arXiv subjects

Huan Zhao

Publications and source records attributed to Huan Zhao.

At least 109 records · Page 6Linked to original sources

arXiv2022

Manipulating Interlayer Excitons for Ultra-pure Near-infrared Quantum Light Generation

Interlayer excitons (IXs) formed at the interface of atomically-thin semiconductors possess various novel properties. In a parallel development, nanoscale strain engineering has emerged as an effective means for creating 2D quantum light sources. Exploring the intersection of these two exciting areas, where strain and defects are exploited for the manipulation of IX toward the emergence of new functionalities, is currently at a nascent stage. Here, using MoS2/WSe2 heterostructure as a model system, we demonstrate how strain, defects, and layering can be utilized to create defect-bound IXs capable of bright, robust, and tunable quantum light emission in the technologically important near-infrared spectral range. We were able to achieve ultra-high single-photon purity with g(2)(0) = 0.01. Our strategy of creating site-controlled QEs from the defect-bound IXs represents a paradigm shift in 2D quantum photonics research, from engineering intralayer exciton in monolayer structures towards IXs at the interface of 2D heterostructures.

cond-mat.mes-hall↗

arXiv2022

Bridging the Gap of AutoGraph between Academia and Industry: Analysing AutoGraph Challenge at KDD Cup 2020

Graph structured data is ubiquitous in daily life and scientific areas and has attracted increasing attention. Graph Neural Networks (GNNs) have been proved to be effective in modeling graph structured data and many variants of GNN architectures have been proposed. However, much human effort is often needed to tune the architecture depending on different datasets. Researchers naturally adopt Automated Machine Learning on Graph Learning, aiming to reduce the human effort and achieve generally top-performing GNNs, but their methods focus more on the architecture search. To understand GNN practitioners' automated solutions, we organized AutoGraph Challenge at KDD Cup 2020, emphasizing on automated graph neural networks for node classification. We received top solutions especially from industrial tech companies like Meituan, Alibaba and Twitter, which are already open sourced on Github. After detailed comparisons with solutions from academia, we quantify the gaps between academia and industry on modeling scope, effectiveness and efficiency, and show that (1) academia AutoML for Graph solutions focus on GNN architecture search while industrial solutions, especially the winning ones in the KDD Cup, tend to obtain an overall solution (2) by neural architecture search only, academia solutions achieve on average 97.3% accuracy of industrial solutions (3) academia solutions are cheap to obtain with several GPU hours while industrial solutions take a few months' labors. Academic solutions also contain much fewer parameters.

cs.LG↗

arXiv2022

Improving Event Representation via Simultaneous Weakly Supervised Contrastive Learning and Clustering

Representations of events described in text are important for various tasks. In this work, we present SWCC: a Simultaneous Weakly supervised Contrastive learning and Clustering framework for event representation learning. SWCC learns event representations by making better use of co-occurrence information of events. Specifically, we introduce a weakly supervised contrastive learning method that allows us to consider multiple positives and multiple negatives, and a prototype-based clustering method that avoids semantically related events being pulled apart. For model training, SWCC learns representations by simultaneously performing weakly supervised contrastive learning and prototype-based clustering. Experimental results show that SWCC outperforms other baselines on Hard Similarity and Transitive Sentence Similarity tasks. In addition, a thorough analysis of the prototype-based clustering method demonstrates that the learned prototype vectors are able to implicitly capture various relations between events.

cs.CL↗

arXiv2022

Revisiting stress-corrosion cracking and hydrogen embrittlement in 7xxx-Al alloys at the near-atomic-scale

Hydrogen embrittlement (HE) affects all major high-strength structural materials and as such is a major impediment to lightweighting e.g. vehicles and help reduce carbon-emissions and reach net-zero. The high-strength 7xxx series aluminium alloys can fulfil the need for light, high strength materials, and are already extensively used in aerospace for weight reduction purposes. However, depending on the thermomechanical and loading state, these alloys can be sensitive to stress-corrosion cracking (SCC) through anodic dissolution and hydrogen embrittlement. Here, we study at the near-atomic-scale the intra- and inter-granular microstructure ahead and in the wake of a propagating SCC crack. Moving away from model cases not strictly relevant to application, we performed an industry-standard test on an engineering Al-7XXX alloy. H is found segregated to planar arrays of dislocations and to grain boundaries that we can associate to the combined effects of hydrogen-enhanced localized plasticity (HELP) and hydrogen-enhanced decohesion (HEDE) mechanisms. We report on a Mg-rich H-rich amorphous oxide on the corroded crack surface and evidence of Mg-related diffusional processes leading to dissolution of the strengthening eta-phase precipitates ahead of the crack. We show ingress of up to 1 at% O, i.e. well above the solubility limit of O in Al, near the oxide-metal interface, while no increased level of H is found in the matrix. We provide an array of discussion points relative to the interplay of structural defects, transport of solutes, thereby changing the resistance against crack propagation, which have been overlooked across the SCC literature and prevent accurate service life predictions.

cond-mat.mtrl-sci↗

arXiv2022

Proximity Induced Chiral Quantum Light Generation in Strain-Engineered WSe2/NiPS3 Heterostructures

Quantum light emitters (QEs) capable of generating single photons of well-defined circular polarization could enable non-reciprocal single photon devices and deterministic spin-photon interfaces critical for realizing complex quantum networks. To date, emission of such chiral quantum light has been achieved via the application of intense external magnetic field electrical/optical injection of spin polarized carriers/excitons, or coupling with complex photonic/meta-structures. Here we report free-space generation of highly chiral single photons from QEs created in monolayer WSe2 - NiPS3 heterostructures at zero external magnetic field. These QEs emit in the 760-800 nm range with a degree of circular polarization and single photon purity as high as 0.71 and 80% respectively, independent of pump laser polarization. QEs are deterministically created by pressing a scanning probe microscope tip into a two-dimensional heterostructure comprising a WSe2 monolayer and a ~50 nm thick layer of the antiferromagnetic (AFM) insulator NiPS3. Temperature dependent magneto-photoluminescence studies indicate that the chiral quantum light emission arises from magnetic proximity interactions between localized excitons in the WSe2 monolayer and the out-of-plane magnetization of AFM defects in NiPS3, both of which are co-localized by the strain field arising from the nanoscale indentations.

cond-mat.mes-hall↗

arXiv2022

Codabench: Flexible, Easy-to-Use and Reproducible Benchmarking Platform

Obtaining standardized crowdsourced benchmark of computational methods is a major issue in data science communities. Dedicated frameworks enabling fair benchmarking in a unified environment are yet to be developed. Here we introduce Codabench, an open-source, community-driven platform for benchmarking algorithms or software agents versus datasets or tasks. A public instance of Codabench (https://www.codabench.org/) is open to everyone, free of charge, and allows benchmark organizers to compare fairly submissions, under the same setting (software, hardware, data, algorithms), with custom protocols and data formats. Codabench has unique features facilitating the organization of benchmarks flexibly, easily and reproducibly, such as the possibility of re-using templates of benchmarks, and supplying compute resources on-demand. Codabench has been used internally and externally on various applications, receiving more than 130 users and 2500 submissions. As illustrative use cases, we introduce 4 diverse benchmarks covering Graph Machine Learning, Cancer Heterogeneity, Clinical Diagnosis and Reinforcement Learning.

cs.LG↗

arXiv2022

Designing the Topology of Graph Neural Networks: A Novel Feature Fusion Perspective

In recent years, Graph Neural Networks (GNNs) have shown superior performance on diverse real-world applications. To improve the model capacity, besides designing aggregation operations, GNN topology design is also very important. In general, there are two mainstream GNN topology design manners. The first one is to stack aggregation operations to obtain the higher-level features but easily got performance drop as the network goes deeper. Secondly, the multiple aggregation operations are utilized in each layer which provides adequate and independent feature extraction stage on local neighbors while are costly to obtain the higher-level information. To enjoy the benefits while alleviating the corresponding deficiencies of these two manners, we learn to design the topology of GNNs in a novel feature fusion perspective which is dubbed F$^2$GNN. To be specific, we provide a feature fusion perspective in designing GNN topology and propose a novel framework to unify the existing topology designs with feature selection and fusion strategies. Then we develop a neural architecture search method on top of the unified framework which contains a set of selection and fusion operations in the search space and an improved differentiable search algorithm. The performance gains on eight real-world datasets demonstrate the effectiveness of F$^2$GNN. We further conduct experiments to show that F$^2$GNN can improve the model capacity while alleviating the deficiencies of existing GNN topology design manners, especially alleviating the over-smoothing problem, by utilizing different levels of features adaptively.

cs.LG↗

arXiv2022

Hydrogen trapping and embrittlement in high-strength Al-alloys

Ever more stringent regulations on greenhouse gas emissions from transportation motivate efforts to revisit materials used for vehicles. High-strength Al-alloys often used in aircrafts could help reduce the weight of automobiles, but are susceptible to environmental degradation. Hydrogen (H) "embrittlement" is often pointed as the main culprit, however, the mechanisms underpinning failure are elusive: atomic-scale analysis of H inside an alloy remains a challenge, and this prevents deploying alloy design strategies to enhance the materials' durability. Here we successfully performed near-atomic scale analysis of H trapped in second-phase particles and at grain boundaries in a high-strength 7xxx Al-alloy. We used these observations to guide atomistic ab-initio calculations which show that the co-segregation of alloying elements and H favours grain boundary decohesion, while the strong partitioning of H into the second-phases removes solute H from the matrix, hence preventing H-embrittlement. Our insights further advance the mechanistic understanding of H-assisted embrittlement in Al-alloys, emphasizing the role of H-traps in retarding cracking and guiding new alloy design.

cond-mat.mtrl-sci↗

arXiv2022

Graph Neural Networks for Double-Strand DNA Breaks Prediction

Double-strand DNA breaks (DSBs) are a form of DNA damage that can cause abnormal chromosomal rearrangements. Recent technologies based on high-throughput experiments have obvious high costs and technical challenges.Therefore, we design a graph neural network based method to predict DSBs (GraphDSB), using DNA sequence features and chromosome structure information. In order to improve the expression ability of the model, we introduce Jumping Knowledge architecture and several effective structural encoding methods. The contribution of structural information to the prediction of DSBs is verified by the experiments on datasets from normal human epidermal keratinocytes (NHEK) and chronic myeloid leukemia cell line (K562), and the ablation studies further demonstrate the effectiveness of the designed components in the proposed GraphDSB framework. Finally, we use GNNExplainer to analyze the contribution of node features and topology to DSBs prediction, and proved the high contribution of 5-mer DNA sequence features and two chromatin interaction modes.

q-bio.QM↗

arXiv2021

Learn Layer-wise Connections in Graph Neural Networks

In recent years, Graph Neural Networks (GNNs) have shown superior performance on diverse applications on real-world datasets. To improve the model capacity and alleviate the over-smoothing problem, several methods proposed to incorporate the intermediate layers by layer-wise connections. However, due to the highly diverse graph types, the performance of existing methods vary on diverse graphs, leading to a need for data-specific layer-wise connection methods. To address this problem, we propose a novel framework LLC (Learn Layer-wise Connections) based on neural architecture search (NAS) to learn adaptive connections among intermediate layers in GNNs. LLC contains one novel search space which consists of 3 types of blocks and learnable connections, and one differentiable search algorithm to enable the efficient search process. Extensive experiments on five real-world datasets are conducted, and the results show that the searched layer-wise connections can not only improve the performance but also alleviate the over-smoothing problem.

cs.LG↗

arXiv2021

Beijing ZKJ-NPU Speaker Verification System for VoxCeleb Speaker Recognition Challenge 2021

In this report, we describe the Beijing ZKJ-NPU team submission to the VoxCeleb Speaker Recognition Challenge 2021 (VoxSRC-21). We participated in the fully supervised speaker verification track 1 and track 2. In the challenge, we explored various kinds of advanced neural network structures with different pooling layers and objective loss functions. In addition, we introduced the ResNet-DTCF, CoAtNet and PyConv networks to advance the performance of CNN-based speaker embedding model. Moreover, we applied embedding normalization and score normalization at the evaluation stage. By fusing 11 and 14 systems, our final best performances (minDCF/EER) on the evaluation trails are 0.1205/2.8160% and 0.1175/2.8400% respectively for track 1 and 2. With our submission, we came to the second place in the challenge for both tracks.

cs.SD↗

arXiv2021

Nondestructive Testing of Composite Fibre Materials with Hyperspectral Imaging : Evaluative Studies in the EU H2020 FibreEUse Project

Through capturing spectral data from a wide frequency range along with the spatial information, hyperspectral imaging (HSI) can detect minor differences in terms of temperature, moisture and chemical composition. Therefore, HSI has been successfully applied in various applications, including remote sensing for security and defense, precision agriculture for vegetation and crop monitoring, food/drink, and pharmaceuticals quality control. However, for condition monitoring and damage detection in carbon fibre reinforced polymer (CFRP), the use of HSI is a relatively untouched area, as existing non-destructive testing (NDT) techniques focus mainly on delivering information about physical integrity of structures but not on material composition. To this end, HSI can provide a unique way to tackle this challenge. In this paper, with the use of a near-infrared HSI camera, applications of HSI for the non-destructive inspection of CFRP products are introduced, taking the EU H2020 FibreEUse project as the background. Technical challenges and solutions on three case studies are presented in detail, including adhesive residues detection, surface damage detection and Cobot based automated inspection. Experimental results have fully demonstrated the great potential of HSI and related vision techniques for NDT of CFRP, especially the potential to satisfy the industrial manufacturing environment.

cs.CV↗

arXiv2021

Second-order Cumulants Ghost Imaging

In the conventional ghost imaging (GI), the image is retrieved by correlating the reference intensity fluctuation at a charge-coupled device (CCD) with the signal intensity fluctuation at a bucket detector. In this letter, we present the protocol of GI, it is called Second-order Cumulants ghost imaging (SCGI). The image is retrieved by the fluctuation information of correlating intensity fluctuation at two detectors, and the resolution limit can be enhanced than conventional GI. The experimental results of SCGI agreement with theoretical results.

quant-ph↗

arXiv2021

DiffMG: Differentiable Meta Graph Search for Heterogeneous Graph Neural Networks

In this paper, we propose a novel framework to automatically utilize task-dependent semantic information which is encoded in heterogeneous information networks (HINs). Specifically, we search for a meta graph, which can capture more complex semantic relations than a meta path, to determine how graph neural networks (GNNs) propagate messages along different types of edges. We formalize the problem within the framework of neural architecture search (NAS) and then perform the search in a differentiable manner. We design an expressive search space in the form of a directed acyclic graph (DAG) to represent candidate meta graphs for a HIN, and we propose task-dependent type constraint to filter out those edge types along which message passing has no effect on the representations of nodes that are related to the downstream task. The size of the search space we define is huge, so we further propose a novel and efficient search algorithm to make the total search cost on a par with training a single GNN once. Compared with existing popular NAS algorithms, our proposed search algorithm improves the search efficiency. We conduct extensive experiments on different HINs and downstream tasks to evaluate our method, and experimental results show that our method can outperform state-of-the-art heterogeneous GNNs and also improves efficiency compared with those methods which can implicitly learn meta paths.

cs.LG↗

arXiv2021

Multi-Relational Graph based Heterogeneous Multi-Task Learning in Community Question Answering

Various data mining tasks have been proposed to study Community Question Answering (CQA) platforms like Stack Overflow. The relatedness between some of these tasks provides useful learning signals to each other via Multi-Task Learning (MTL). However, due to the high heterogeneity of these tasks, few existing works manage to jointly solve them in a unified framework. To tackle this challenge, we develop a multi-relational graph based MTL model called Heterogeneous Multi-Task Graph Isomorphism Network (HMTGIN) which efficiently solves heterogeneous CQA tasks. In each training forward pass, HMTGIN embeds the input CQA forum graph by an extension of Graph Isomorphism Network and skip connections. The embeddings are then shared across all task-specific output layers to compute respective losses. Moreover, two cross-task constraints based on the domain knowledge about tasks' relationships are used to regularize the joint learning. In the evaluation, the embeddings are shared among different task-specific output layers to make corresponding predictions. To the best of our knowledge, HMTGIN is the first MTL model capable of tackling CQA tasks from the aspect of multi-relational graphs. To evaluate HMTGIN's effectiveness, we build a novel large-scale multi-relational graph CQA dataset with over two million nodes from Stack Overflow. Extensive experiments show that: $(1)$ HMTGIN is superior to all baselines on five tasks; $(2)$ The proposed MTL strategy and cross-task constraints have substantial advantages.

cs.CL↗

arXiv2021

Neural PathSim for Inductive Similarity Search in Heterogeneous Information Networks

PathSim is a widely used meta-path-based similarity in heterogeneous information networks. Numerous applications rely on the computation of PathSim, including similarity search and clustering. Computing PathSim scores on large graphs is computationally challenging due to its high time and storage complexity. In this paper, we propose to transform the problem of approximating the ground truth PathSim scores into a learning problem. We design an encoder-decoder based framework, NeuPath, where the algorithmic structure of PathSim is considered. Specifically, the encoder module identifies Top T optimized path instances, which can approximate the ground truth PathSim, and maps each path instance to an embedding vector. The decoder transforms each embedding vector into a scalar respectively, which identifies the similarity score. We perform extensive experiments on two real-world datasets in different domains, ACM and IMDB. Our results demonstrate that NeuPath performs better than state-of-the-art baselines in the PathSim approximation task and similarity search task.

cs.SI↗

arXiv2021

Pooling Architecture Search for Graph Classification

Graph classification is an important problem with applications across many domains, like chemistry and bioinformatics, for which graph neural networks (GNNs) have been state-of-the-art (SOTA) methods. GNNs are designed to learn node-level representation based on neighborhood aggregation schemes, and to obtain graph-level representation, pooling methods are applied after the aggregation operation in existing GNN models to generate coarse-grained graphs. However,due to highly diverse applications of graph classification, and the performance of existing pooling methods vary on different graphs. In other words, it is a challenging problem to design a universal pooling architecture to perform well in most cases, leading to a demand for data-specific pooling methods in real-world applications. To address this problem, we propose to use neural architecture search (NAS) to search for adaptive pooling architectures for graph classification. Firstly we designed a unified framework consisting of four modules: Aggregation, Pooling, Readout, and Merge, which can cover existing human-designed pooling methods for graph classification. Based on this framework, a novel search space is designed by incorporating popular operations in human-designed architectures. Then to enable efficient search, a coarsening strategy is proposed to continuously relax the search space, thus a differentiable search method can be adopted. Extensive experiments on six real-world datasets from three domains are conducted, and the results demonstrate the effectiveness and efficiency of the proposed framework.

cs.LG↗

arXiv2021

TabGNN: Multiplex Graph Neural Network for Tabular Data Prediction

Tabular data prediction (TDP) is one of the most popular industrial applications, and various methods have been designed to improve the prediction performance. However, existing works mainly focus on feature interactions and ignore sample relations, e.g., users with the same education level might have a similar ability to repay the debt. In this work, by explicitly and systematically modeling sample relations, we propose a novel framework TabGNN based on recently popular graph neural networks (GNN). Specifically, we firstly construct a multiplex graph to model the multifaceted sample relations, and then design a multiplex graph neural network to learn enhanced representation for each sample. To integrate TabGNN with the tabular solution in our company, we concatenate the learned embeddings and the original ones, which are then fed to prediction models inside the solution. Experiments on eleven TDP datasets from various domains, including classification and regression ones, show that TabGNN can consistently improve the performance compared to the tabular solution AutoFE in 4Paradigm.

cs.LG↗