Search

Directionally solidified Al2O3/Y3Al5O12 (YAG), Al2O3/YAG/ZrO2, Al2O3/GdAlO3 (GAP), Al2O3/GAP/ZrO2 and Al2O3/ZrO2 eutectic/non eutectic ceramics were prepared with laser zone melting. Their microstructure, phase component and shape were investigated. The results indicated that the molar ratio of Zr4+ to rare earth ions (Y3+, Gd3+) in the ZrO2 phase of the ternary eutectic was closed to 2:1. The free energy of chemical reaction was the key to determine the precipitation order of phase. The growth of the firstly-precipitated phase could be divided into the formation of ordered Bravais lattice composed by the oxygen ions, the diffusion of metal cations in this lattice and the change of phase morphology. The unfilled cation gap might be a fast channel for the ion transport. On the same growth plane, the nucleation of the later-precipitated phase began only after the firstly-precipitated phase was fully formed. The triangular was the main ion arrangement way for the stable nucleation and growth of phases. The effects of the precipitation temperature, the phase size, the eutectic type and the growth rate on the microstructure were discussed. This improved theory provided a more reasonable explanation for the formation of rod-shaped, layered, triangular, Chinese script (CS) and cellular microstructures, which might be of significance to regulate the morphology of microstructures.

Directionally solidified oxide eutectic ceramic (DSOEC) has been one of the focuses of ceramic materials because of its excellent mechanical properties and oxidation resistance at ultra-high temperature (above 1600 ℃). Al2O3/GdAlO3/ZrO2 ternary DSOEC was prepared with induction heating zone melting method, of which microstructure and mechanical properties’ involution with the growth rate was revealed. As the growth rate increased, the microstructure of Al2O3/GdAlO3/ZrO2 DSOEC became into a complex regular structure from an irregular network structure, and the eutectic spacing decreased continuously. Its fracture toughness increased first from 4.75 to 6.55 MPa m1/2 and then decreased to 5.48 MPa m1/2, while its hardness increased continuously from 12.20 to 15.56 GPa. The third component ZrO2 made GdAlO3 phase microstructure in the binary eutectic Al2O3/GdAlO3 from a complex regular structure transformed into an irregular network one in the ternary Al2O3/GdAlO3/ZrO2 for the low growth rate, while from the regular layered microstructure to the regular triangular one in the ternary eutectic for the high growth rate, ZrO2 phase also increased the eutectic spacing, improved the fracture toughness. The fracture toughness at room temperature (4.75–6.55 MPa m1/2) of Al2O3/GdAlO3/ZrO2 ternary DSOEC was much higher than that of Al2O3/GdAlO3 binary DSOEC (1.47–3.24 MPa m1/2).

Al2O3/Er3Al5O12(EAG)/ZrO2 directionally solidified eutectic ceramics (DSECs) were prepared with high frequency induction heating zone melting (IHZM), and heat treated at high temperature. Their phase components, microstructures, flexural and tensile strengths were investigated. The results indicated that the microstructures were composed of the mutually interpenetrated Al2O3, EAG and ZrO2 (stabilized by Er3+ ion) phases. As the growth rate increased from 2 mm/h to 20 mm/h, the phase size decreased and the irregular microstructure became regular. When the heat treatment temperature was up to 1400 °C, the ionic diffusion was triggered and the phase began to grow. Under the action of surface energy and special microstructure, the growth of phase stopped after it reached a certain threshold. The sample with smaller phase size exhibited higher strength at room temperature, which was dependent on the crack propagation resistance. However, the speed of strength reduction was faster. The formation of surface groove, microcrack and malignant growth phase was responsible for this rapidly decreasing of strength. The flexural and tensile strengths of eutectic ceramic with larger phase size remained at 671.6 MPa and 359.45 MPa after long-term heat treatment at 1500 °C, respectively. The smaller difference between the original phase size and the maximum coarsened phase size was beneficial to avoid the occurrence of the above mentioned harmful factors and was the source of strength stability.

Transition metal nitrides (e.g., TiN) have shown tremendous promise in optical metamaterials for nanophotonic devices due to their plasmonic properties comparable to noble metals and superior high temperature stability. Vertically aligned nanocomposites (VANs) offer a great platform for combining two dissimilar functional materials with a one-step deposition technique toward multifunctionality integration and strong structural/property anisotropy. Here we report a two-phase nanocomposite design combining ferromagnetic CoFe2 nanosheets in the plasmonic TiN matrix as a new hybrid plasmonic metamaterial. The hybrid metamaterials exhibit obvious anisotropic optical and magnetic responses, as well as a pronounced magneto-optical coupling response evidenced by MOKE measurement, owing to the novel vertically aligned structure. This work demonstrates a new TiN-based metamaterial with anisotropic properties and multi-functionality towards optical switchable spintronics, magnetic sensors and integrated optics

Various distributed deep neural network (DNN) training technologies lead to increasingly complicated use of collective communications on GPU. The deadlock-prone collectives on GPU force researchers to guarantee that collectives are enqueued in a consistent order on each GPU to prevent deadlocks. In complex distributed DNN training scenarios, manual hardcoding is the only practical way for deadlock prevention, which poses significant challenges to the development of artificial intelligence. This paper presents OCCL, which is, to the best of our knowledge, the first deadlock-free collective communication library for GPU supporting dynamic decentralized preemption and gang-scheduling for collectives. Leveraging the preemption opportunity of collectives on GPU, OCCL dynamically preempts collectives in a decentralized way via the deadlock-free collective execution framework and allows dynamic decentralized gang-scheduling via the stickiness adjustment scheme. With the help of OCCL, researchers no longer have to struggle to get all GPUs to launch collectives in a consistent order to prevent deadlocks. We implement OCCL with several optimizations and integrate OCCL with a distributed deep learning framework OneFlow. Experimental results demonstrate that OCCL achieves comparable or better latency and bandwidth for collectives compared to NCCL, the state-of-the-art. When used in distributed DNN training, OCCL can improve the peak training throughput by up to 78% compared to statically sequenced NCCL, while introducing overheads of less than 6.5% across various distributed DNN training approaches.

As the size of deep learning models gets larger and larger, training takes longer time and more resources, making fault tolerance more and more critical. Existing state-of-the-art methods like CheckFreq and Elastic Horovod need to back up a copy of the model state (i.e., parameters and optimizer states) in memory, which is costly for large models and leads to non-trivial overhead. This paper presents SWIFT, a novel recovery design for distributed deep neural network training that significantly reduces the failure recovery overhead without affecting training throughput and model accuracy. Instead of making an additional copy of the model state, SWIFT resolves the inconsistencies of the model state caused by the failure and exploits the replicas of the model state in data parallelism for failure recovery. We propose a logging-based approach when replicas are unavailable, which records intermediate data and replays the computation to recover the lost state upon a failure. The re-computation is distributed across multiple machines to accelerate failure recovery further. We also log intermediate data selectively, exploring the trade-off between recovery time and intermediate data storage overhead. Evaluations show that SWIFT significantly reduces the failure recovery time and achieves similar or better training throughput during failure-free execution compared to state-of-the-art methods without degrading final model accuracy. SWIFT can also achieve up to 1.16x speedup in total training time compared to state-of-the-art methods.

Graph Neural Networks (GNNs) have shown superior performance for semi-supervised learning of numerous web applications, such as classification on web services and pages, analysis of online social networks, and recommendation in e-commerce. The state of the art derives representations for all nodes in graphs following the same diffusion (message passing) model without discriminating their uniqueness. However, (i) labeled nodes involved in model training usually account for a small portion of graphs in the semisupervised setting, and (ii) different nodes locate at different graph local contexts and it inevitably degrades the representation qualities if treating them undistinguishedly in diffusion. To address the above issues, we develop NDM, a universal node-wise diffusion model, to capture the unique characteristics of each node in diffusion, by which NDM is able to yield high-quality node representations. In what follows, we customize NDM for semisupervised learning and design the NIGCN model. In particular, NIGCN advances the efficiency significantly since it (i) produces representations for labeled nodes only and (ii) adopts well-designed neighbor sampling techniques tailored for node representation generation. Extensive experimental results on various types of web datasets, including citation, social and co-purchasing graphs, not only verify the state-of-the-art effectiveness of NIGCN but also strongly support the remarkable scalability of NIGCN. In particular, NIGCN completes representation generation and training within 10 seconds on the dataset with hundreds of millions of nodes and billions of edges, up to orders of magnitude speedups over the baselines, while achieving the highest F1-scores on classification.

Recently, implicit graph neural networks (GNNs) have been proposed to capture long-range dependencies in underlying graphs. In this paper, we introduce and justify two weaknesses of implicit GNNs: the constrained expressiveness due to their limited effective range for capturing long-range dependencies, and their lack of ability to capture multiscale information on graphs at multiple resolutions. To show the limited effective range of previous implicit GNNs, We first provide a theoretical analysis and point out the intrinsic relationship between the effective range and the convergence of iterative equations used in these models. To mitigate the mentioned weaknesses, we propose a multiscale graph neural network with implicit layers (MGNNI) which is able to model multiscale structures on graphs and has an expanded effective range for capturing long-range dependencies. We conduct comprehensive experiments for both node classification and graph classification to show that MGNNI outperforms representative baselines and has a better ability for multiscale modeling and capturing of long-range dependencies., Comment: NeurIPS 2022

Recent literature focuses on utilizing the entity information in the sentence-level relation extraction (RE), but this risks leaking superficial and spurious clues of relations. As a result, RE still suffers from unintended entity bias, i.e., the spurious correlation between entity mentions (names) and relations. Entity bias can mislead the RE models to extract the relations that do not exist in the text. To combat this issue, some previous work masks the entity mentions to prevent the RE models from overfitting entity mentions. However, this strategy degrades the RE performance because it loses the semantic information of entities. In this paper, we propose the CORE (Counterfactual Analysis based Relation Extraction) debiasing method that guides the RE models to focus on the main effects of textual context without losing the entity information. We first construct a causal graph for RE, which models the dependencies between variables in RE models. Then, we propose to conduct counterfactual analysis on our causal graph to distill and mitigate the entity bias, that captures the causal effects of specific entity mentions in each instance. Note that our CORE method is model-agnostic to debias existing RE systems during inference without changing their training processes. Extensive experimental results demonstrate that our CORE yields significant gains on both effectiveness and generalization for RE. The source code is provided at: https://github.com/vanoracai/CoRE., Comment: NAACL 2022

We study dangling-aware entity alignment in knowledge graphs (KGs), which is an underexplored but important problem. As different KGs are naturally constructed by different sets of entities, a KG commonly contains some dangling entities that cannot find counterparts in other KGs. Therefore, dangling-aware entity alignment is more realistic than the conventional entity alignment where prior studies simply ignore dangling entities. We propose a framework using mixed high-order proximities on dangling-aware entity alignment. Our framework utilizes both the local high-order proximity in a nearest neighbor subgraph and the global high-order proximity in an embedding space for both dangling detection and entity alignment. Extensive experiments with two evaluation settings shows that our framework more precisely detects dangling entities, and better aligns matchable entities. Further investigations demonstrate that our framework can mitigate the hubness problem on dangling-aware entity alignment., Comment: NAACL 2022 (Findings)

Graph neural networks (GNNs) are widely used for modelling graph-structured data in numerous applications. However, with their inherently finite aggregation layers, existing GNN models may not be able to effectively capture long-range dependencies in the underlying graphs. Motivated by this limitation, we propose a GNN model with infinite depth, which we call Efficient Infinite-Depth Graph Neural Networks (EIGNN), to efficiently capture very long-range dependencies. We theoretically derive a closed-form solution of EIGNN which makes training an infinite-depth GNN model tractable. We then further show that we can achieve more efficient computation for training EIGNN by using eigendecomposition. The empirical results of comprehensive experiments on synthetic and real-world datasets show that EIGNN has a better ability to capture long-range dependencies than recent baselines, and consistently achieves state-of-the-art performance. Furthermore, we show that our model is also more robust against both noise and adversarial perturbations on node features., Comment: NeurIPS 2021 Camera Ready

Deep learning frameworks such as TensorFlow and PyTorch provide a productive interface for expressing and training a deep neural network (DNN) model on a single device or using data parallelism. Still, they may not be flexible or efficient enough in training emerging large models on distributed devices, which require more sophisticated parallelism beyond data parallelism. Plugins or wrappers have been developed to strengthen these frameworks for model or pipeline parallelism, but they complicate the usage and implementation of distributed deep learning. Aiming at a simple, neat redesign of distributed deep learning frameworks for various parallelism paradigms, we present OneFlow, a novel distributed training framework based on an SBP (split, broadcast and partial-value) abstraction and the actor model. SBP enables much easier programming of data parallelism and model parallelism than existing frameworks, and the actor model provides a succinct runtime mechanism to manage the complex dependencies imposed by resource constraints, data movement and computation in distributed deep learning. We demonstrate the general applicability and efficiency of OneFlow for training various large DNN models with case studies and extensive experiments. The results show that OneFlow outperforms many well-known customized libraries built on top of the state-of-the-art frameworks. The code of OneFlow is available at: https://github.com/Oneflow-Inc/oneflow.

InstaHide is a state-of-the-art mechanism for protecting private training images, by mixing multiple private images and modifying them such that their visual features are indistinguishable to the naked eye. In recent work, however, Carlini et al. show that it is possible to reconstruct private images from the encrypted dataset generated by InstaHide. Nevertheless, we demonstrate that Carlini et al.'s attack can be easily defeated by incorporating data augmentation into InstaHide. This leads to a natural question: is InstaHide with data augmentation secure? In this paper, we provide a negative answer to this question, by devising an attack for recovering private images from the outputs of InstaHide even when data augmentation is present. The basic idea is to use a comparative network to identify encrypted images that are likely to correspond to the same private image, and then employ a fusion-denoising network for restoring the private image from the encrypted ones, taking into account the effects of data augmentation. Extensive experiments demonstrate the effectiveness of the proposed attack in comparison to Carlini et al.'s attack., Comment: 15 pages

We introduce a deep recursive octree network for the compression of 3D voxel data. Our network compresses a voxel grid of any size down to a very small latent space in an autoencoder-like network. We show results for compressing 32, 64 and 128 grids down to just 80 floats in the latent space. We demonstrate the effectiveness and efficiency of our proposed method on several publicly available datasets with three experiments: 3D shape classification, 3D shape reconstruction, and shape generation. Experimental results show that our algorithm maintains accuracy while consuming less memory with shorter training times compared to existing methods, especially in 3D reconstruction tasks.

Node classification on graphs is an important task in many practical domains. It usually requires labels for training, which can be difficult or expensive to obtain in practice. Given a budget for labelling, active learning aims to improve performance by carefully choosing which nodes to label. Previous graph active learning methods learn representations using labelled nodes and select some unlabelled nodes for label acquisition. However, they do not fully utilize the representation power present in unlabelled nodes. We argue that the representation power in unlabelled nodes can be useful for active learning and for further improving performance of active learning for node classification. In this paper, we propose a latent space clustering-based active learning framework for node classification (LSCALE), where we fully utilize the representation power in both labelled and unlabelled nodes. Specifically, to select nodes for labelling, our framework uses the K-Medoids clustering algorithm on a latent space based on a dynamic combination of both unsupervised features and supervised features. In addition, we design an incremental clustering module to avoid redundancy between nodes selected at different steps. Extensive experiments on five datasets show that our proposed framework LSCALE consistently and significantly outperforms the stateof-the-art approaches by a large margin., Comment: ECML-PKDD 2022

Given a graph G where each node is associated with a set of attributes, attributed network embedding (ANE) maps each node v in G to a compact vector Xv, which can be used in downstream machine learning tasks. Ideally, Xv should capture node v's affinity to each attribute, which considers not only v's own attribute associations, but also those of its connected nodes along edges in G. It is challenging to obtain high-utility embeddings that enable accurate predictions; scaling effective ANE computation to massive graphs pushes the difficulty of the problem to a whole new level. Existing solutions largely fail on such graphs, leading to prohibitive costs, low-quality embeddings, or both. This paper proposes PANE, an effective and scalable approach to ANE computation for massive graphs that achieves state-of-the-art result quality on multiple benchmark datasets. PANE obtains high scalability and effectiveness through 3 main algorithmic designs. First, it formulates the learning objective based on a novel random walk model for attributed networks. Second, PANE includes a highly efficient solver for the above optimization problem, whose key module is a carefully designed initialization of the embeddings, which drastically reduces the number of iterations required to converge. Finally, PANE utilizes multi-core CPUs through non-trivial parallelization of the above solver, which achieves scalability while retaining the high quality of the resulting embeddings. The performance of PANE depends upon the number of attributes in the input network. To handle large networks with numerous attributes, we further extend PANE to PANE++. Extensive experiments, comparing 10 existing approaches on 8 real datasets, demonstrate that PANE and PANE++ consistently outperform all existing methods in terms of result quality, while being orders of magnitude faster., Comment: Paper published in VLDBJ 2023. Extended version of PVLDB Vol. 14 (VLDB 2021) paper titled "Scaling Attributed Network Embedding to Massive Graphs"

Sketching is a natural and intuitive communication tool used for expressing concepts or ideas which are difficult to communicate through text or speech alone. Sketching is therefore used for a variety of purposes, from the expression of ideas on two-dimensional (2D) physical media, to object creation, manipulation, or deformation in three-dimensional (3D) immersive environments. This variety in sketching activities brings about a range of technologies which, while having similar scope, namely that of recording and interpreting the sketch gesture to effect some interaction, adopt different interpretation approaches according to the environment in which the sketch is drawn. In fields such as product design, sketches are drawn at various stages of the design process, and therefore, designers would benefit from sketch interpretation technologies which support these differing interactions. However, research typically focuses on one aspect of sketch interpretation and modeling such that literature on available technologies is fragmented and dispersed. In this paper, we bring together the relevant literature describing technologies which can support the product design industry, namely technologies which support the interpretation of sketches drawn on 2D media, sketch-based search interactions, as well as sketch gestures drawn in 3D media. This paper, therefore, gives a holistic view of the algorithmic support that can be provided in the design process. In so doing, we highlight the research gaps and future research directions required to provide full sketch-based interaction support., NA

This paper summarizes the state of discussions in CIGRE/CIRED joined working group C4.24, concerning expected impacts on the power quality of future methods for volt-var-control in the distribution grid. The positive impacts include the reduction of the number of overvoltage and undervoltages and also a reduction of voltage unbalance, when some control schemes are applied. Potential negative impacts include an increased number of short-duration undervoltages, rapid voltage changes, flicker, and voltage transients; a higher risk of harmonic resonances; and increased emission of supraharmonics. All these potential impacts are discussed in the paper., Godkänd; 2015; 20150625 (matbol)

This paper summarizes the state of discussions in CIGRE/CIRED joined working group C4.24, concerning expected impacts on the power quality of future methods for volt-var-control in the distribution grid. The positive impacts include the reduction of the number of overvoltage and undervoltages and also a reduction of voltage unbalance, when some control schemes are applied. Potential negative impacts include an increased number of short-duration undervoltages, rapid voltage changes, flicker, and voltage transients; a higher risk of harmonic resonances; and increased emission of supraharmonics. All these potential impacts are discussed in the paper., Godkänd; 2015; 20150625 (matbol)

Zhao X, Li Z, He B, et al. Onco Targets Ther. 2013;6:1751–1760.In the author affiliation section on the first page, the affiliations were shown incorrectly. Read the original article

Xiaohong Zhao,1,* Zuoqing Li,2,* Benfu He,3 Juncheng Liu,2 Suisheng Li,2 Li Zhou,2 Cuiling Pan,2 Zhe Yu,4 Zhe Xu21State Key Laboratory of Oncology in Southern China, Department of Experimental Research, Cancer Center, Sun Yat-sen University, 2Department of Pediatric Surgery, First Affiliated Hospital of Sun Yat-sen University, 3Oncology Department, PLA421 Hospital, 4Laura Biotech Co, Ltd, Guangzhou, People’s Republic of China*These authors contributed equally to this workBackground: Neuroblastoma (NB) is the most common solid extracranial tumor in children. However, the molecular mechanism and progression of NB is largely unknown, and unfortunately, the prognosis is poor. Src-associated in mitosis with a molecular weight of 68 kDa (Sam68) is associated with carcinogenesis and neurogenesis. The present study aimed to investigate the clinical and prognostic significance of Sam68 in NB.Methods: The expression of Sam68 in immortalized normal epithelial cells, NB cell lines, and in four cases of paired NB tissue and adjacent normal tissue from the same patient was examined using Western blotting, reverse transcription-polymerase chain reaction (PCR) and real-time reverse transcription-PCR. The proliferation of NB cells was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Furthermore, Sam68 protein expression was analyzed in 90 NB cases characterized as clinicopathological using immunohistochemistry. Statistical analyses were applied to evaluate the diagnostic value and associations of Sam68 with clinical parameters.Results: Western blotting and reverse transcription-PCR showed that the expression level of Sam68 was markedly higher in NB cell lines than in the immortalized normal epithelial cells at both messenger RNA and protein levels. The MTT assay revealed that Sam68 expression supported proliferation of NB cells. Sam68 expression levels were significantly up-regulated in tumor tissues in comparison to the matched adjac