Unsteady heat transfer of long gas insulated transmission lines in tunnel: Model and analysis.

The temperature of the cable core and cable casing is crucial for the safety and efficiency of the transmission system. Accurately predicting the distribution and variation of the transmission temperature field in the gas insulated transmission lines (GIL) tunnel is essential to ensure the long GIL tunnel transmission system's safe and stable operation. This paper addresses the challenge of calculating unsteady heat transfer flow in extra-long GIL transmission tunnels. The paper proposes a model, and a rapid solution method for extra-long GIL transmission heat transfer flow. In addition, the temperature variation during the operation of 1000 kV GIL has been analyzed. The results indicated that the conductor and cable casing temperature gradually increases and tends to be relatively stable with the operation time. The research results can provide theoretical basis and data support for the regulation of GIL tunnel transmission operation.

Performance-Aware Approximation of Global Channel Pruning for Multitask CNNs.

Global channel pruning (GCP) aims to remove a subset of channels (filters) across different layers from a deep model without hurting the performance. Previous works focus on either single task model pruning or simply adapting it to multitask scenario, and still face the following problems when handling multitask pruning: 1) Due to the task mismatch, a well-pruned backbone for classification task focuses on preserving filters that can extract category-sensitive information, causing filters that may be useful for other tasks to be pruned during the backbone pruning stage; 2) For multitask predictions, different filters within or between layers are more closely related and interacted than that for single task prediction, making multitask pruning more difficult. Therefore, aiming at multitask model compression, we propose a Performance-Aware Global Channel Pruning (PAGCP) framework. We first theoretically present the objective for achieving superior GCP, by considering the joint saliency of filters from intra- and inter-layers. Then a sequentially greedy pruning strategy is proposed to optimize the objective, where a performance-aware oracle criterion is developed to evaluate sensitivity of filters to each task and preserve the globally most task-related filters. Experiments on several multitask datasets show that the proposed PAGCP can reduce the FLOPs and parameters by over 60% with minor performance drop, and achieves 1.2x  âˆ¼ 3.3x acceleration on both cloud and mobile platforms. Our code is available at http://www.github.com/HankYe/PAGCP.git.

A biochip based on shell-isolated Au@MnO2 nanoparticle array-enhanced fluorescence effect for simple and sensitive exosome assay.

Exosomes, carrying specific molecular information of their parent cells, have been regarded as a kind of promising noninvasive biomarker for liquid biopsy. Plentiful fluorescence methods have been proposed for exosome assay. However, most of them are dependent on nucleic acid signal amplification strategies, which require complicated sequence design and experimental operation. Herein, a metal-enhanced fluorescence (MEF) biochip based on shell-isolated Au@MnO2 nanoparticle array was designed for simple and sensitive assay of exosomes. The designed method consists of only two parts: signal conversion and MEF amplification. The conversion of exosome signals to DNA signals was realized by means of chain displacement reaction. The subtle conversion effectively averts the effect of steric hindrance on MEF while amplifying the signal easily for the first time. The MEF biochip based on shell-isolated Au@MnO2 nanoparticle array achieves a second signal amplification in a simple way. Profiting from the two signal amplifications, this strategy displays high sensitivity toward exosomes with a detection limit of 4.5 × 103 particles µL-1. Compared with the result without MEF, the sensitivity is enhanced about thirty times. As far as we know, this is the first attempt for exosome assay by using MEF strategy. In addition to the favorable fluorescence enhancement, both shell-isolated Au@MnO2 nanoparticles and Au@MnO2 nanoparticle array show excellent stability in buffer solutions, which is conducive to practical application. Moreover, the proposed method is able to distinguish breast cancer patients from healthy people, showing its potential for exosome-based liquid biopsy.

Sample-Centric Feature Generation for Semi-Supervised Few-Shot Learning.

Semi-supervised few-shot learning aims to improve the model generalization ability by means of both limited labeled data and widely-available unlabeled data. Previous works attempt to model the relations between the few-shot labeled data and extra unlabeled data, by performing a label propagation or pseudo-labeling process using an episodic training strategy. However, the feature distribution represented by the pseudo-labeled data itself is coarse-grained, meaning that there might be a large distribution gap between the pseudo-labeled data and the real query data. To this end, we propose a sample-centric feature generation (SFG) approach for semi-supervised few-shot image classification. Specifically, the few-shot labeled samples from different classes are initially trained to predict pseudo-labels for the potential unlabeled samples. Next, a semi-supervised meta-generator is utilized to produce derivative features centering around each pseudo-labeled sample, enriching the intra-class feature diversity. Meanwhile, the sample-centric generation constrains the generated features to be compact and close to the pseudo-labeled sample, ensuring the inter-class feature discriminability. Further, a reliability assessment (RA) metric is developed to weaken the influence of generated outliers on model learning. Extensive experiments validate the effectiveness of the proposed feature generation approach on challenging one- and few-shot image classification benchmarks.

Point Cloud Instance Segmentation With Semi-Supervised Bounding-Box Mining.

Point cloud instance segmentation has achieved huge progress with the emergence of deep learning. However, these methods are usually data-hungry with expensive and time-consuming dense point cloud annotations. To alleviate the annotation cost, unlabeled or weakly labeled data is still less explored in the task. In this paper, we introduce the first semi-supervised point cloud instance segmentation framework (SPIB) using both labeled and unlabelled bounding boxes as supervision. To be specific, our SPIB architecture involves a two-stage learning procedure. For stage one, a bounding box proposal generation network is trained under a semi-supervised setting with perturbation consistency regularization (SPCR). The regularization works by enforcing an invariance of the bounding box predictions over different perturbations applied to the input point clouds, to provide self-supervision for network learning. For stage two, the bounding box proposals with SPCR are grouped into some subsets, and the instance masks are mined inside each subset with a novel semantic propagation module and a property consistency graph module. Moreover, we introduce a novel occupancy ratio guided refinement module to refine the instance masks. Extensive experiments on the challenging ScanNet v2 dataset demonstrate our method can achieve competitive performance compared with the recent fully-supervised methods.

Label-free detection of exosomes based on ssDNA-modulated oxidase-mimicking activity of CuCo2O4 nanorods.

A label-free method for exosome detection was proposed. It is based on the target-responsive controllability of oxidase-like activity of Cu/Co bimetallic metal-organic frameworks (CuCo2O4 nanorods). In the absence of exosomes, the oxidase-like activity was inhibited due to the adsorption of CD63 aptamer onto nanorods' surface. In the presence of exosomes, CD63 aptamer was disassembled from CuCo2O4 nanorods by virtue of CD63 aptamer-exosome recognition, which resulted in the recovery of oxidase-like activity. The activity inhibition is attributed to the fact that the ssDNA adsorption hindered the electron transfer between CuCo2O4 nanorods and colorimetric substrates. Under optimal conditions, a sensitive colorimetric method for detecting exosomes was established over a range of 5.6 × 104 to 8.9 × 105 particles µL-1 with a detection limit of 4.5 × 103 particles µL-1. The method was further applied in distinguishing healthy people and breast cancer patients by testing exosomes in the serum samples and showed satisfying differentiation ability.

Two D-π-A Schiff-Base-Functionalized Silica Gel Adsorbents for Preconcentration of Copper Ions in Foods and Water for Detection.

Copper is an essential element in many biological processes and plays an important role in carbohydrate and lipid metabolism. Excess or deficiency of Cu ions can cause disturbances in cellular homeostasis and damage the central nervous system. Here, for the first time, two functionalized silica gel (SG-A and SG-B) adsorbents were prepared and tested for copper detection via the reactions of chlorinated silica gel with two novel D-π-A Schiff base compounds: 2-amino-3-(quinolin-2-ylmethyleneamino)maleonitrile (A) and 2-(4-(diethylamino)-2-hydroxybenzylideneamino)-3-aminomaleonitrile (B) in the thionyl chloride solution, respectively. SG-A and SG-B as adsorbents filled in a microcolumn were used to enrich trace Cu ions in foods and water with the detection of flame atomic absorption spectrometry. Because of the strong coordination between two D-π-A Schiff base compounds and Cu2+ ions, the stable heterocyclic Cu2+-SG-A/B complex is formed. For a sample volume of 30 mL, detection limits of 0.09 µg L-1 and 0.15 µg L-1 have been achieved. The results of selectivity study show that the two adsorbents can selectively extract Cu2+ in complex matrixes with other metal cations. The methods have been successfully applied to the determination of Cu2+ content in various real samples, and the detection sensitivity that we report here is better than most results reported using modified silica gels.

SS-HCNN: Semi-Supervised Hierarchical Convolutional Neural Network for Image Classification.

The availability of large-scale annotated data and uneven separability of different data categories become two major impediments of deep learning for image classification. In this paper, we present a Semi-Supervised Hierarchical Convolutional Neural Network (SS-HCNN) to address these two challenges. A large-scale unsupervised maximum margin clustering technique is designed, which splits images into a number of hierarchical clusters iteratively to learn cluster-level CNNs at parent nodes and category-level CNNs at leaf nodes. The splitting uses the similarity of CNN features to group visually similar images into the same cluster, which relieves the uneven data separability constraint. With the hierarchical cluster-level CNNs capturing certain high-level image category information, the category-level CNNs can be trained with a small amount of labelled images, and this relieves the data annotation constraint. A novel cluster splitting criterion is also designed which automatically terminates the image clustering in the tree hierarchy. The proposed SS-HCNN has been evaluated on the CIFAR-100 and ImageNet classification datasets. Experiments show that the SS-HCNN trained using a portion of labelled training images can achieve comparable performance with other fully trained CNNs using all labelled images. Additionally, the SS-HCNN trained using all labelled images clearly outperforms other fully trained CNNs.

S-CNN: Subcategory-Aware Convolutional Networks for Object Detection.

The marriage between the deep convolutional neural network (CNN) and region proposals has made breakthroughs for object detection in recent years. While the discriminative object features are learned via a deep CNN for classification, the large intra-class variation and deformation still limit the performance of the CNN based object detection. We propose a subcategory-aware CNN (S-CNN) to solve the object intra-class variation problem. In the proposed technique, the training samples are first grouped into multiple subcategories automatically through a novel instance sharing maximum margin clustering process. A multi-component Aggregated Channel Feature (ACF) detector is then trained to produce more latent training samples, where each ACF component corresponds to one clustered subcategory. The produced latent samples together with their subcategory labels are further fed into a CNN classifier to filter out false proposals for object detection. An iterative learning algorithm is designed for the joint optimization of image subcategorization, multi-component ACF detector, and subcategory-aware CNN classifier. Experiments on INRIA Person dataset, Pascal VOC 2007 dataset and MS COCO dataset show that the proposed technique clearly outperforms the state-of-the-art methods for generic object detection.