High-quality and efficient large-area copper removal utilizing laser-induced active mechanical peeling.

Large-area copper layer removal is one of the essential processes in manufacturing printed circuit boards (PCB) and frequency selective surfaces (FSS). However, laser direct ablation (LDA) with one-step scanning is challenging in resolving excessive substrate damage and material residue. Here, this study proposes a laser scanning strategy based on the laser-induced active mechanical peeling (LIAMP) effect generated by resin decomposition. This scanning strategy allows the removal of large-area copper layers from FR-4 copper-clad laminates (FR-4 CCL) in one-step scanning without additional manual intervention. During the removal process, the resin decomposition in the laser-irradiated area provides the mechanical tearing force, while the resin decomposition in the laser-unirradiated area reduces the interfacial adhesion force and provides recoil pressure. By optimizing scanning parameters to control the laser energy deposition, the substrate damage and copper residue can be effectively avoided. In our work, the maximum removal efficiency with different energy densities, pulse duration, and repetition frequency are 31.8 mm2/ms, 30.25 mm2/ms, and 82.8 mm2/ms, respectively. Compared with the reported copper removal using laser direct write lithography technology combined with wet chemical etching (LDWL+WCE) and LDA, the efficiency improved by 8.3 times and 66 times. Predictably, the laser scanning strategy and the peeling mechanism are simple and controllable, which have potential in electronics, communications, and aerospace.

The Influence of Laser Process Parameters on the Adhesion Strength between Electroless Copper and Carbon Fiber Composites Determined Using Response Surface Methodology.

Laser process technology provides a feasible method for directly manufacturing surface-metallized carbon fiber composites (CFCs); however, the laser's process parameters strongly influence on the adhesion strength between electroless copper and CFCs. Here, a nanosecond ultraviolet laser was used to fabricate electroless copper on the surface of CFCs. In order to achieve good adhesion strength, four key process parameters, namely, the laser power, scanning line interval, scanning speed, and pulse frequency, were optimized experimentally using response surface methodology, and a central composite design was utilized to design the experiments. An analysis of variance was conducted to evaluate the adequacy and significance of the developed regression model. Also, the effect of the process parameters on the adhesion strength was determined. The numerical analysis indicated that the optimized laser power, scanning line interval, scanning speed, and pulse frequency were 5.5 W, 48.2 µm, 834.0 mm/s, and 69.5 kHz, respectively. A validation test confirmed that the predicted results were consistent with the actual values; thus, the developed mathematical model can adequately predict responses within the limits of the laser process parameters being used.

Incremental Incomplete Concept-Cognitive Learning Model: A Stochastic Strategy.

Concept-cognitive learning is an emerging area of cognitive computing, which refers to continuously learning new knowledge by imitating the human cognition process. However, the existing research on concept-cognitive learning is still at the level of complete cognition as well as cognitive operators, which is far from the real cognition process. Meanwhile, the current classification algorithms based on concept-cognitive learning models (CCLMs) are not mature enough yet since their cognitive results highly depend on the cognition order of attributes. To address the above problems, this article presents a novel concept-cognitive learning method, namely, stochastic incremental incomplete concept-cognitive learning method (SI2CCLM), whose cognition process adopts a stochastic strategy that is independent of the order of attributes. Moreover, a new classification algorithm based on SI2CCLM is developed, and the analysis of the parameters and convergence of the algorithm is made. Finally, we show the cognitive effectiveness of SI2CCLM by comparing it with other concept-cognitive learning methods. In addition, the average accuracy of our model on 24 datasets is 82.02%, which is higher than the compared 20 classification algorithms, and the elapsed time of our model also has advantages.

Typicality-Aware Adaptive Similarity Matrix for Unsupervised Learning.

Graph-based clustering approaches, especially the family of spectral clustering, have been widely used in machine learning areas. The alternatives usually engage a similarity matrix that is constructed in advance or learned from a probabilistic perspective. However, unreasonable similarity matrix construction inevitably leads to performance degradation, and the sum-to-one probability constraints may make the approaches sensitive to noisy scenarios. To address these issues, the notion of typicality-aware adaptive similarity matrix learning is presented in this study. The typicality (possibility) rather than the probability of each sample being a neighbor of other samples is measured and adaptively learned. By introducing a robust balance term, the similarity between any pairs of samples is only related to the distance between them, yet it is not affected by other samples. Therefore, the impact caused by the noisy data or outliers can be alleviated, and meanwhile, the neighborhood structures can be well captured according to the joint distance between samples and their spectral embeddings. Moreover, the generated similarity matrix has block diagonal properties that are beneficial to correct clustering. Interestingly, the results optimized by the typicality-aware adaptive similarity matrix learning share the common essence with the Gaussian kernel function, and the latter can be directly derived from the former. Extensive experiments on synthetic and well-known benchmark datasets demonstrate the superiority of the proposed idea when comparing with some state-of-the-art methods.

Interval Dominance-Based Feature Selection for Interval-Valued Ordered Data.

Dominance-based rough approximation discovers inconsistencies from ordered criteria and satisfies the requirement of the dominance principle between single-valued domains of condition attributes and decision classes. When the ordered decision system (ODS) is no longer single-valued, how to utilize the dominance principle to deal with multivalued ordered data is a promising research direction, and it is the most challenging step to design a feature selection algorithm in interval-valued ODS (IV-ODS). In this article, we first present novel thresholds of interval dominance degree (IDD) and interval overlap degree (IOD) between interval values to make the dominance principle applicable to an IV-ODS, and then, the interval-valued dominance relation in the IV-ODS is constructed by utilizing the above two developed parameters. Based on the proposed interval-valued dominance relation, the interval-valued dominance-based rough set approach (IV-DRSA) and their corresponding properties are investigated. Moreover, the interval dominance-based feature selection rules based on IV-DRSA are provided, and the relevant algorithms for deriving the interval-valued dominance relation and the feature selection methods are established in IV-ODS. To illustrate the effectiveness of the parameters variation on feature selection rules, experimental evaluation is performed using 12 datasets coming from the University of California-Irvine (UCI) repository.

A Review of Generalized Zero-Shot Learning Methods.

Generalized zero-shot learning (GZSL) aims to train a model for classifying data samples under the condition that some output classes are unknown during supervised learning. To address this challenging task, GZSL leverages semantic information of the seen (source) and unseen (target) classes to bridge the gap between both seen and unseen classes. Since its introduction, many GZSL models have been formulated. In this review paper, we present a comprehensive review on GZSL. First, we provide an overview of GZSL including the problems and challenges. Then, we introduce a hierarchical categorization for the GZSL methods and discuss the representative methods in each category. In addition, we discuss the available benchmark data sets and applications of GZSL, along with a discussion on the research gaps and directions for future investigations.

Adversarial Learning With Cost-Sensitive Classes.

It is necessary to improve the performance of some special classes or to particularly protect them from attacks in adversarial learning. This article proposes a framework combining cost-sensitive classification and adversarial learning together to train a model that can distinguish between the protected and unprotected classes, such that the protected classes are less vulnerable to adversarial examples. We find in this framework an interesting phenomenon during the training of deep neural networks, called the Min-Max property, that is, the absolute values of most parameters in the convolutional layer approach 0 while the absolute values of a few parameters are significantly larger, becoming bigger. Based on this Min-Max property which is formulated and analyzed in a view of random distribution, we further build a new defense model against adversarial examples for adversarial robustness improvement. An advantage of the built model is that it performs better than the standard one and can combine with adversarial training to achieve improved performance. It is experimentally confirmed that, regarding the average accuracy of all classes, our model is almost as same as the existing models when an attack does not occur and is better than the existing models when an attack occurs. Specifically, regarding the accuracy of protected classes, the proposed model is much better than the existing models when an attack occurs.

Semantic Correlation Transfer for Heterogeneous Domain Adaptation.

Heterogeneous domain adaptation (HDA) is expected to achieve effective knowledge transfer from a label-rich source domain to a heterogeneous target domain with scarce labeled data. Most prior HDA methods strive to align the cross-domain feature distributions by learning domain invariant representations without considering the intrinsic semantic correlations among categories, which inevitably results in the suboptimal adaptation performance across domains. Therefore, to address this issue, we propose a novel semantic correlation transfer (SCT) method for HDA, which not only matches the marginal and conditional distributions between domains to mitigate the large domain discrepancy, but also transfers the category correlation knowledge underlying the source domain to target by maximizing the pairwise class similarity across source and target. Technically, the domainwise and classwise centroids (prototypes) are first computed and aligned according to the feature embeddings. Then, based on the derived classwise prototypes, we leverage the cosine similarity of each two classes in both domains to transfer the supervised source semantic correlation knowledge among different categories to target effectively. As a result, the feature transferability and category discriminability can be simultaneously improved during the adaptation process. Comprehensive experiments and ablation studies on standard HDA tasks, such as text-to-image, image-to-image, and text-to-text, have demonstrated the superiority of our proposed SCT against several state-of-the-art HDA methods.

MRGAT: Multi-Relational Graph Attention Network for knowledge graph completion.

One of the most effective ways to solve the problem of knowledge graph completion is embedding-based models. Graph neural networks (GNNs) are popular and promising embedding models which can exploit and use the structural information of neighbors in knowledge graphs. The current GNN-based knowledge graph completion methods assume that all neighbors of a node have equal importance. This assumption which cannot assign different weights to neighbors is pointed out in our study to be unreasonable. In addition, since the knowledge graph is a kind of heterogeneous graph with multiple relations, multiple complex interactions between nodes and neighbors can bring challenges to the effective message passing of GNNs. We then design a multi-relational graph attention network (MRGAT) which can adapt to different cases of heterogeneous multi-relational connections and then calculate the importance of different neighboring nodes through a self-attention layer. The incorporation of self-attention mechanism into the network with different node weights optimizes the network structure, and therefore, significantly results in a promotion of performance. We experimentally validate the rationality of our models on multiple benchmark knowledge graphs, where MRGAT achieves the best performance on various evaluation metrics including MRR score, Hits@ score compared with other state-of-the-art baseline models.

Towards improving fast adversarial training in multi-exit network.

Adversarial examples are usually generated by adding adversarial perturbations on clean samples, designed to deceive the model to make wrong classifications. Adversarial robustness refers to the ability of a model to resist adversarial attacks. And currently, a mainstream method to enhance adversarial robustness is the Projected Gradient Descent (PGD). However, PGD is often criticized for being time-consuming during constructing adversarial examples. Fast adversarial training can improve the adversarial robustness in shorter time, but it only can train for a limited number of epochs, leading to sub-optimal performance. This paper demonstrates that the multi-exit network can reduce the impact of adversarial perturbations by outputting easily identified samples at early exits. Therefore, we can improve the adversarial robustness. Further, we find that the multi-exit network can prevent catastrophic overfitting existing in single-step adversarial training. Specifically, we find that, in the multi-exit network, (1) the norm of weights at a fully connected layer in a non-overfitted exit is much smaller than that in an overfitted exit; and (2) catastrophic overfitting occurs when the late exits have weight norms larger than the early exits. Based on these findings, we propose an approach to alleviating the catastrophic overfitting of the multi-exit network. Compared to PGD adversarial training, our approach can train a model with decreased time complexity and increased empirical robustness. Extensive experiments have been conducted to evaluate our approach against various adversarial attacks, and the experimental results demonstrate superior robustness accuracies on CIFAR-10, CIFAR-100 and SVHN.

Bidirectional stochastic configuration network for regression problems.

To adapt to the reality of limited computing resources of various terminal devices in industrial applications, a randomized neural network called stochastic configuration network (SCN), which can conduct effective training without GPU, was proposed. SCN uses a supervisory random mechanism to assign its input weights and hidden biases, which makes it more stable than other randomized algorithms but also leads to time-consuming model training. To alleviate this problem, we propose a novel bidirectional SCN algorithm (BSCN) in this paper, which divides the way of adding hidden nodes into two modes: forward learning and backward learning. In the forward learning mode, BSCN still uses the supervisory mechanism to configure the parameters of the newly added nodes, which is the same as SCN. In the backward learning mode, BSCN calculates the parameters at one time based on the residual error feedback of the current model. The two learning modes are performed iteratively until the prediction error of the model reaches an acceptable level or the number of hidden nodes reaches its maximum value. This semi-random learning mechanism greatly speeds up the training efficiency of the BSCN model and significantly improves the quality of the hidden nodes. Extensive experiments on ten benchmark regression problems, two real-life air pollution prediction problems, and a classical image processing problem show that BSCN can achieve faster training speed, higher stability, and better generalization ability than SCN.

Overexpressed P75CUX1 promotes EMT in glioma infiltration by activating ß-catenin.

The homeobox protein cut-like 1 (CUX1) comprises three isoforms and has been shown to be involved in the development of various types of malignancies. However, the expression and role of the CUX1 isoforms in glioma remain unclear. Herein, we first identified that P75CUX1 isoform exhibited consistent expression among three isoforms in glioma with specifically designed antibodies to identify all CUX1 isoforms. Moreover, a significantly higher expression of P75CUX1 was found in glioma compared with non-tumor brain (NB) tissues, analyzed with western blot and immunohistochemistry, and the expression level of P75CUX1 was positively associated with tumor grade. In addition, Kaplan-Meier survival analysis indicated that P75CUX1 could serve as an independent prognostic indicator to identify glioma patients with poor overall survival. Furthermore, CUX1 knockdown suppressed migration and invasion of glioma cells both in vitro and in vivo. Mechanistically, this study found that P75CUX1 regulated epithelial-mesenchymal transition (EMT) process mediated via ß-catenin, and CUX1/ß-catenin/EMT is a novel signaling cascade mediating the infiltration of glioma. Besides, CUX1 was verified to promote the progression of glioma via multiple other signaling pathways, such as Hippo and PI3K/AKT. In conclusion, we suggested that P75CUX1 could serve as a potential prognostic indicator as well as a novel treatment target in malignant glioma.

AdvKin: Adversarial Convolutional Network for Kinship Verification.

Kinship verification in the wild is an interesting and challenging problem. The goal of kinship verification is to determine whether a pair of faces are blood relatives or not. Most previous methods for kinship verification can be divided as handcrafted features-based shallow learning methods and convolutional neural network (CNN)-based deep-learning methods. Nevertheless, these methods are still facing the challenging task of recognizing kinship cues from facial images. The reason is that the family ID information and the distribution difference of pairwise kin-faces are rarely considered in kinship verification tasks. To this end, a family ID-based adversarial convolutional network (AdvKin) method focused on discriminative Kin features is proposed for both small-scale and large-scale kinship verification in this article. The merits of this article are four-fold: 1) for kin-relation discovery, a simple yet effective self-adversarial mechanism based on a negative maximum mean discrepancy (NMMD) loss is formulated as attacks in the first fully connected layer; 2) a pairwise contrastive loss and family ID-based softmax loss are jointly formulated in the second and third fully connected layer, respectively, for supervised training; 3) a two-stream network architecture with residual connections is proposed in AdvKin; and 4) for more fine-grained deep kin-feature augmentation, an ensemble of patch-wise AdvKin networks is proposed (E-AdvKin). Extensive experiments on 4 small-scale benchmark KinFace datasets and 1 large-scale families in the wild (FIW) dataset from the first Large-Scale Kinship Recognition Data Challenge, show the superiority of our proposed AdvKin model over other state-of-the-art approaches.

Fast Supervised Topic Models for Short Text Emotion Detection.

With the development of social network platforms, discussion forums, and question answering websites, a huge number of short messages that typically contain a few words for an individual document are posted by online users. In these short messages, emotions are frequently embedded for communicating opinions, expressing friendship, and promoting influence. It is quite valuable to detect emotions from short messages, but the corresponding task suffers from the sparsity of feature space. In this article, we first generate term groups co-occurring in the same context to enrich the number of features. Then, two basic supervised topic models are proposed to associate emotions with topics accurately. To reduce the time cost of parameter estimation, we further propose an accelerated algorithm for our basic models. Extensive evaluations using three short corpora validate the efficiency and effectiveness of the accelerated models for predicting the emotions of unlabeled documents, in addition to generate the topic-level emotion lexicons.

MiR-4310 induced by SP1 targets PTEN to promote glioma progression.

BACKGROUND: miRNAs have been reported to be involved in multiple biological processes of gliomas. Here, we aimed to analyze miR-4310 and its correlation genes involved in the progression of human glioma. METHODS: miR-4310 expression levels were examined in glioma and non-tumor brain (NB) tissues. The molecular mechanisms of miR-4310 expression and its effects on cell proliferation, migration, and invasion were explored using 3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide, Transwell chamber, Boyden chamber, and western blot analyses, as well as its effect on tumorigenesis was explored in vivo in nude mice. The relationships between miR-4310, SP1, phosphatase, and tensin homolog (PTEN) were explored using chromatin immunoprecipitation, agarose gel electrophoresis, electrophoresis mobility shift, and dual-luciferase reporter gene assays. RESULTS: miR-4310 expression was upregulated in glioma tissues compared to that in NB tissues. Overexpressed miR-4310 promoted glioma cell proliferation, migration, and invasion in vitro, as well as tumorigenesis in vivo. The inhibition of miR-4310 expression was sufficient to reverse these results. Mechanistic analyses revealed that miR-4310 promoted glioma progression through the PI3K/AKT pathway by targeting PTEN. Additionally, SP1 induced the expression of miR-4310 by binding to its promoter region. CONCLUSION: miR-4310 promotes the progression of glioma by targeting PTEN and activating the PI3K/AKT pathway; meanwhile, the expression of miR-4310 was induced by SP1.

Identification of Prognostic Signatures of Alternative Splicing in Glioma.

Alternative splicing (AS) is a ubiquitous mechanism in which pre-mRNA can be spliced into divergent variants and involved in carcinogenesis and progression in several cancers. In the present study, we systematically profiled prognostic AS signatures involving both low grade glioma (LGG) and glioblastoma (GBM) and investigated the association of AS signatures with tumor grade and IDH1 status in glioma. Percent spliced in (PSI) values and corresponding clinical data were obtained from TCGA SpliceSeq and TCGA data portal, respectively. Prognostic AS signatures were identified using univariate and stepwise multivariate Cox regression. Heatmap analysis was performed based on prognostic AS signatures. A prognostic signature was established with 69 and 88 AS events, including specific splicing events of MUTYH, STEAP3, and CTNNB1, in LGG and GBM cohorts, respectively. The area under the curve (AUC) of the prediction model was 0.968 at 2000 days of overall survival (OS) in the LGG cohort and 0.966 at 450 days of OS in the GBM cohort. In addition, these prognostic AS signatures could complement current molecular classification, such as IDH1 mutation, 1p/19q codeletion, and ATRX loss, of glioma and further identify potential subgroups of glioma with the same molecular features. In conclusion, our study systematically profiled prognostic AS events involving both low grade glioma and glioblastoma for the first time, which also shed light on the crosstalk between AS signatures and molecular features of glioma.

MicroRNA-4476 promotes glioma progression through a miR-4476/APC/ß-catenin/c-Jun positive feedback loop.

Glioma has been a major healthcare burden; however, the specific molecular regulatory mechanism underlying its initiation and progression remains to be elucidated. Although it is known that many miRNAs are involved in the regulation of malignant phenotypes of glioma, the role of miR-4476 has not been reported yet. In the present study, we identify miR-4476 as an upregulated microRNA, which promotes cell proliferation, migration, and invasion in glioma. Further mechanistic analyses indicate that the adenomatous polyposis coli (APC), a negative regulator of the Wnt/ß-catenin signaling pathway, is a direct target of miR-4476 and mediates the oncogenic effects of miR-4476 in glioma. C-Jun, a downstream effector of the Wnt/ß-catenin signaling, is upregulated by miR-4476 overexpression. In turn, c-Jun could positively regulate miR-4476 expression by binding to the upstream of its transcription start site (TSS). Furthermore, in our clinical samples, increased miR-4476 is an unfavorable prognostic factor, and its expression positively correlates with c-Jun expression but negatively correlates with that of APC. In conclusion, our study demonstrates that miR-4476 acts as a tumor enhancer, directly targeting APC to stimulate its own expression and promoting the malignant phenotypes of glioma.

Decreased CDKL2 expression is correlated with the progression and poor prognosis of glioma.

Glioma is the most common form of malignant intracranial tumors. Cyclin-dependent kinase-like 2 (CDKL2) was observed in various regions of the brain, but the specific role of CDKL2 in glioma has not been reported yet. In the present study, the expression of CDKL2 mRNA was detected by real-time QPCR in freshly collected glioma and para-carcinoma tissues, and we collected genomic and clinical data from The Cancer Genome Atlas to determine mRNA expression levels of CDKL2 in the normal brain and glioma samples. Moreover, western blot assay and immunohistochemistry experiments were implemented to identify CDKL2 protein expression, and clinical pathology characteristics from 151 glioma cases and thirty-four para-carcinoma tissues were also examined. The relationship between the levels of CDKL2 expression and clinical data was analyzed. Low mRNA and protein expression of CDKL2 was observed in glioma tissues compared to non-cancerous tissues. In addition, low levels of CDKL2 correlated with Astrocytic type, higher clinical WHO grade, and higher Ki-67 expression in glioma. Low mRNA and protein expression of CDKL2 in glioma predicted an observably shorter overall survival time than high expression. However, as revealed by multivariate analysis, CDKL2 protein expression was not an independent prognostic biomarker for the survival of patients with glioma. Our study firstly determined that low levels of CDKL2 expression are associated with poor clinical diagnosis. Thus, CDKL2 may serve as a prognostic factor of glioma.

SKA1 promotes malignant phenotype and progression of glioma via multiple signaling pathways.

BACKGROUND: Spindle and kinetochore associated protein 1 (SKA1) is a protein involved in chromosome congression and mitosis. It has been found to be upregulated and oncogenic in several human cancers. Herein, we investigated the precise role of SKA1 in the progression and malignant phenotype of human glioma. METHODS: Bioinformatic analysis was carried out based on the RNA-seq data and corresponding clinical data from GEO, TCGA and CGGA databases. Western blot was performed to analyze the expression of SKA1 in clinical samples and signaling pathway proteins in glioma cells, respectively. CCK8 assay, colony forming assay and EdU assay were performed to assess the cell viability. Cell migration and invasion assays were also performed. Moreover, xenograft model was established and the expression of SKA1 was assessed in the xenograft by immunohistochemistry. RESULTS: SKA1 expression is positively correlated with glioma grade and could be a promising biomarker for GBM. Moreover, overexpression of SKA1 may lead to poor prognosis in glioma. Downregulation of SKA1 attenuated cell viability, migration, and invasion in U251, U87, LN229 and T98 cells. Furthermore, GSEA analysis demonstrated that SKA1 was involved in the cell cycle, EMT pathway as well as Wnt/ß-catenin signaling pathway, which were then confirmed with Western blot analysis. CONCLUSION: SKA1 promotes malignant phenotype and progression of glioma via multiple pathways, including cell cycle, EMT, Wnt/ß-catenin signaling pathway. Therefore, SKA1 could be a promising therapeutic target for the treatment of human gliomas.

Spectral Clustering of Customer Transaction Data With a Two-Level Subspace Weighting Method.

Finding customer groups from transaction data is very important for retail and e-commerce companies. Recently, a "Purchase Tree" data structure is proposed to compress the customer transaction data and a local PurTree spectral clustering method is proposed to cluster the customer transaction data. However, in the PurTree distance, the node weights for the children nodes of a parent node are set as equal and the differences between different nodes are not distinguished. In this paper, we propose a two-level subspace weighting spectral clustering (TSW) algorithm for customer transaction data. In the new method, a PurTree subspace metric is proposed to measure the dissimilarity between two customers represented by two purchase trees, in which a set of level weights are introduced to distinguish the importance of different tree levels and a set of sparse node weights are introduced to distinguish the importance of different tree nodes in a purchase tree. TSW learns an adaptive similarity matrix from the local distances in order to better uncover the cluster structure buried in the customer transaction data. Simultaneously, it learns a set of level weights and a set of sparse node weights in the PurTree subspace distance. An iterative optimization algorithm is proposed to optimize the proposed model. We also present an efficient method to compute a regularization parameter in TSW. TSW was compared with six clustering algorithms on ten benchmark data sets and the experimental results show the superiority of the new method.