulation of intricate phenomena such as vortical wakes and interacting vortex filaments. In this article, we prove the global regularity and asymptotic behaviors for incompressible Schrödinger flow with small and localized data in three dimensions. We choose a suitable gauge to rewrite the system, and then use Fourier analysis and vector field method to prove global existence and asymptotic behavior. [ABSTRACT FROM AUTHOR]

Playing a crucial role in ocean activities, internal solitary waves (ISWs) are of significant importance. Currently, the use of deep learning for detecting ISWs in synthetic aperture radar (SAR) imagery is gaining growing attention. However, these approaches often demand a considerable number of labeled images, which can be challenging to acquire in practice. In this study, we propose an innovative method employing a pyramidal conditional generative adversarial network (PCGAN). At each scale, it employs the framework of a conditional generative adversarial network (CGAN), comprising a generator and a discriminator. The generator works to produce internal wave patterns as authentically as possible, while the discriminator is designed to differentiate between images generated by the generator and reference images. The architecture based on pyramids adeptly captures the encompassing as well as localized characteristics of internal waves. The incorporation of upsampling further bolsters the model's ability to recognize fine-scale internal wave stripes. These attributes endow the PCGAN with the capacity to learn from a limited amount of internal wave observation data. Experimental results affirm that the PCGAN, trained with just four internal wave images, can accurately detect internal wave stripes in the test set. Through comparative experiments with other segmentation models, we demonstrate the effectiveness and robustness of PCGAN. [ABSTRACT FROM AUTHOR]

Artificial intelligence is gaining traction in bearing fault detection and diagnosis. Generally, signal processing and feature selection are carried out to facilitate the fault classification process; however, classification accuracy tends to degrade under limited training data. In this paper, various artificial intelligence (AI) classification models are studied and compared using raw vibration data without signal processing and feature engineering. A Cosine k-Nearest Neighbours (CosKNN)- based classification model is optimized by integrating a Segmentive Mechanism, resulting in an overall classification F1-score improvement to 90.8% compared to the original classifier's 76.9%. The comparative findings show that the proposed model is suitable for circumstances with limited availability of training data, signal processing tools, and feature engineering tuning. [ABSTRACT FROM AUTHOR]

The purpose of this study was to investigate the possibility of implementing an artificial intelligence (AI) approach for the analysis of fluorescence lifetime imaging ophthalmoscopy (FLIO) data even with small data. FLIO data, including the fluorescence intensity and mean fluorescence lifetime (τm) of two spectral channels, as well as OCT-A data from 26 non-smokers and 28 smokers without systemic and ocular diseases were used. The analysis was performed with support vector machines (SVMs), a well-known AI method for small datasets, and compared with the results of convolutional neural networks (CNNs) and autoencoder networks. The SVM was the only tested AI method, which was able to distinguish τm between non-smokers and heavy smokers. The accuracy was about 80%. OCT-A data did not show significant differences. The feasibility and usefulness of the AI in analyzing FLIO and OCT-A data without any apparent retinal diseases were demonstrated. Although further studies with larger datasets are necessary to validate the results, the results greatly suggest that AI could be useful in analyzing FLIO-data even from healthy subjects without retinal disease and even with small datasets. AI-assisted FLIO is expected to greatly advance early retinal diagnosis. [ABSTRACT FROM AUTHOR]

This study proposes a discrete mathematical Susceptible–Exposed–Preinfectious–Infectious–Recovered (SEPIR) states model for seasonal influenza. In a previous study, focusing on infections by preinfectious people using preexisting data, the author showed that the super-spreading of seasonal influenza occurred before the day that the first patients were discovered (D-day). In addition, when people do not take precautionary measures, the infectivity rate (from preinfected people) was determined as 0.041. After D-day in the community, the implementation of countermeasures was observed to reduce the infectivity rate to 0.002 and 0.013 in working and living spaces, respectively. The number of infectious people can be estimated by summing up each group in the community. This study performed a multiagent simulation (MAS) of seasonal influenza from preinfectious people in closed spaces based on decomposability. Then, the basic simulation is validated and the appropriateness of infective rates, changed infectivity rates, and near decomposability is confirmed. [ABSTRACT FROM AUTHOR]

Consider the second order nonlinear partial differential equation:... Given small analytic data, Yamane was able to obtain the order of the lifespan of the solution with respect to the smallness parameter e. On the other hand, Gourdin and Mechab studied the lifespan of the solution given small Gevrey data, but under the assumption that F is independent of u. In this paper, we considered non-vanishing Gevrey data and used the method of successive approximations to obtain a solution and constructively estimate its lifespan. [ABSTRACT FROM AUTHOR]

The emergence of SARS-CoV-2 variants during the COVID-19 pandemic caused frequent global outbreaks that confounded public health efforts across many jurisdictions, highlighting the need for better understanding and prediction of viral evolution. Predictive models have been shown to support disease prevention efforts, such as with the seasonal influenza vaccine, but they require abundant data. For emerging viruses of concern, such models should ideally function with relatively sparse data typically encountered at the early stages of a viral outbreak. Conventional discrete approaches have proven difficult to develop due to the spurious and reversible nature of amino acid mutations and the overwhelming number of possible protein sequences adding computational complexity. We hypothesized that these challenges could be addressed by encoding discrete protein sequences into continuous numbers, effectively reducing the data size while enhancing the resolution of evolutionarily relevant differences. To this end, we developed a viral protein evolution prediction model (VPRE), which reduces amino acid sequences into continuous numbers by using an artificial neural network called a variational autoencoder (VAE) and models their most statistically likely evolutionary trajectories over time using Gaussian process (GP) regression. To demonstrate VPRE, we used a small amount of early SARS-CoV-2 spike protein sequences. We show that the VAE can be trained on a synthetic dataset based on this data. To recapitulate evolution along a phylogenetic path, we used only 104 spike protein sequences and trained the GP regression with the numerical variables to project evolution up to 5 months into the future. Our predictions contained novel variants and the most frequent prediction mapped primarily to a sequence that differed by only a single amino acid from the most reported spike protein within the prediction timeframe. Novel variants in the spike receptor binding domain (RBD) were capable of binding human angiotensin-converting enzyme 2 (ACE2) in silico, with comparable or better binding than previously resolved RBD-ACE2 complexes. Together, these results indicate the utility and tractability of combining deep learning and regression to model viral protein evolution with relatively sparse datasets, toward developing more effective medical interventions.

Recently, fault diagnosing supervised classifiers have been widely proposed to diagnose both electric and mechanical faults in induction motors (IM). However, many of them require a large amount of data, which implies a great effort required for processing fault-related features and building the training set. Furthermore, in real-world datasets, it is required to deal with highly skewed data distributions, also known as class imbalance, which is a limiting issue and can misguide the tuning of machine learning algorithms. Resampling techniques based on a synthetic generation of minority class observations aim to address this problem. Last but not least is the fact that inverter-fed IM introduces undesired harmonics in the monitoring signal altering the diagnosis patterns. This diagnosis scheme is evaluated on experimental imbalanced data oriented to deal with the diagnosis of a rotor in situations where it is fed with an inverter. The results show how this imbalanced approach determines the actual diagnosis performance on a small amount of data. The experimental results demonstrate that balanced training sets built with class balancing techniques improve the classifier and therefore its performance for diagnosing incipient rotor faults in inverter-fed IM with studied and interpretable features recently proposed in this field of study.

There has always been high interest in predicting the solder joint fatigue life in advanced packaging with high accuracy and efficiency. Artificial Intelligence Plus (AI+) is becoming increasingly popular as computational facilities continue to develop. This study will introduce machine learning (a core component of AI). With machine learning, metamodels that approximate the attributes of systems or functions are created to predict the fatigue life of advanced packaging. However, the prediction ability is highly dependent on the size and distribution of the training data. Increasing the amount of training data is the most intuitive approach to improve prediction performance, but this implies a higher computational cost. In this research, the adaptive sampling methods are applied to build the machine learning model with a small dataset sampled from an existing database. The performance of the model will be visualized using predefined criteria. Moreover, ensemble learning can be used to improve the performance of AI models after they have been fully trained.

Quality control is essential for modern business success. The traditional statistical process control (SPC), however, lacks efficacy in current high-variety low-volume industrial practices since the historical reference data in Phase I are usually too scarce to infer the in-control process parameters accurately. To solve this 'small data' challenge, a novel Bayesian process monitoring scheme via transfer learning is proposed to facilitate a cross-product data sharing. In particular, a joint prior distribution is taken to explicitly capture the relatedness between the process data of two similar products, through which the process information can be transferred from one product (source domain) to improve the Bayesian inference for the other product (target domain). The posteriors can be derived analytically in closed forms by using generalised hypergeometric functions, thereby leading to a computationally efficient control chart for the online real-time monitoring in Phase II. A user-specified parameter is also provided to enable a better theoretical understanding of the transferability matter and a free practical control of the transferred information across domains. Extensive numerical simulations and real example studies of an assembly process validate the superiority of our proposed scheme in terms of both the false alarm rate and detection capability. [ABSTRACT FROM AUTHOR]

Measurement is a typical way of gathering information about an investigated object, generalized by a finite set of characteristic parameters. The result of each iteration of the measurement is an instance of the class of the investigated object in the form of a set of values of characteristic parameters. An ordered set of instances forms a collection whose dimensionality for a real object is a factor that cannot be ignored. Managing the dimensionality of data collections, as well as classification, regression, and clustering, are fundamental problems for machine learning. Compactification is the approximation of the original data collection by an equivalent collection (with a reduced dimension of characteristic parameters) with the control of accompanying information capacity losses. Related to compactification is the data completeness verifying procedure, which is characteristic of the data reliability assessment. If there are stochastic parameters among the initial data collection characteristic parameters, the compactification procedure becomes more complicated. To take this into account, this study proposes a model of a structured collection of stochastic data defined in terms of relative entropy. The compactification of such a data model is formalized by an iterative procedure aimed at maximizing the relative entropy of sequential implementation of direct and reverse projections of data collections, taking into account the estimates of the probability distribution densities of their attributes. The procedure for approximating the relative entropy function of compactification to reduce the computational complexity of the latter is proposed. To qualitatively assess compactification this study undertakes a formal analysis that uses data collection information capacity and the absolute and relative share of information losses due to compaction as its metrics. Taking into account the semantic connection of compactification and completeness, the proposed metric is also relevant for the task of assessing data reliability. Testing the proposed compactification procedure proved both its stability and efficiency in comparison with previously used analogues, such as the principal component analysis method and the random projection method. [ABSTRACT FROM AUTHOR]

In this work, we present a novel methodology for performing the supervised classification of time-ordered noisy data; we call this methodology Entropic Sparse Probabilistic Approximation with Markov regularization (eSPA-Markov). It is an extension of entropic learning methodologies, allowing the simultaneous learning of segmentation patterns, entropy-optimal feature space discretizations, and Bayesian classification rules. We prove the conditions for the existence and uniqueness of the learning problem solution and propose a one-shot numerical learning algorithm that—in the leading order—scales linearly in dimension. We show how this technique can be used for the computationally scalable identification of persistent (metastable) regime affiliations and regime switches from high-dimensional non-stationary and noisy time series, i.e., when the size of the data statistics is small compared to their dimensionality and when the noise variance is larger than the variance in the signal. We demonstrate its performance on a set of toy learning problems, comparing eSPA-Markov to state-of-the-art techniques, including deep learning and random forests. We show how this technique can be used for the analysis of noisy time series from DNA and RNA Nanopore sequencing.

Background: The distinction between small data and big data is increasingly muted and has caused challenges and confusion in many quarters. Objective: The objective of the study is to gain a deeper understanding of the confounded confusion that exists between small data and big data. Firstly, to develop a taxonomy that distinguishes between small data and big data. Secondly, it seeks to extract the value from the concepts, which can be of fundamental importance to an organisation. Methods: This study follows the interpretive approach and employs qualitative methods, based on which 57 related materials were gathered, covering big data and small data, and analysed. Results: The study reveals the factors that differentiate the concepts, which are of a technical front, business logic and data processing. Conclusion: This study addresses the challenges which are increasingly of prohibitive ramifications for both academic and business domains. By removing the confusion, the classifications of small data and big data including associated attributes will be better understood. This increases their business use towards enhancement and competitive advantage. Contribution: The article distinguishes between small data and big data, which has been missing, from both academic and business perspectives, since the emergence of the latter. The differentiation between small data and big data provides a guide to organisations in developing strategic frameworks and operational plans.

Big data in drug development may not satisfactorily address the demands of precision medicine in a rare disease population, making the use of smaller clinical trials necessary. Consequently, the use of innovative design and analysis of these clinical trials using model‐informed approaches have become indispensable. This requires informative exposure–outcome analysis, together with formal statistical analysis, which should include the strength of evidence for a study outcome. We demonstrate how knowledge can be gained, with supporting strength of evidence, from a small (data) clinical trial with a low dose of blarcamesine in the treatment of Rett syndrome. Based on a small data paradigm, pharmacometrics item response theory modelling and Bayes factor analysis were used to demonstrate the efficacy of blarcamesine in Rett syndrome. [ABSTRACT FROM AUTHOR]

The specific context and characteristics of ecological studies mean datasets are often small and poorly suited to many advanced analytical approaches. Some important ecological insights can only be derived from Small Data, so its collection should not be neglected. Ongoing advances in study design, data analysis, and machine learning methods increase the value of 'Small Data' beyond the intrinsic purposes for which they are originally collected. These methods include transfer learning, generating synthetic data, using causal model structures and new approaches to collating Small Data. The greatest opportunities will arise if the reusability of Small Data is improved. Big Data science has significantly furthered our understanding of complex systems by harnessing large volumes of data, generated at high velocity and in great variety. However, there is a risk that Big Data collection is prioritised to the detriment of 'Small Data' (data with few observations). This poses a particular risk to ecology where Small Data abounds. Machine learning experts are increasingly looking to Small Data to drive the next generation of innovation, leading to development in methods for Small Data such as transfer learning, knowledge graphs, and synthetic data. Meanwhile, meta-analysis and causal reasoning approaches are evolving to provide new insights from Small Data. These advances should add value to high-quality Small Data catalysing future insights for ecology. [ABSTRACT FROM AUTHOR]

3D reconstruction is a useful tool for surgical planning and guidance. Supervised methods for disparity/depth estimation are the state of the art, with demonstrated performances far superior to all alternatives, such as self supervised and traditional geometric methods. However, supervised training requires large datasets, and in this field, data is lacking. In this paper, we investigate the learning of structured light projections to enhance the development of disparity estimation networks. Improving supervised learning on small datasets without needing to collect extra data. We first show that it is possible to learn the projection of structured light on a scene. Secondly, we show that the joint training of structured light and disparity, using a multi-task learning (MTL) framework, improves the learning of disparity. Our MTL setup outperformed the single task learning (STL) network in every validation test. Notably, in the generalisation test, the STL error was 1.4 times worse than that of the best MTL performance. A dataset containing stereoscopic images, disparity maps and structured light projections on medical phantoms and ex vivo tissue was created for evaluation together with virtual scenes. This dataset will be made publicly available in the future. [ABSTRACT FROM AUTHOR]

Playing a crucial role in ocean activities, internal solitary waves (ISWs) are of significant importance. Currently, the use of deep learning for detecting ISWs in synthetic aperture radar (SAR) imagery is gaining growing attention. However, these approaches often demand a considerable number of labeled images, which can be challenging to acquire in practice. In this study, we propose an innovative method employing a pyramidal conditional generative adversarial network (PCGAN). At each scale, it employs the framework of a conditional generative adversarial network (CGAN), comprising a generator and a discriminator. The generator works to produce internal wave patterns as authentically as possible, while the discriminator is designed to differentiate between images generated by the generator and reference images. The architecture based on pyramids adeptly captures the encompassing as well as localized characteristics of internal waves. The incorporation of upsampling further bolsters the model’s ability to recognize fine-scale internal wave stripes. These attributes endow the PCGAN with the capacity to learn from a limited amount of internal wave observation data. Experimental results affirm that the PCGAN, trained with just four internal wave images, can accurately detect internal wave stripes in the test set. Through comparative experiments with other segmentation models, we demonstrate the effectiveness and robustness of PCGAN.

The purpose of this study was to investigate the possibility of implementing an artificial intelligence (AI) approach for the analysis of fluorescence lifetime imaging ophthalmoscopy (FLIO) data even with small data. FLIO data, including the fluorescence intensity and mean fluorescence lifetime (τm) of two spectral channels, as well as OCT-A data from 26 non-smokers and 28 smokers without systemic and ocular diseases were used. The analysis was performed with support vector machines (SVMs), a well-known AI method for small datasets, and compared with the results of convolutional neural networks (CNNs) and autoencoder networks. The SVM was the only tested AI method, which was able to distinguish τm between non-smokers and heavy smokers. The accuracy was about 80%. OCT-A data did not show significant differences. The feasibility and usefulness of the AI in analyzing FLIO and OCT-A data without any apparent retinal diseases were demonstrated. Although further studies with larger datasets are necessary to validate the results, the results greatly suggest that AI could be useful in analyzing FLIO-data even from healthy subjects without retinal disease and even with small datasets. AI-assisted FLIO is expected to greatly advance early retinal diagnosis.

The article offers information on the hypothesis that committing a son to lifelong religious celibacy may be advantageous for parents in increasing their reproductive success. The authors formalize this hypothesis with an inclusive fitness model and test it in a Tibetan population. They find evidence supporting their model assumptions, suggesting that the decision to commit to celibacy is often made by parents, and men with a monk brother have higher reproductive success.

Small data and big data are distinct approaches to data analysis and utilization in various applications. While big data has been the focus of many research and business efforts for more than ten years, small data is increasingly being recognized as having potential value in certain settings. We systematically review literature and conclude that small data can be indeed valuable in certain scenarios. This paper incorporates the data value perspective of small data within various application areas. For this, we apply the data-information-knowledge-wisdom (DIKW) hierarchy to categorize papers and findings, and discuss the papers from the view point of "data value". Our review identifies various contexts where small data can be used to create value, such as data pre-processing, classification tasks, anomaly detection, forecasting and decision support. We also highlight industries that may be particularly promising areas for practitioners and researchers focused on small data. In addition, we provide an overview of methods and tools for small data analysis, including statistical techniques, visualization, and machine learning algorithms. Finally, based on our results, we suggest, that further research should focus on small data analysis. [ABSTRACT FROM AUTHOR]

The need for and interest in very-short-term load forecasting (VSTLF) is increasing and important for goals such as energy pricing markets. There is greater challenge in predicting load consumption for residential-load-type data, which is highly variable in nature and does not form visible patterns present in aggregated nodal-type load data. Previous works have used methods such as LSTM and CNN for VSTLF; however, the use of DNN has yet to be investigated. Furthermore, DNNs have been effectively used in STLF but have not been applied to very-short-term time frames. In this work, a deep network architecture is proposed and applied to very-short-term forecasting of residential load patterns that exhibit high variability and abrupt changes. The method extends previous work by including delayed load demand as an input, as well as working for 1 min data resolution. The deep model is trained on the load demand data of selected days—one, two, and a week—prior to the targeted day. Test results on real-world residential load patterns encompassing a set of 32 days (a sample from different seasons and special days) exhibit the efficiency of the deep network in providing high-accuracy residential forecasts, as measured with three different error metrics, namely MSE, RMSE, and MAPE. On average, MSE and RMSE are lower than 0.51 kW and 0.69 kW, and MAPE lower than 0.51%. [ABSTRACT FROM AUTHOR]

The application of the traditional k-nearest neighbours in regression analysis suffers from several difficulties when only a limited number of samples are available. In this paper, two decision models based on density are proposed. In order to reduce testing time, a k-nearest neighbours table (kNN-Table) is maintained to keep the neighbours of each object x along with their weighted Manhattan distance to x and a binary vector representing the increase or the decrease in each dimension compared to x's values. In the first decision model, if the unseen sample having a distance to one of its neighbours x less than the farthest neighbour of x's neighbour then its label is estimated using linear interpolation otherwise linear extrapolation is used. In the second decision model, for each neighbour x of the unseen sample, the distance of the unseen sample to x and the binary vector are computed. Also, the set S of nearest neighbours of x are identified from the kNN-Table. For each sample in S, a normalized distance to the unseen sample is computed using the information stored in the kNN-Table and it is used to compute the weight of each neighbor of the neighbors of the unseen object. In the two models, a weighted average of the computed label for each neighbour is assigned to the unseen object. The diversity between the two proposed decision models and the traditional kNN regressor motivates us to develop an ensemble of the two proposed models along with traditional kNN regressor. The ensemble is evaluated and the results showed that the ensemble achieves significant increase in the performance compared to its base regressors and several related algorithms. [ABSTRACT FROM AUTHOR]