In data mining, large differences between multi-class distributions regarded as class imbalance issues have been known to hinder the classification performance. Unfortunately, existing sampling methods have shown their deficiencies such as causing the problems of over-generation and over-lapping by oversampling techniques, or the excessive loss of significant information by undersampling techniques. This paper presents three proposed sampling approaches for imbalanced learning: the first one is the entropy-based oversampling (EOS) approach; the second one is the entropy-based undersampling (EUS) approach; the third one is the entropy-based hybrid sampling (EHS) approach combined by both oversampling and undersampling approaches. These three approaches are based on a new class imbalance metric, termed entropy-based imbalance degree (EID), considering the differences of information contents between classes instead of traditional imbalance-ratio. Specifically, to balance a data set after evaluating the information influence degree of each instance, EOS generates new instances around difficult-to-learn instances and only remains the informative ones. EUS removes easy-to-learn instances. While EHS can do both simultaneously. Finally, we use all the generated and remaining instances to train several classifiers. Extensive experiments over synthetic and real-world data sets demonstrate the effectiveness of our approaches.

In term of data-driven techniques for reliability assessment of transmission gears, it is necessary to collect sufficient data of each class in practice. However, the fact that highest and lowest reliability events rarely occur would result in a class imbalance problem. To resolve such an issue, we propose conditional generative adversarial network-mean-covariance balancing labeling (CGAN-MBL) to improve reliability assessment of transmission gears with insufficient and imbalanced data. First, we establish a CGAN-based model to generate credible instances with a similar distribution of original data. In the model, it is proven that by introducing a network weight initialization scheme, the ability to escape from local optimum is enhanced and global search ability of CGAN is strengthened; minibatch discriminator is designed to decrease mode collapse by distinguishing a batch of original and generated data iteratively. Second, in order to overcome the drawback of CGAN with unlabeled generated data, MBL is introduced to tag the generated data through exploring the nearest distance between generated instances and original class centers. Finally, synthetic data combined generated data with original ones train classifiers. Simulations show the effectiveness of the proposed approach for imbalance learning in real-world transmission gear data.

Reliability classification of gear safety has long been a challenging issue in transmission industry because of complicated calculations and great classification errors of coupled parameters with insufficient data. This paper proposes a model based on generative adversarial network (GAN) as pretreatment to improve the accuracy of reliability classification. First, we present bounded-GAN to generate gear data within required boundaries without massive computations. In bounded-GAN, three bounded layers are designed to bound generated data in terms of different data characteristics; smooth targets are developed to enhance the ability of generating high-quality instances by the generator; Adam optimizer is used to train both generator and discriminator to avoid nonconvergence. Second, to overcome unlabeling defect of bounded-GAN, a mean-covariance labeling scheme is introduced to label the data according to the nearest classes of gear reliability within specific ranges. Finally, original and qualified data are combined to train classifiers. Simulations on gear data from industry show that our proposed model outperforms other methods on operational metrics.

Bridge health assessment has been a challenging problem due to great evaluation errors caused by heterogeneous and different dimensional bridge factors. Most approaches employ monitoring data of one bridge factor to simplify the evaluation problem resulting in poor assessment performance. To address this issue, this paper proposes an end-to-end framework to evaluate the health of bridges by exploring objective features and correlations of multiple monitoring factors. This model aims to learn representative features from raw monitoring data of bridge factors (i.e., strain, temperature, traffic flow, and heavy vehicle number), and classify the comprehensive features into different health degrees in a single framework. Specifically, in terms of characteristics of bridge factors (e.g., time sequence and heterogeneity), a hierarchical learning structure with multiple convolutional, pooling, and dense layers is presented to learn the representations of the whole bridge monitoring data. The structure contributes to capturing rich information from different observed factors and improving feature learning ability through designing particular neural networks for each bridge factor in accordance with their corresponding data structure. Additionally, a classification scheme with multiple fully connection layers and support vector machine in the framework is designed to achieve higher evaluation accuracy. Experiments on real-world monitoring data of a specific bridge validate the superiority of the proposed model.

The prediction of gear safety factors has long been a challenging issue in vehicle transmission industry due to its difficulty in investigating meaningful representation from insufficient and highly coupled parameters. This paper presents a framework to find a high-quality solution that improves prediction accuracy of gear safety factors. In the framework, to cope with the insufficiency of gear data, a generative model based on generative adversarial networks (GAN) is established and proven to generate acceptable data with Adam optimizer. Then, eigen-error principal component analysis (E-PCA) is proposed to extract features of gear parameters by reconstructing error function with eigenvectors and eigenvalues. Finally, particle swarm optimization and back propagation (PSO-BP) are applied to predict safety factors with these extracted features. Experimental results on real-world gear data of vehicle transmissions have validated the effectiveness of our proposed framework.

Power efficiency for license plate recognition (LPR) under unconstrained scenarios is a crucial factor in many edge-based real-world applications, e.g., autonomous vehicles whose power budget is limited. While a bulk of prior works have explored LPR approaches for unconstrained situations on CPU and GPU servers, these methods result in huge power dissipation, and are ineffective in challenging scenes. In this work, we present a mixed-precision (MP) Transformer hardware architecture to meet the requirements of power efficiency and satisfactory LPR accuracy in unconstrained scenarios, dedicated to implementing power-efficient edge accelerators for difficult LPR tasks. Firstly, MP-LPR Transformer model is proposed, where novel mixed-precision quantization and non-linear approximation techniques are tailored for low bit-width inference. Secondly, vector tiling systolic array Transformer architecture (VTSATA) is proposed with unique vector tiling systolic array (VTSA) and vector ALU (VALU) designs, where VTSA is used to accelerate matrix multiplications, and VALU is used to process non-linear operations. Finally, a FPGA accelerator prototype based on our approach is developed. Experimental results demonstrate that our hardware platform can reduce power consumption more than $20\times$ compared to GPU platforms, and for challenge subset on CCPD dataset, our method can further improve nearly $1\%$ accuracy with respect to the state-of-the-art performance.