Disambiguation Enabled Linear Discriminant Analysis for Partial Label Dimensionality Reduction

As an emerging weakly supervised learning framework, partial label learning considers inaccurate supervision where each training example is associated with multiple candidate labels among which only one is valid. In this article, a first attempt toward employing dimensionality reduction to help improve the generalization performance of partial label learning system is investigated. Specifically, the popular linear discriminant analysis (LDA) techniques are endowed with the ability of dealing with partial label training examples. To tackle the challenge of unknown ground-truth labeling information, a novel learning approach named Delin is proposed which alternates between LDA dimensionality reduction and candidate label disambiguation based on estimated labeling confidences over candidate labels. On one hand, the (kernelized) projection matrix of LDA is optimized by utilizing disambiguation-guided labeling confidences. On the other hand, the labeling confidences are disambiguated by resorting to k NN aggregation in the LDA-induced feature space. Extensive experiments over a broad range of partial label datasets clearly validate the effectiveness of Delin in improving the generalization performance of well-established partial label learning algorithms.

An Advanced Attention-Enhanced Hybrid Deep Learning Model for WBSNs’ Gait Pattern Recognition

Background: The deep learning techniques have been attracted increasing attention on wireless body sensor networks (WBSNs) gait pattern recognition that has a great contribution to monitoring gait change in clinical application. However, in existing studies, there are some challenging issues such as low generalization performance and no potential interpretation for gait variability. It is necessary to search for the advanced deep learning models to resolve these issues. Method: A public WARD database including acceleration and gyroscope data acquired from each subject wearing five sensors was selected, and the gait with different combination of on-body multi-sensors is considered as a WBSNs’ gait pattern. An advanced attention-enhanced hybrid deep learning model of DCNN and LSTM for WBSNs’ gait pattern recognition was proposed. In our proposed technique, the combination model of DCNN with LSTM is firstly to discover the spatial-temporary gait correlation features. And then the attention mechanism is introduced to exploit the more valuable intrinsic nonlinear dynamic correlation gait characteristics associated with gait variability hidden in spatial-temporary gait space obtained. This significantly contributes to enhancing the generalization performance and taking insight on gait variability in a certain anatomical region. Results: The ten gait patterns are randomly selected from WARD database to evaluate the feasibility of our proposed method. Our experiments demonstrated the superior generalization ability of our method to some models such as CNN-LSTM, DCNN-LSTM. Our proposed model could classify ten gait patterns with the highest accuracy and F1-score of 91.48% and 91.46%, respectively. Moreover, we also found that the classification performance of a certain gait pattern was almost same best when the combinations of three or five on-body sensors were employed respectively, suggesting that our method possibly take insight on gait variability in a certain anatomical region. Conclusion: Our proposed technique could feasibly discover the more intrinsic nonlinear dynamic correlation gait characteristics associated with gait variability from on-body multi-sensors gait data, which greatly contributed to best generalization performance and potential clinical interpretation. Our proposed technique would hopefully become a powerful tool of monitoring gait change in clinical application.

A CNN-ABiGRU method for Gearbox Fault Diagnosis

As a key part of modern industrial machinery, there has been a lot of fault diagnosis methods for gearbox. However, traditional fault diagnosis methods suffer from dependence on prior knowledge. This paper proposed an end-to-end method based on convolutional neural network (CNN), Bidirectional gated recurrent unit (BiGRU), and Attention Mechanism. Among them, the application of BiGRU not only made perfect use of the time sequence of signal, but also saved computing resources more than the same type of networks because of the low amount of calculation. In order to verify the effectiveness and generalization performance of the proposed method, experiments are carried out on two datasets, and the accuracy is calculated by the ten-fold crossvalidation. Compared with the existing fault diagnosis methods, the experimental results show that the proposed model has higher accuracy.

A Robust Training Method for Pathological Cellular Detector via Spatial Loss Calibration

Computer-aided diagnosis of pathological images usually requires detecting and examining all positive cells for accurate diagnosis. However, cellular datasets tend to be sparsely annotated due to the challenge of annotating all the cells. However, training detectors on sparse annotations may be misled by miscalculated losses, limiting the detection performance. Thus, efficient and reliable methods for training cellular detectors on sparse annotations are in higher demand than ever. In this study, we propose a training method that utilizes regression boxes' spatial information to conduct loss calibration to reduce the miscalculated loss. Extensive experimental results show that our method can significantly boost detectors' performance trained on datasets with varying degrees of sparse annotations. Even if 90% of the annotations are missing, the performance of our method is barely affected. Furthermore, we find that the middle layers of the detector are closely related to the generalization performance. More generally, this study could elucidate the link between layers and generalization performance, provide enlightenment for future research, such as designing and applying constraint rules to specific layers according to gradient analysis to achieve “scalpel-level” model training.

Equivalent Modeling of Microgrids Based on Optimized Broad Learning System

The DC microgrid is an important structure of microgrids. Aiming at the problem of the grid-connected DC microgrid modeling, a grid-connected DC microgrid equivalent modeling method based on the optimized Broad Learning System (BLS) is proposed. Taking the electrical parameter data of the grid-connected DC microgrid access point as the training data set of BLS, the equivalent model of the grid-connected equivalent model is constructed. In order to further improve the accuracy and generalization performance of the model, the shark smell optimization (SSO) algorithm is used to optimize the input weights and thresholds of the BLS. Furthermore, the shark smell optimization-Broad Learning System (SSO-BLS) algorithm is proposed. SSO-BLS is compared with RBF, BLS, BFO-ELM, and other algorithms. The results show that the grid-connected DC microgrid model based on SSO-BLS has good accuracy and generalization characteristics.

Neural Network for Intelligent and Efficient Volleyball Passing Training

Passing is a relatively basic technique in volleyball. In volleyball passing teaching, training the correct passing technique plays a very important role. The correct pass can not only accurately grasp the direction of the ball point and the drop point but also effectively connect the defense and the offense. In order to improve the efficiency and quality of volleyball passing training, improve the precise extraction of sport targets, reduce redundant feature information, and improve the generalization performance and nonlinear fitting capabilities of the algorithm, this paper studies volleyball based on the nested convolutional neural network model and passing training wrong movement detection method. The structure of the convolutional neural network is improved by nesting mlpconv layers, and the Gaussian mixture model is used to effectively and accurately extract the foreground objects in the video. The nested multilayer mlpconv layer automatically learns the deep-level features of the foreground target, and the generated feature map is vectorized and input to the Softmax classifier connected to the fully connected layer for passing wrong behavior detection in volleyball training. Based on the detection of nearly 1,000 athletes’ action datasets, the simulation experiment results show that the algorithm reduces the acquisition of redundant information and shortens the calculation time and learning time of the algorithm, and the improved convolutional neural network has generalization performance and nonlinearity. The fitting ability has been improved, and the detection of abnormal volleyball passing behaviors has achieved a higher accuracy rate.

Deepfake-Image Anti-Forensics with Adversarial Examples Attacks

Many deepfake-image forensic detectors have been proposed and improved due to the development of synthetic techniques. However, recent studies show that most of these detectors are not immune to adversarial example attacks. Therefore, understanding the impact of adversarial examples on their performance is an important step towards improving deepfake-image detectors. This study developed an anti-forensics case study of two popular general deepfake detectors based on their accuracy and generalization. Herein, we propose the Poisson noise DeepFool (PNDF), an improved iterative adversarial examples generation method. This method can simply and effectively attack forensics detectors by adding perturbations to images in different directions. Our attacks can reduce its AUC from 0.9999 to 0.0331, and the detection accuracy of deepfake images from 0.9997 to 0.0731. Compared with state-of-the-art studies, our work provides an important defense direction for future research on deepfake-image detectors, by focusing on the generalization performance of detectors and their resistance to adversarial example attacks.

Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices

This study addresses a series of methodological questions that arise when modeling inflectional morphology with Linear Discriminative Learning. Taking the semi-productive German noun system as example, we illustrate how decisions made about the representation of form and meaning influence model performance. We clarify that for modeling frequency effects in learning, it is essential to make use of incremental learning rather than the end-state of learning. We also discuss how the model can be set up to approximate the learning of inflected words in context. In addition, we illustrate how in this approach the wug task can be modeled. The model provides an excellent memory for known words, but appropriately shows more limited performance for unseen data, in line with the semi-productivity of German noun inflection and generalization performance of native German speakers.