Search

A bike-sharing system is a shared mobility mechanism that provides an alternative transportation mode for short trips with almost no added travel speed loss. However, this model’s low usage ratio and high depreciation rate pose a risk to the sustainable development of the bike-sharing industry. Our study proposes a new integrated station location and rebalancing vehicle service design model. This model aims to maximize daily revenue under a given total investment for station locations and bike acquisition. To address demand ambiguity due to possible bias and loss of data, we present a two-stage robust optimization model with a demand-related uncertainty set. The first stage of our model determines the station locations, initial bike inventory, and service areas of rebalancing vehicles. In contrast to the literature, which either simplifies the rebalancing process as an inventory transshipment problem or formulates it as a complex dynamic bike rebalancing problem, we assign each rebalancing vehicle to a service area composed of several specified stations. An approximate maximum travel distance for each rebalancing vehicle is also designed and constrained to ensure that the rebalancing operation can be performed within each period. In the second stage, our model optimizes the daily fleet operation and maximizes the total revenue minus the rebalancing cost. To solve our model, we design a customized row generation approach. Our numerical studies demonstrate that our algorithm can efficiently obtain exact solutions in small instances. For a real-size problem, the nearly optimal solutions of our model also reveal a high-quality worst-case performance with a small loss in mean performance, particularly when the value of the budget ratio (that is, the average number of bikes per station) is at a medium level. Moreover, the distribution of service areas depends on the bike supply and demand level at each station. The optimal fleet rebalancing operation does not have to be confined to one geographical area. Furthermore, our robust model can achieve larger mean and worst-case revenues and a higher revenue stability than a stochastic model with a small data set.

Representation learning with small labeled data have emerged in many problems, since the success of deep neural networks often relies on the availability of a huge amount of labeled data that is expensive to collect. To address it, many efforts have been made on training sophisticated models with few labeled data in an unsupervised and semi-supervised fashion. In this paper, we will review the recent progresses on these two major categories of methods. A wide spectrum of models will be categorized in a big picture, where we will show how they interplay with each other to motivate explorations of new ideas. We will review the principles of learning the transformation equivariant, disentangled, self-supervised and semi-supervised representations, all of which underpin the foundation of recent progresses. Many implementations of unsupervised and semi-supervised generative models have been developed on the basis of these criteria, greatly expanding the territory of existing autoencoders, generative adversarial nets (GANs) and other deep networks by exploring the distribution of unlabeled data for more powerful representations. We will discuss emerging topics by revealing the intrinsic connections between unsupervised and semi-supervised learning, and propose in future directions to bridge the algorithmic and theoretical gap between transformation equivariance for unsupervised learning and supervised invariance for supervised learning, and unify unsupervised pretraining and supervised finetuning. We will also provide a broader outlook of future directions to unify transformation and instance equivariances for representation learning, connect unsupervised and semi-supervised augmentations, and explore the role of the self-supervised regularization for many learning problems., Comment: published in IEEE Transactions on Pattern Analysis and Machine Intelligence

Predicting materials properties from composition or structure is of great interest to the materials science community. Deep learning has recently garnered considerable interest in materials predictive tasks with low model errors when dealing with large materials data. However, deep learning models suffer in the small data regime that is common in materials science. Here we leverage the transfer learning concept and the graph network deep learning framework and develop the AtomSets machine learning framework for consistent high model accuracy at both small and large materials data. The AtomSets models can work with both compositional and structural materials data. By combining with transfer learned features from graph networks, they can achieve state-of-the-art accuracy from using small compositional data (130,000). The AtomSets models show much lower errors than the state-of-the-art graph network models at small data limits and the classical machine learning models at large data limits. They also transfer better in the simulated materials discovery process where the targeted materials have property values out of the training data limits. The models require minimal domain knowledge inputs and are free from feature engineering. The presented AtomSets model framework opens new routes for machine learning-assisted materials design and discovery.

We present an efficient coreset construction algorithm for large-scale Support Vector Machine (SVM) training in Big Data and streaming applications. A coreset is a small, representative subset of the original data points such that a model trained on the coreset is provably competitive with that trained on the original data set. Since the size of the coreset is generally much smaller than the original set, our preprocess-then-train scheme has potential to lead to significant speedups when training SVM models. We prove lower and upper bounds on the size of the coreset required to obtain small data summaries for the SVM problem. As a corollary, we show that our algorithm can be used to extend the applicability of any off-the-shelf SVM solver to streaming, distributed, and dynamic data settings. We evaluate the performance of our algorithm on real-world and synthetic data sets. Our experimental results reaffirm the favorable theoretical properties of our algorithm and demonstrate its practical effectiveness in accelerating SVM training.

In this paper several statistical learning algorithms are used to predict the maximal length of fatigue cracks based on a sample composed of 31 observations. The small-data regime is still a problem for many professionals, especially in the areas where failures occur rarely. The analyzed object is a high-pressure Nozzle of a heavy-duty gas turbine. Operating parameters of the engines are used for the regression analysis. The following algorithms are used in this work: multiple linear and polynomial regression, random forest, kernel-based methods, AdaBoost and extreme gradient boosting and artificial neural networks. A substantial part of the paper provides advice on the effective selection of features. The paper explains how to process the dataset in order to reduce uncertainty; thus, simplifying the analysis of the results. The proposed loss and cost functions are custom and promote solutions accurately predicting the longest cracks. The obtained results confirm that some of the algorithms can accurately predict maximal lengths of the fatigue cracks, even if the sample is small.

As a high-efficiency and high-quality welding process, hybrid laser-MAG welding (HLMW) has significant potential of application in welding thick plates. In the present study, based on thermal elastic–plastic theory, a three-dimensional finite element model is developed to predict the weld shape characteristics, residual stress and distortion in HLMW for a butt joint of 12-mm-thick high strength steel plate. Metal active gas arc welding (MAG) heat input and laser energy are modeled as one double-ellipsoid body heat source and one cone body heat source with enhanced peak density along the central axis, respectively. The comparison between calculated molten pool shapes and those obtained by the experiment shows a good agreement. Then weld shape characteristic, residual stresses and distortions are calculated in four different welding process parameters. The results show that the increase of laser power and current can effectively increase the weld penetration width when the welding speed is fixed. At the top surfaces of weldment, the peak stress of high laser beam power is more significant than that of low laser beam power. A high compressive transverse stress of low laser beam power can be found at the welding zone and the surrounding heat affected zone. However, at the bottom surfaces of weldment, the peak stress of low laser beam power is larger than that of high laser beam power. The peak stress of low laser beam power is much larger than that of high laser beam power. A high compressive transverse stress of low laser beam power can be found at the welding zone. The vertical deformation in low laser beam power and low arc current welding has the lowest value.

Image transfer based on deep learning methods can achieve good results in face illumination processing. However, data constraints and generalization ability restrict the further development of these methods in this field. In this paper, we propose small data assisting face image illumination normalization. For data constraints, we train the network model on a small number of image pairs. In terms of generalization ability, the proposed normalization network parameters are different for processing different face images. Small data learning can provide prior knowledge, and the reconstruction process can guide detail generation. Therefore, the small data learning network and the reconstruction network are complementary to each other in image generating mode when only a small quantity of data is available. We use this network mode to normalize the illumination and reconstruct the super-resolution of face images. After illumination normalization, super-resolution reconstruction can obtain more precise face information and further improve the face recognition rate. Experiments show that the proposed method has good normalization performance when only 500 face image pairs are used for training.

With the development of smart grid information physical systems, some of the data processing functions gradually approach the edge layer of end-users. To better realize the energy theft detection function at the edge, we proposed an energy theft detection method based on the power consumption information acquisition system of power enterprises. The method involves the following steps. In the centralized data center, K-means is used to decompose a large amount of data into small data and then input and train neural network parameters to realize feature extraction. We design a neural network named DWMCNN, which can extract features from the day, week, and month and can extract more accurate features. In the edge data center, the random forest (RF) algorithm is used to classify the extracted features. The experimental results show that the clustering method accords with the idea of edge computing-distributed processing and improves the operation speed and that the feature extractor has good convergence performance. In addition, compared with the methods based on various classifiers, this method has higher accuracy and lower computational complexity, which is suitable for the deployment of edge data centers.

This paper provides a re-signification of industrial risk as a slow-burning issue (Mah [2017] "Environmental justice in the age of big data: challenging toxic blind spots of voice, speed, and expertise." Environmental Sociology 3 (2): 122-133.), invisibly and violently diffusing across time and space and affecting relational entanglements between human and non-human components of risk landscapes. As an alternative to a planning approach based on quantitative and objective data, the authors propose to build strategic planning of riskscapes upon what they call small data, that is, the ensemble of qualitative and embodied data that can be gathered through street science (Corburn [2005]. Street Science: Community Knowledge and Environmental Health Justice. Cambridge: MIT Press.) and toxic autobiographies (Armiero et al. [2019]. "Toxic Bios: Toxic Autobiographies - A Public Environmental Humanities Project." Environmental Justice, 1-5. ). In order to discuss the potential role of both small data and toxic autobiographies in the planning field, the authors present the results of an ongoing empirical case study in Gela, a Sicilian town converted into one of the main Italian petrochemical poles in the 1960s by a multinational oil company. The authors analyse Gela's risk landscapes through the perceptions of citizens and their initiatives to tackle environmental injustices. Finally, the authors argue that small data can provide a better understanding of the landscape of risk through four lenses that allow seeing the slow and diffused change brought by industrial risk: memories of injustice, memories of smell, trans-corporeal stories, and relational stories.

The paper considers the problem of handling short sets of medical data. Effectively solving this problem will provide the ability to solve numerous classification and regression tasks in case of limited data in health decision support systems. Many similar tasks arise in various fields of medicine. The authors improved the regression method of data analysis based on artificial neural networks by introducing additional elements into the formula for calculating the output signal of the existing RBF-based input-doubling method. This improvement provides averaging of the result, which is typical for ensemble methods, and allows compensating for the errors of different signs of the predicted values. These two advantages make it possible to significantly increase the accuracy of the methods of this class. It should be noted that the duration of the training algorithm of the advanced method remains the same as for existing method. Experimental modeling was performed using a real short medical data. The regression task in rheumatology was solved based on only 77 observations. The optimal parameters of the method, which provide the highest prediction accuracy based on MAE and RMSE, were selected experimentally. A comparison of its efficiency with other methods of this class has been performed. The highest accuracy of the proposed RBF-based additive input-doubling method among the considered ones is established. The method can be modified by using other nonlinear artificial intelligence tools to implement its training and application algorithms and such methods can be applied in various fields of medicine.

The issue of sentiment classification in short-term and small-scale data scenarios is considered in this paper. It is a hot topic because the text sentiment classification task in the public opinion analysis scene has two characteristics: short time and small data scale. Existing work focused on improving the accuracy at the cost of data and training time, without considering scenarios where time and data are lacked. The most commonly used method to solve the problem of small data scale is to use multi-modal information such as pictures, sounds and videos, which will lead to unbearable training time. The shorter training time determines that the classification model is generally selected as a deep neural network with fewer layers, such as TextCNN, TextRNN, and so on. However, such models are limited by the structure and have a low classification accuracy. In order to solve both short-term and small-scale data problems, a common information user attribute on social media is added to the model as multimodal information, which includes twelve attributes such as user age, location, and posting time. This paper proposed a sentiment classification algorithm based on multi-modal social media text information. The algorithm makes use of parallel convolutional neural networks (CNN) and recurrent neural network (RNN) to process text information and user attributes respectively, and combines the feature vectors of the two models for classification, which is called User attributes Convolutional and Recurrent Neural Network (UCRNN). The addition of user attributes can improve accuracy, and the CNN network used to extract user attributes features has fewer parameters, which proves that the algorithm can achieve high accuracy under short-term and small-scale data. Experiments verify that the training time of this model is slightly less than TextRNN. The classification accuracy can reach 90.2%, which is the state-of-the-art in the field of short-term and small-scale data sentiment classification.

In recent years, machine-learning-based approaches have become of considerable interest to the efficient processing of short or limited data samples. Its so-called small data approach. This is due to the significant growth of new intellectual analysis tasks in various industries, which are characterized by limited historical data. These include Materials Science, Economics, Medicine, and so on. An effective processing of short datasets is especially acute in medicine. Insufficient number of vectors, significant gaps in the data collected during the supervision of patient’s treatment or rehabilitation, reduces the effectiveness or prevents effective intellectual analysis based on them. This paper presents a new approach to processing short medical data samples. The basis of the developed method is SVR with RBF kernel. The algorithmic implementation of the method in both operation modes is described. Experimental modeling on a real short data set (Trabecular bone data) is conducted. It contained only 35 observations. A comparison of the method with a number of existing machine learning methods is conducted. It is experimental established the highest accuracy of the method among those considered. The developed method has potential opportunities for wide application in various fields of medicine.

A small data approach is an important area of research in the medical field. Many tasks in this area require the development of such direction due to the high cost of obtaining dataset or the inability to obtain sufficient observations for the analysis. Intelligent analysis of small datasets, which are very common in this area, will improve the quality of diagnosis, prevention or treatment. In this paper, the authors developed a new method of handling small datasets. The method is based on the use of a GRNN. The main steps of algorithms for preparation and application of the developed input doubling method are described. The modelling of the method is carried out; the optimal parameters for its functioning are selected. The efficiency of the method is established using a short set of clinical data by comparing its accuracy with a number of existing methods. It is shown that the developed input doubling method based on GRNN demonstrates the highest accuracy based on RMSE and MAE among all the methods considered. In addition, it provides unambiguous solutions, which avoids the use of such an approach as the "multiple runs" to choose the best solution. The main disadvantages of the method are analyzed, among which its considerable time delays of functioning in comparison with the basic GRNN that should be noted.

Optimization applications often depend on a huge number of uncertain parameters. In many contexts, however, the amount of relevant data per parameter is small, and hence, we may only have imprecise estimates. We term this setting—in which the number of uncertainties is large but all estimates have low precision—the small-data, large-scale regime. We formalize a model for this new regime, focusing on optimization problems with uncertain linear objectives. We show that common data-driven methods, such as sample average approximation, data-driven robust optimization, and certain regularized policies, may perform poorly in this new setting. We then propose a novel framework for selecting a data-driven policy from a given policy class. As with the aforementioned data-driven methods, our new policy enjoys provably good performance in the large-sample regime. Unlike these methods, we show that in the small-data, large-scale regime, our data-driven policy performs comparably to an oracle best-in-class policy under some mild conditions. We strengthen this result for linear optimization problems and two natural policy classes, the first inspired by the empirical Bayes literature and the second by regularization techniques. For both classes, the suboptimality gap between our proposed policy and the oracle policy decays exponentially fast in the number of uncertain parameters even for a fixed amount of data. Thus, these policies retain the strong large-sample performance of traditional methods and additionally enjoy provably strong performance in the small-data, large-scale regime. Numerical experiments confirm the significant benefits of our methods. This paper was accepted by Yinyu Ye, optimization.

Fire detection technologies remain a critical component of building automation. With the recent significant advances in computer vision, visual fire detection methods have been developed and integrated into building surveillance systems. Overfitting and accuracy challenges remain in fire detection when training datasets are limited. In this work, we tackle these challenges by developing a deep convolutional generative adversarial network (DCGAN) for highly accurate visual fire detection when training images are limited. Our model addresses three types of errors in visual fire detection with small training datasets: model overfitting, fire probability overestimation, and fire probability underestimation. The DCGAN includes a generator of fake fire images for self-supervised learning (SSL) and a discriminator for image classification. We designed computational experiments with high-quality datasets to test and validate our model against other supervised learning approaches. We also benchmarked the performance of the DCGAN against a best-in-class deep visual fire detection model. The results show that our model significantly outperforms other fire detection models on all performance metrics when trained with the same small dataset. The results demonstrate that the DCGAN effectively mitigates the three types of error when the training dataset is limited.

Prediction methods of glioblastoma tumours growth constitute a hard task due to the lack of medical data, which is mostly related to the patients' privacy, the cost of collecting a large medical data set, and the availability of related notations by experts.In this study, the authors propose a synthetic medical image generator (SMIG) with the purpose of generating synthetic data based on the generative adversarial network in order to provide anonymised data. In addition, to predict the glioblastoma multiform tumour growth the authors developed a tumour growth predictor based on end to end convolution neural network architecture that allows training on a public data set from the cancer imaging archive (TCIA), combined with the generated synthetic data. The authors also highlighted the impact of implicating a synthetic data generated using SMIG as a data augmentation tool. Despite small data size provided by TCIA data set, the obtained results demonstrate valuable tumour growth prediction accuracy.

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 th...

The outbreak of COVID-19 has caused more than 200,000 deaths so far in the USA alone, which instigates the necessity of initial screening to control the spread of the onset of COVID-19. However, screening for the disease becomes laborious with the available testing kits as the number of patients increases rapidly. Therefore, to reduce the dependency on the limited test kits, many studies suggested a computed tomography (CT) scan or chest radiograph (X-ray) based screening system as an alternative approach. Thereby, to reinforce these approaches, models using both CT scan and chest X-ray images need to develop to conduct a large number of tests simultaneously to detect patients with COVID-19 symptoms. In this work, patients with COVID-19 symptoms have been detected using eight distinct deep learning techniques, which are VGG16, InceptionResNetV2, ResNet50, DenseNet201, VGG19, MobilenetV2, NasNetMobile, and ResNet15V2, using two datasets: one dataset includes 400 CT scan and another 400 chest X-ray images. Results show that NasNetMobile outperformed all other models by achieving an accuracy of 82.94% in CT scan and 93.94% in chest X-ray datasets. Besides, Local Interpretable Model-agnostic Explanations (LIME) is used. Results demonstrate that the proposed models can identify the infectious regions and top features, ultimately, it provides a potential opportunity to distinguish between COVID-19 patients with others.

In order to address the clustering problem of intelligent medical data, the data sets were not preprocessed using the traditional method, leading to a large amount of calculation, low efficiency, and large data cluster center offset distance. We proposed a balanced clustering algorithm for intelligent medical heterogeneous big data set using deep learning. Firstly, a deep neural network model based on incremental updating was constructed, and adaptive training and adjustment were made according to data scale, and the multi-layer feature learning of heterogeneous big data sets of intelligent medical care. Secondly, under-sampling preprocessing was carried out on the data set so that the data of the heterogeneous big data set was in a balanced state, and on this basis, clustering calculation of the heterogeneous big data was conducted. Then, the clustering center was set according to the kernel density estimation results, and the data cluster center was updated iteratively until convergence by combining the data features obtained from deep learning and euclidean distance calculation, so as to complete the balanced clustering of the heterogeneous big data set of intelligent medical treatment. The results show that the proposed algorithm has the advantages of small data cluster center offset distance, short clustering time, low energy consumption, high Macro-F1 value and NMI value, and the accuracy of clustering can be as high as 95%, the calculational cost is low, which has certain advantages. 2020 Elsevier Ltd. All rights reserved.

Automatic extracting of knowledge from massive data samples, i.e., big data analytics (BDA), has emerged as a vital task in almost all scientific research fields. The BDA problems are rather difficult to solve due to their large-scale, high-dimensional, and dynamic properties, while the problems with small data are usually hard to handle due to insufficient data samples and incomplete information. Such difficulties lead to the search-based data analytics problem, where a data analysis task is modeled as a complex, dynamic, and computationally expensive optimization problem and then solved by using an iterative algorithm. In this paper, we intend to present an extensive and in-depth discussion on the utilizing of evolutionary computation (EC) based optimization methods [including evolutionary algorithms (EAs) and swarm intelligence (SI)] for solving search-based data analysis problems. Then, as an example for illustration, we provide a comprehensive review of the applications of state-of-the-art EC methods for different types of data mining problems in bioinformatics. Here, the detailed analysis and discussion are conducted on three types of data samples, which include sequences data, network data, and image data. Finally, we survey the challenges faced by EC methods and the trend for future directions. Based on the applications of EC methods for search-based data analysis problems involving inexact and uncertain information, the insights of data analytics are able to understand better, and more efficient algorithms could be designed to solve real-world complex BDA problems.

Under the plane wave setting, the existence of small Cauchy data global solution (or local solution) of Landau–Lifshitz–Gilbert equation is proved. Some variable separation type solutions (include some small data global solution) and self-similar type solutions are constructed for the Harmonic map heat flow on S 2 . As far as we know, there is not any literature that presents the exact blowup solution of this equation. Some explicit solutions which include some finite time gradient-blowup solutions are provided. These blowup examples indicate a finite time blowup will happen in any spacial dimension.

Switched systems, i.e., systems changing the parameter values (even structural forms) abruptly and randomly at arbitrary instants, have been extensively utilized in many fields of modern industries. Rapid identification of switched systems, i.e., capturing all the changing instants and reconstructing the mathematical models rapidly, is of great significance for behavior prediction, performance evaluation and possible control, but is restricted by small data amount available. Here, the rapid identification problem is successfully solved by a data-driven method in variational framework. The data-driven method only requires a small amount of data due to the compact form of the variational description, and is robust to data noise due to the holistic viewpoint. Two numerical examples, i.e., Duffing oscillator and van der Pol system (as two representative systems in nonlinear dynamics), are adopted to illustrate its application, efficiency and robustness to noise.

This work aims to reduce queries on big data to computations on small data, and hence make querying big data possible under bounded resources. A query Q is boundedly evaluable when posed on any big dataset $${\cal D}$$ D , there exists a fraction $${{\cal D}_Q}$$ D Q of $${\cal D}$$ D such that $$Q({\cal D}) = Q({{\cal D}_Q})$$ Q ( D ) = Q ( D Q ) , and the cost of identifying $${{\cal D}_Q}$$ D Q is independent of the size of $${\cal D}$$ D . It has been shown that with an auxiliary structure known as access schema, many queries in relational algebra (RA) are boundedly evaluable under the set semantics of RA. This paper extends the theory of bounded evaluation to RAaggr, i.e., RA extended with aggregation, under the bag semantics. (1) We extend access schema to bag access schema, to help us identify $${{\cal D}_Q}$$ D Q for RAaggr queries Q. (2) While it is undecidable to determine whether an RAaggr query is boundedly evaluable under a bag access schema, we identify special cases that are decidable and practical. (3) In addition, we develop an effective syntax for bounded RAaggr queries, i.e., a core subclass of boundedly evaluable RAaggr queries without sacrificing their expressive power. (4) Based on the effective syntax, we provide efficient algorithms to check the bounded evaluability of RAaggr queries and to generate query plans for bounded RAaggr queries. (5) As proof of concept, we extend PostgreSQL to support bounded evaluation. We experimentally verify that the extended system improves performance by orders of magnitude.

Radar cross-section (RCS) of an object is a complex function of various geometric variables, frequency and angles of incidence. In this work, an artificial intelligence solution is provided to predict the non-deterministic characteristics of RCS using the supervised machine learning algorithm that involves Gaussian process (GP) regression. A parametrised aircraft model is used to generate training data where five variables are selected as predictors while the response is chosen to be monostatic RCS in the azimuth plane. To provide a comparison of GP modelling-based predictions, shooting and bouncing rays-based multi-frequency RCS simulations are used and the results show good agreement. To further validate the GP-based modelling approach, the data of a design point is compared with the measured RCS of 1:8 scaled-down aircraft model, which confirms the accuracy of the proposed methodology. Good prediction capabilities of GP regression for RCS evaluation of complex geometries and requirement of small data set make it an excellent tool for exploring the large design space as well as integration into multi-disciplinary design optimisation environments.

For predicting catchment runoff with data-driven methods, a long historical database of measurements is required. The current study focuses on the assessment of a deep learning model named long short-term memory (LSTM) for rainfall–runoff relationship with different training data size. The developed model has been evaluated on twenty catchments with diverse hydrological conditions obtained from the freely available CAMELS dataset. In order to prove the efficiency of the proposed model for runoff prediction, we test its performances against the traditional feed-forward neural network model. The studied models have been trained with the same input parameters and different size of training data to show the effect of data length on the prediction performances. To this end, the length of training data was varied from 3 to 15 years, while the model was tested on 10 years of data. The results show that the deep LSTM outperforms the traditional model in terms of statistical indicators over different size of training sets. The proposed deep LSTM model can predict runoff with acceptable performances using 9 years of data length in the training procedure, a result that improves when using 12 years. In addition, it has been proven that the deep LSTM model may be efficient even when using small data size (3 years) compared to its benchmarked model which require 9 years for similar results. Thus, the LSTM network is a powerful deep learning model able to learn the behavior of rainfall–runoff relationship with a minimum data length.

—Nowadays, many devices require a wireless Internet connection, while the available frequency spectrum is limited. The nonorthogonal multiple access (NOMA) is a multiplexing technology that can significantly improve spectral efficiency and thus solve this problem. In this paper, the use of NOMA in Wi-Fi networks with two traffic types, Web traffic and Internet of Things (IoT) traffic, is investigated. Web traffic is one of the main Internet traffic types. It consists of relatively small data flows that need to be quickly delivered to users. IoT traffic consists of very short data flows compared to Web traffic, but there can be a lot of IoT flows due to a large number of IoT devices. In this paper, we propose an algorithm for resource allocation in Wi-Fi networks with the NOMA. This algorithm takes into account the traffic of both types and its performance is investigated by simulation.

Purpose Considering the problem that the high recognition rate of deep learning requires the support of mass data, this study aims to propose an insulating fault identification method based on small data set convolutional neural network (CNN). Design/methodology/approach Because of the chaotic characteristics of partial discharge (PD) signals, the equivalent transformation of the PD signal of unit power frequency period is carried out by phase space reconstruction to derive the chaotic features. At the same time, geometric, fractal, entropy and time domain features are extracted to increase the volume of feature data. Finally, the combined features are constructed and imported into CNN to complete PD recognition. Findings The results of the case study show that the proposed method can realize the PD recognition of small data set and make up for the shortcomings of the methods based on CNN. Also, the 1-CNN built in this paper has better recognition performance for four typical insulation faults of cable accessories. The recognition performance is improved by 4.37% and 1.25%, respectively, compared with similar methods based on support vector machine and BPNN. Originality/value In this paper, a method of insulation fault recognition based on CNN with small data set is proposed, which can solve the difficulty to realize insulation fault recognition of cable accessories and deep data mining because of insufficient measure data.

The present study suggests an integrated approach for determining the optimal release policy for the Mula reservoir situated at the Godavari basin, India. The proposed integrated algorithm named DP-PSO is a hybridization of Dynamic Programming (DP) and Particle Swarm Optimization (PSO). The reservoir operation problem is demonstrated in the form of a nonlinear optimization model subject to various constraints. Two case studies are considered. In the first case the efficiency of the proposed algorithm is tested on a small data set of 1 year and in the second case, data set taken is for 10 years. The results obtained are compared in terms of objective function value as well CPU time for performance evaluation of the integrated methods.

Understanding waste generation flow is vital to any evidence-based effort by policy-makers and practitioners to successfully manage construction project waste. Previous research has found that accumulative waste generation in construction projects follows an S-curve, but improving our understanding of waste generation requires its investigation at a higher level of granularity. Such efforts, however, are often constrained by lack of quality “bigger” data, i.e. data that is bigger than normal small data. This research aims to provide an anatomy of waste generation flow in building projects by making use of a large set of data on waste generation in 19 demolition, 59 foundation, and 54 new building projects undertaken in Hong Kong between 2011 and 2019. We know that waste is generated in far from a steady stream as it is always impacted by contingent factors. However, we do find that peaks of waste generation in foundation projects appear when project duration is at 50–85%, and in new building projects at 40–70% of total project time. Our research provides useful information for waste managers in developing their waste management plans, arranging waste hauling logistics, and benchmarking waste management performance.

Recently, artificial neural networks (ANNs) have been applied to various robot-related research areas due to their powerful spatial feature abstraction and temporal information prediction abilities. Decision-making has also played a fundamental role in the research area of robotics. How to improve ANNs with the characteristics of decision-making is a challenging research issue. ANNs are connectionist models, which means they are naturally weak in long-term planning, logical reasoning, and multistep decision-making. Considering that a small refinement of the inner network structures of ANNs will usually lead to exponentially growing data costs, an additional planning module seems necessary for the further improvement of ANNs, especially for small data learning. In this paper, we propose a state operator and result (SOAR) improved ANN (SANN) model, which takes advantage of both the long-term cognitive planning ability of SOAR and the powerful feature detection ability of ANNs. It mimics the cognitive mechanism of the human brain to improve the traditional ANN with an additional logical planning module. In addition, a data fusion module is constructed to combine the probability vector obtained by SOAR planning and the original data feature array. A data fusion module is constructed to convert the information from the logical sequences in SOAR to the probabilistic vector in ANNs. The proposed architecture is validated in two types of robot multistep decision-making experiments for a grasping task: a multiblock simulated experiment and a multicup experiment in a real scenario. The experimental results show the efficiency and high accuracy of our proposed architecture. The integration of SOAR and ANN is a good compromise between logical planning with small data and probabilistic classification with big data. It also has strong potential for more complicated tasks that require robust classification, long-term planning, and fast learning. Some potential applications include recognition of grasping order in multiobject environment and cooperative grasping of multiagents.

A data-oriented approach has been recently proposed that simultaneously supports ultra-high reliability and low latency in wireless transmission systems. In this letter, we analyze the performance of small data transmissions over fading channels in terms of a data-oriented performance metric. Through analytical results and associated numerical examples, we demonstrate the effectiveness of rate/power adaptive transmission and diversity combining in achieving ultra-reliable low-latency transmissions.

Recent progress in deep learning is essentially based on a "big data for small tasks" paradigm, under which massive amounts of data are used to train a classifier for a single narrow task. In this paper, we call for a shift that flips this paradigm upside down. Specifically, we propose a "small data for big tasks" paradigm, wherein a single artificial intelligence (AI) system is challenged to develop "common sense", enabling it to solve a wide range of tasks with little training data. We illustrate the potential power of this new paradigm by reviewing models of common sense that synthesize recent breakthroughs in both machine and human vision. We identify functionality, physics, intent, causality, and utility (FPICU) as the five core domains of cognitive AI with humanlike common sense. When taken as a unified concept, FPICU is concerned with the questions of "why" and "how", beyond the dominant "what" and "where" framework for understanding vision. They are invisible in terms of pixels but nevertheless drive the creation, maintenance, and development of visual scenes. We therefore coin them the "dark matter" of vision. Just as our universe cannot be understood by merely studying observable matter, we argue that vision cannot be understood without studying FPICU. We demonstrate the power of this perspective to develop cognitive AI systems with humanlike common sense by showing how to observe and apply FPICU with little training data to solve a wide range of challenging tasks, including tool use, planning, utility inference, and social learning. In summary, we argue that the next generation of AI must embrace "dark" humanlike common sense for solving novel tasks., For high quality figures, please refer to http://wellyzhang.github.io/attach/dark.pdf

Thermal properties of sand are of importance in numerous engineering and scientific applications ranging from energy storage and transportation infrastructures to underground construction. All these applications require knowledge of the effective thermal parameters for proper operation. The traditional approaches for determination of the effective thermal property, such as the thermal conductivity are based on very costly, tedious and time-consuming experiments. The recent developments in computer science have allowed the use of soft and hard computational methods to compute the effective thermal conductivity (ETC). Here, two computation methods are presented based on soft and hard computing approaches, namely, the deep neural network (DNN) and the thermal lattice element method (TLEM), respectively, to compute the ETC of sands with varying porosity and moisture content values. The developed models are verified and validated with a small data set reported in the literature. The computation results are compared with the experiments, and the numerical results are found to be within reasonable error bounds. The deep learning method offers fast and robust implementation and computation, even with a small data set due to its superior backpropagation algorithm. However, the TLEM based on micro and meso physical laws outperforms it at accuracy.

In order to make a good compromise of cost and safety with small data in the early structural design stage, a practical decoupled credibility-based design optimization method is developed in the presence of fuzzy uncertainty. In the proposed approach, failure credibility is constructed as optimization constraints estimated by fuzzy advanced first-order second-moment method. By approximating the fuzzy credibility constraint by the adaptive Kriging surrogate model, a fuzzy credibility-based design is decoupled to a common deterministic optimization so that various existing optimization algorithms can be easily applied. Compared to the traditional double-loop approach, the newly proposed method is more efficient and strongly practical for complicated engineering problems. Design results of three structural engineering examples also show advantages in accuracy and computation speed of the proposed method over the traditional double-loop approach.

In this paper, we propose a novel architecture that combines the convolutional neural network (CNN) with a fuzzy neural network (FNN). We utilize the fuzzy neural network with semi-connected layers to sum up feature information. During the training process, to map membership values, the CNN generates feature maps as outputs and feeds into fuzzifier layers, alternatively called fuzzy maps. The proposed method increases classification accuracy, because fuzzy neural networks can generate not only crisp values but also fuzzy values; this means that there is potentially more information contained in the fuzzy set. Our model is evaluated by cross-validation tests. While big data is necessary for training in general, we train our model with small data and test with big data to demonstrate its ability of object classification in cases where sufficient data are not available.

Purpose With the development of deep learning-based analytical techniques, increased research has focused on fatigue data analysis methods based on deep learning, which are gaining in popularity. However, the application of deep neural networks in the material science domain is mainly inhibited by data availability. In this paper, to overcome the difficulty of multifactor fatigue life prediction with small data sets, Design/methodology/approach A multiple neural network ensemble (MNNE) is used, and an MNNE with a general and flexible explicit function is developed to accurately quantify the complicated relationships hidden in multivariable data sets. Moreover, a variational autoencoder-based data generator is trained with small sample sets to expand the size of the training data set. A comparative study involving the proposed method and traditional models is performed. In addition, a filtering rule based on the R2 score is proposed and applied in the training process of the MNNE, and this approach has a beneficial effect on the prediction accuracy and generalization ability. Findings A comparative study involving the proposed method and traditional models is performed. The comparative experiment confirms that the use of hybrid data can improve the accuracy and generalization ability of the deep neural network and that the MNNE outperforms support vector machines, multilayer perceptron and deep neural network models based on the goodness of fit and robustness in the small sample case. Practical implications The experimental results imply that the proposed algorithm is a sophisticated and promising multivariate method for predicting the contact fatigue life of a coating when data availability is limited. Originality/value A data generated model based on variational autoencoder was used to make up lack of data. An MNNE method was proposed to apply in the small data case of fatigue life prediction.

Synthetic aperture radar (SAR) tomography (TomoSAR) has been well established for the 3-D reconstruction of urban buildings. Many methods have been proposed in the literature for TomoSAR inversion. These methods usually require fairly large data stacks (more than 20 images) for reliable reconstruction. Hence, they cannot be applied to the 3-D reconstruction of urban buildings using Gaofen-3 (GF-3) data directly, because there are few images available on average in each city. This article proposes a novel workflow for the 3-D reconstruction of high-rise buildings using small data stacks. In this workflow, we combine the methods of contour line extraction (CLE) and reference-elevation multilooking relaxation (RM-RELAX) for the 3-D reconstruction of high-rise buildings. The CLE method extracts contour lines of buildings relying on a data-driven approach and does not need to rely on any external data, which can provide prior knowledge for subsequent processing. The RM-RELAX method is an extension of RELAX, which can obtain more precise 3-D inversion with multilooking and reject outliers with the reference elevation (which can be obtained by the prior knowledge of contour lines) of each pixel. The applicability of this workflow is demonstrated by using simulated data and real data with six SAR images acquired by the GF-3 satellite over an area in Beijing, China.

Molecular biology studies on cancer, using gene expression datasets, have revealed that the datasets have a very small number of samples. Obtaining medical data is difficult and expensive due to privacy constraints. Accuracy of classifiers depends greatly on the quality and quantity of input data. The problem of small sample size or small data size has been addressed by augmentation. Owing to the sensitivity of synthetic data samples for the cancer data classification for gene expression data, this paper is motivated to investigate data augmentation using GAN. GAN is based on the principle of two blocks (generator and discriminator) working in a collaborative yet adversarial way. This paper proposes modified generator GAN (MG-GAN) where the generator is fed with original data and multivariate noise to generate data with Gaussian distribution. As the generated data lie within latent space, we reach saddle point faster. GAN has been widely used in data augmentation for image datasets. As per our understanding, this is the first attempt of using GAN for augmentation on gene expression dataset. The performance merit of proposed MG-GAN was compared with KNN and Basic GAN. As compared to KNN and GAN, MG-GAN improves classification accuracy by 18.8% and 11.9%, respectively. The loss value of the error function for MG-GAN is drastically reduced, from 0.6978 to 0.0082, ensuring sensitivity of the generated data. Improved classification accuracy and reduction in the loss value make our improved MG-GAN method better suited for critical applications with sensitive data.

In this paper, a hierarchical frequency regulation system is presented for real-time operation of the electrical grids. It includes a decentralized transactive energy system for clearing transactions of producers/consumers in real-time. The proposed system is implemented in a distributed fashion with small data communication requirement. It is proved that the proposed transactive energy system converges to the optimal power flow (OPF). An iterative method for optimizing control parameters of the proposed hierarchical frequency regulation system is also given. The proposed method is developed for high penetration of renewable energy sources (RESs) as the sizes, locations, and uncertainties of RESs are considered. Finally, the performance of proposed transactive energy system is evaluated in the case study section.

Malicious HTTP traffic detection plays an important role in web application security. Most existing work applies machine learning and deep learning techniques to build the malicious HTTP traffic detection model. However, they still suffer from the problems of huge training data collection cost and low cross-dataset generalization ability. Aiming at these problems, this paper proposes DeepPTSD, a deep learning method for payload based malicious HTTP traffic detection. First, it treats the malicious HTTP traffic detection as a text classification problem and trains the initial detection model using TextCNN on a public dataset, and then adapts the initial detection model to the target dataset based on a transfer learning algorithm. Second, in the transfer learning procedure, it uses a semi-supervised learning algorithm to accomplish the model adaptation task. The semi-supervised learning algorithm enhances the target dataset based on a HTTP payload data augmentation mechanism to exploit both the labeled and unlabeled data. We evaluate DeepPTSD on two real HTTP traffic datasets. The results show that DeepPTSD has competitive performance under the small data condition.

Computer-aided systems for skin lesion diagnosis is a growing area of research. Recently, researchers have shown an increasing interest in developing computer-aided diagnosis systems. This paper aims to review, synthesize and evaluate the quality of evidence for the diagnostic accuracy of computer-aided systems. This study discusses the papers published in the last five years in ScienceDirect, IEEE, and SpringerLink databases. It includes 53 articles using traditional machine learning methods and 49 articles using deep learning methods. The studies are compared based on their contributions, the methods used and the achieved results. The work identified the main challenges of evaluating skin lesion segmentation and classification methods such as small datasets, ad hoc image selection and racial bias.

International audience; In this paper, we outline a diffractive practice of machine learning (ML) in the frame of material-centered interaction design. To this aim, we review related work in ML, HCI, design, new interfaces for musical expression, and computational art, and introduce two practice-based studies of music performance and robotic art based on interactive machine learning tools, with the hope of revealing the computational materiality of ML, and the potential of embodiment to craft prototypes of ML that reconfigure conceptual or technical approaches to ML. We derive five interference conditions for such art-based ML prototypes—situational whole, small data, shallow model, learnable algorithm, and somaesthetic behaviour—and describe their widening of design and engineering practices of ML prototyping. Finally, we sketch how a process of intra-active machine learning could complement that of interactive machine learning to take materiality as an entry point for ML design within HCI.

There is a growing need to perform a diverse set of real-time analytics (batch and stream analytics) on evolving graphs to deliver the values of big data to users. The key requirement from such applications is to have a data store to support their diverse data access efficiently, while concurrently ingesting fine-grained updates at a high velocity. Unfortunately, current graph systems, either graph databases or analytics engines, are not designed to achieve high performance for both operations; rather, they excel in one area that keeps a private data store in a specialized way to favor their operations only. To address this challenge, we have designed and developed G raph O ne , a graph data store that abstracts the graph data store away from the specialized systems to solve the fundamental research problems associated with the data store design. It combines two complementary graph storage formats (edge list and adjacency list) and uses dual versioning to decouple graph computations from updates. Importantly, it presents a new data abstraction, GraphView , to enable data access at two different granularities of data ingestions (called data visibility ) for concurrent execution of diverse classes of real-time graph analytics with only a small data duplication. Experimental results show that G raph O ne is able to deliver 11.40× and 5.36× average speedup in ingestion rate against LLAMA and Stinger, the two state-of-the-art dynamic graph systems, respectively. Further, they achieve an average speedup of 8.75× and 4.14× against LLAMA and 12.80× and 3.18× against Stinger for BFS and PageRank analytics (batch version), respectively. G raph O ne also gains over 2,000× speedup against Kickstarter, a state-of-the-art stream analytics engine in ingesting the streaming edges and performing streaming BFS when treating first half as a base snapshot and rest as streaming edge in a synthetic graph. G raph O ne also achieves an ingestion rate of two to three orders of magnitude higher than graph databases. Finally, we demonstrate that it is possible to run concurrent stream analytics from the same data store.

In this article, we comment on the drawbacks of the existing AI-based Bayesian network (BN) cyber-vulnerability analysis (C-VA) model proposed in Mukhopadhyay et al. (2013) to assess cyber-risk in IT firms, where this quantity is usually a joint distribution of multiple risk (random) variables (e.g., quality of antivirus, frequency of monitoring, etc.) coming from heterogeneous distribution families. As a major modeling drawback, Mukhopadhyay et al. (2013) assume that any pair of random variables in the BN are linearly correlated with each other. This simplistic assumption might not always hold true for general IT organizational environments. Thus, the use of the C-VA model in general will result in loose estimates of correlated IT risk and will subsequently affect cyber-insurance companies in framing profitable coverage policies for IT organizations. To this end, we propose methods to (1) find a closed-form expression for the maximal correlation arising between pairs of discrete random variables, whose value finds importance in getting robust estimates of copula-induced computations of organizational cyber-risk, and (2) arrive at a computationally effective mechanism to compute nonlinear correlations among pairs of discrete random variables in the correlation matrix of the CBBN model (Mukhopadhyay et al. 2013). We also prove that an empirical computation of MC using our method converges rapidly, that is, exponentially fast, to the true correlation value in the number of samples. Our proposed method contributes to a tighter estimate of IT cyber-risk under environments of low-risk data availability and will enable insurers to better assess organizational risks and subsequently underwrite profitable cyber-insurance policies.

Protein phosphorylation is the most extensive and important post-translational modification in eukaryotes, regulating the activity of almost all cells. Experimental methods used to identify phosphorylation sites, such as mass spectrometry, are costly and time-consuming. A number of algorithms have been developed to predict phosphorylation sites. However, they often select small data volume by random sampling. This cannot make full use of the characteristics of the entire data set to build a prediction model. According to the granularity calculation combined with the kernel fuzzy C-means clustering, this paper maps the massive raw data to a high-dimensional kernel space, and then divides the grains by clustering to obtain high-dimensional equilibrium grains. In particular, a specific granular support vector machine (KFCC–GSVM) prediction model is built in equilibrium grain data. This novel model improves the rationality and reliability of phosphorylation site data compression, so that the compressed data has the same distribution in the kernel space as the pre-compression data when applying the traditional SVM algorithm classification. Experimental results demonstrate that our method is better than the SVM-based non-kinase-specific phosphorylation site prediction method—Musite and the traditional GSVM method.

PurposeThe purpose of this article is to examine the ways in which self-tracking data have meaning and value in and after the life of the creator, including how such data could become part of the larger historical record, curated in an institutional archive. In doing so, the article expands upon existing shared interests among researchers working in the areas of self-tracking, human–computer interaction and archival science.Design/methodology/approachA total of 18 people who had self-tracked for six months or more were recruited for the study. Participants completed a survey which gathered demographic data and characteristicsvis-à-vistheir self-tracking behavior. In-person semi-structured interviews were then conducted to ascertain the beliefs of the participants regarding the long-term use and value of personal quantified-self data.FindingsThe findings reveal the value that people place on self-tracking data, their thoughts on proper modes for accessing their archive once it moves from the private to the public space, and how to provide fidelity within the system such that their experiences are represented while also enabling meaning making on the part of subsequent users of the archive.Originality/valueToday’s quantified-self data are generally embedded in systems that create a pipeline from the individual source to that of the corporate warehouse, bent on absorbing and extracting insight from a totality of big data. This article posits that new opportunities for knowing and for design can be revealed when a public interest rationale is appended to rich personalized collections of small data.

PurposeThe purpose of this paper is to propose a framework for data analytics where everything is grey in nature and the associated uncertainty is considered as an essential part in data collection, profiling, imputation, analysis and decision making.Design/methodology/approachA comparative study is conducted between the available uncertainty models and the feasibility of grey systems is highlighted. Furthermore, a general framework for the integration of grey systems and grey sets into data analytics is proposed.FindingsGrey systems and grey sets are useful not only for small data, but also big data as well. It is complementary to other models and can play a significant role in data analytics.Research limitations/implicationsThe proposed framework brings a radical change in data analytics. It may bring a fundamental change in our way to deal with uncertainties.Practical implicationsThe proposed model has the potential to avoid the mistake from a misleading data imputation.Social implicationsThe proposed model takes the philosophy of grey systems in recognising the limitation of our knowledge which has significant implications in our way to deal with our social life and relations.Originality/valueThis is the first time that the whole data analytics is considered from the point of view of grey systems.

We study a Klein-Gordon-Maxwell system, in a bounded spatial domain, under Neumann boundary conditions on the electric potential. We allow a nonconstant coupling coefficient. For sufficiently small data, we find infinitely many standing waves., 18 pages

Recently, auto-encoder-based generative models have been widely used successfully for image processing. However, there are few studies on the realization of continuous input–output mappings for regression problems. Lack of a sufficient amount of training data plagues regression problems, which is also a notable problem in machine learning, which affects its application in the field of materials science. Using variational auto-encoders (VAEs) as generative models for data augmentation, we address the issue of small data size for regression problems. VAEs are popular and powerful auto-encoder-based generative models. Generative auto-encoder models such as VAEs use multilayer neural networks to generate sample data. In this study, we demonstrate the effectiveness of multi-task learning (auto-encoding and regression tasks) relating to regression problems. We conducted experiments on seven benchmark datasets and on one ionic conductivity dataset as an application in materials science. The experimental results show that the multi-task learning for VAEs improved the generalization performance of multivariable linear regression model trained with augmented data.

Single point incremental forming (SPIF) has revolutionized sheet shaping for small-batch production, providing an economical and effective alternative to sheet stamping and pressing, which can be cumbersome and expensive. Efficient data-driven prediction of forming forces can significantly benefit process design, development and optimization in SPIF. However, the nature of localized plastic deformation makes SPIF a time-consuming process which means it is difficult to obtain rich experimental data (or samples) in the early period of forming process design. To build an efficient data-driven model for forming force prediction using the back propagation neural networks, this paper proposes a virtual data generation approach based on mega trend diffusion function and particle swarm optimization algorithm to improve the accuracy of SPIF force prediction given small experimental data problems. The proposed modeling methodology is verified using small amount of force data obtained from pyramidal shape forming. It is found that the accuracy of the established prediction model can be improved by adding the generated virtual data to actual experimental small datasets, which provides a good predictive capability in modeling the forming force of SPIF under different process conditions.