arXiv subjects

Jing Zhang

Publications and source records attributed to Jing Zhang.

At least 775 records · Page 43Linked to original sources

arXiv2021

Hierarchically Modeling Micro and Macro Behaviors via Multi-Task Learning for Conversion Rate Prediction

Conversion Rate (\emph{CVR}) prediction in modern industrial e-commerce platforms is becoming increasingly important, which directly contributes to the final revenue. In order to address the well-known sample selection bias (\emph{SSB}) and data sparsity (\emph{DS}) issues encountered during CVR modeling, the abundant labeled macro behaviors ($i.e.$, user's interactions with items) are used. Nonetheless, we observe that several purchase-related micro behaviors ($i.e.$, user's interactions with specific components on the item detail page) can supplement fine-grained cues for \emph{CVR} prediction. Motivated by this observation, we propose a novel \emph{CVR} prediction method by Hierarchically Modeling both Micro and Macro behaviors ($HM^3$). Specifically, we first construct a complete user sequential behavior graph to hierarchically represent micro behaviors and macro behaviors as one-hop and two-hop post-click nodes. Then, we embody $HM^3$ as a multi-head deep neural network, which predicts six probability variables corresponding to explicit sub-paths in the graph. They are further combined into the prediction targets of four auxiliary tasks as well as the final $CVR$ according to the conditional probability rule defined on the graph. By employing multi-task learning and leveraging the abundant supervisory labels from micro and macro behaviors, $HM^3$ can be trained end-to-end and address the \emph{SSB} and \emph{DS} issues. Extensive experiments on both offline and online settings demonstrate the superiority of the proposed $HM^3$ over representative state-of-the-art methods.

cs.LG↗

arXiv2021

Production of dual species Bose-Einstein condensates of $^{39}$K and $^{87}$Rb

We report the production of $^{39}$K and $^{87}$Rb Bose-Einstein condensates (BECs) in the lowest hyperfine states $| F=1,m_{F}=1 \rangle$ simultaneously. We collect atoms in bright/dark magneto-optical traps (MOTs) of $^{39}$K/$^{87}$Rb to overcome the light-assisted losses of $^{39}$K atoms. Gray molasses cooling on the D1 line of the $^{39}$K is used to effectively increase the phase density, which improves the loading efficiency of $^{39}$K into the quadrupole magnetic trap. Simultaneously, the normal molasses are employed for $^{87}$Rb. After the microwave evaporation cooling on $^{87}$Rb in the optically plugged magnetic trap, the atoms mixture is transferred to a crossed optical dipole trap, where the collisional properties of the two species in different combinations of the hyperfine states are studied. The dual species BECs of $^{39}$K and $^{87}$Rb are obtained by further evaporative cooling in optical dipole trap at a magnetic field of 372.6 G with the background repulsive interspecies scattering length $a_{KRb}$ = 34 $a_{0}$ ($a_{0}$ is the Bohr radius) and the intraspecies scattering length $a_{K}$ = 20.05 $a_{0}$.

cond-mat.quant-gas↗

arXiv2021

Learning structure-aware semantic segmentation with image-level supervision

Compared with expensive pixel-wise annotations, image-level labels make it possible to learn semantic segmentation in a weakly-supervised manner. Within this pipeline, the class activation map (CAM) is obtained and further processed to serve as a pseudo label to train the semantic segmentation model in a fully-supervised manner. In this paper, we argue that the lost structure information in CAM limits its application in downstream semantic segmentation, leading to deteriorated predictions. Furthermore, the inconsistent class activation scores inside the same object contradicts the common sense that each region of the same object should belong to the same semantic category. To produce sharp prediction with structure information, we introduce an auxiliary semantic boundary detection module, which penalizes the deteriorated predictions. Furthermore, we adopt smoothness loss to encourage prediction inside the object to be consistent. Experimental results on the PASCAL-VOC dataset illustrate the effectiveness of the proposed solution.

cs.CV↗

arXiv2021

Simultaneously Localize, Segment and Rank the Camouflaged Objects

Camouflage is a key defence mechanism across species that is critical to survival. Common strategies for camouflage include background matching, imitating the color and pattern of the environment, and disruptive coloration, disguising body outlines [35]. Camouflaged object detection (COD) aims to segment camouflaged objects hiding in their surroundings. Existing COD models are built upon binary ground truth to segment the camouflaged objects without illustrating the level of camouflage. In this paper, we revisit this task and argue that explicitly modeling the conspicuousness of camouflaged objects against their particular backgrounds can not only lead to a better understanding about camouflage and evolution of animals, but also provide guidance to design more sophisticated camouflage techniques. Furthermore, we observe that it is some specific parts of the camouflaged objects that make them detectable by predators. With the above understanding about camouflaged objects, we present the first ranking based COD network (Rank-Net) to simultaneously localize, segment and rank camouflaged objects. The localization model is proposed to find the discriminative regions that make the camouflaged object obvious. The segmentation model segments the full scope of the camouflaged objects. And, the ranking model infers the detectability of different camouflaged objects. Moreover, we contribute a large COD testing set to evaluate the generalization ability of COD models. Experimental results show that our model achieves new state-of-the-art, leading to a more interpretable COD network.

cs.CV↗

arXiv2021

Bootstrapping Your Own Positive Sample: Contrastive Learning With Electronic Health Record Data

Electronic Health Record (EHR) data has been of tremendous utility in Artificial Intelligence (AI) for healthcare such as predicting future clinical events. These tasks, however, often come with many challenges when using classical machine learning models due to a myriad of factors including class imbalance and data heterogeneity (i.e., the complex intra-class variances). To address some of these research gaps, this paper leverages the exciting contrastive learning framework and proposes a novel contrastive regularized clinical classification model. The contrastive loss is found to substantially augment EHR-based prediction: it effectively characterizes the similar/dissimilar patterns (by its "push-and-pull" form), meanwhile mitigating the highly skewed class distribution by learning more balanced feature spaces (as also echoed by recent findings). In particular, when naively exporting the contrastive learning to the EHR data, one hurdle is in generating positive samples, since EHR data is not as amendable to data augmentation as image data. To this end, we have introduced two unique positive sampling strategies specifically tailored for EHR data: a feature-based positive sampling that exploits the feature space neighborhood structure to reinforce the feature learning; and an attribute-based positive sampling that incorporates pre-generated patient similarity metrics to define the sample proximity. Both sampling approaches are designed with an awareness of unique high intra-class variance in EHR data. Our overall framework yields highly competitive experimental results in predicting the mortality risk on real-world COVID-19 EHR data with a total of 5,712 patients admitted to a large, urban health system. Specifically, our method reaches a high AUROC prediction score of 0.959, which outperforms other baselines and alternatives: cross-entropy(0.873) and focal loss(0.931).

cs.LG↗

arXiv2021

Weakly Supervised Video Salient Object Detection

Significant performance improvement has been achieved for fully-supervised video salient object detection with the pixel-wise labeled training datasets, which are time-consuming and expensive to obtain. To relieve the burden of data annotation, we present the first weakly supervised video salient object detection model based on relabeled "fixation guided scribble annotations". Specifically, an "Appearance-motion fusion module" and bidirectional ConvLSTM based framework are proposed to achieve effective multi-modal learning and long-term temporal context modeling based on our new weak annotations. Further, we design a novel foreground-background similarity loss to further explore the labeling similarity across frames. A weak annotation boosting strategy is also introduced to boost our model performance with a new pseudo-label generation technique. Extensive experimental results on six benchmark video saliency detection datasets illustrate the effectiveness of our solution.

cs.CV↗

arXiv2021

Uncertainty-aware Joint Salient Object and Camouflaged Object Detection

Visual salient object detection (SOD) aims at finding the salient object(s) that attract human attention, while camouflaged object detection (COD) on the contrary intends to discover the camouflaged object(s) that hidden in the surrounding. In this paper, we propose a paradigm of leveraging the contradictory information to enhance the detection ability of both salient object detection and camouflaged object detection. We start by exploiting the easy positive samples in the COD dataset to serve as hard positive samples in the SOD task to improve the robustness of the SOD model. Then, we introduce a similarity measure module to explicitly model the contradicting attributes of these two tasks. Furthermore, considering the uncertainty of labeling in both tasks' datasets, we propose an adversarial learning network to achieve both higher order similarity measure and network confidence estimation. Experimental results on benchmark datasets demonstrate that our solution leads to state-of-the-art (SOTA) performance for both tasks.

cs.CV↗

arXiv2021

Neural, Symbolic and Neural-Symbolic Reasoning on Knowledge Graphs

Knowledge graph reasoning is the fundamental component to support machine learning applications such as information extraction, information retrieval, and recommendation. Since knowledge graphs can be viewed as the discrete symbolic representations of knowledge, reasoning on knowledge graphs can naturally leverage the symbolic techniques. However, symbolic reasoning is intolerant of the ambiguous and noisy data. On the contrary, the recent advances of deep learning promote neural reasoning on knowledge graphs, which is robust to the ambiguous and noisy data, but lacks interpretability compared to symbolic reasoning. Considering the advantages and disadvantages of both methodologies, recent efforts have been made on combining the two reasoning methods. In this survey, we take a thorough look at the development of the symbolic, neural and hybrid reasoning on knowledge graphs. We survey two specific reasoning tasks, knowledge graph completion and question answering on knowledge graphs, and explain them in a unified reasoning framework. We also briefly discuss the future directions for knowledge graph reasoning.

cs.AI↗

arXiv2021

VDM-DA: Virtual Domain Modeling for Source Data-free Domain Adaptation

Domain adaptation aims to leverage a label-rich domain (the source domain) to help model learning in a label-scarce domain (the target domain). Most domain adaptation methods require the co-existence of source and target domain samples to reduce the distribution mismatch, however, access to the source domain samples may not always be feasible in the real world applications due to different problems (e.g., storage, transmission, and privacy issues). In this work, we deal with the source data-free unsupervised domain adaptation problem, and propose a novel approach referred to as Virtual Domain Modeling (VDM-DA). The virtual domain acts as a bridge between the source and target domains. On one hand, we generate virtual domain samples based on an approximated Gaussian Mixture Model (GMM) in the feature space with the pre-trained source model, such that the virtual domain maintains a similar distribution with the source domain without accessing to the original source data. On the other hand, we also design an effective distribution alignment method to reduce the distribution divergence between the virtual domain and the target domain by gradually improving the compactness of the target domain distribution through model learning. In this way, we successfully achieve the goal of distribution alignment between the source and target domains by training deep networks without accessing to the source domain data. We conduct extensive experiments on benchmark datasets for both 2D image-based and 3D point cloud-based cross-domain object recognition tasks, where the proposed method referred to Domain Adaptation with Virtual Domain Modeling (VDM-DA) achieves the state-of-the-art performances on all datasets.

cs.CV↗

arXiv2021

Self-supervised Learning: Generative or Contrastive

Deep supervised learning has achieved great success in the last decade. However, its deficiencies of dependence on manual labels and vulnerability to attacks have driven people to explore a better solution. As an alternative, self-supervised learning attracts many researchers for its soaring performance on representation learning in the last several years. Self-supervised representation learning leverages input data itself as supervision and benefits almost all types of downstream tasks. In this survey, we take a look into new self-supervised learning methods for representation in computer vision, natural language processing, and graph learning. We comprehensively review the existing empirical methods and summarize them into three main categories according to their objectives: generative, contrastive, and generative-contrastive (adversarial). We further investigate related theoretical analysis work to provide deeper thoughts on how self-supervised learning works. Finally, we briefly discuss open problems and future directions for self-supervised learning. An outline slide for the survey is provided.

cs.LG↗

arXiv2021

Understanding WeChat User Preferences and "Wow" Diffusion

WeChat is the largest social instant messaging platform in China, with 1.1 billion monthly active users. "Top Stories" is a novel friend-enhanced recommendation engine in WeChat, in which users can read articles based on preferences of both their own and their friends. Specifically, when a user reads an article by opening it, the "click" behavior is private. Moreover, if the user clicks the "wow" button, (only) her/his direct connections will be aware of this action/preference. Based on the unique WeChat data, we aim to understand user preferences and "wow" diffusion in Top Stories at different levels. We have made some interesting discoveries. For instance, the "wow" probability of one user is negatively correlated with the number of connected components that are formed by her/his active friends, but the click probability is the opposite. We further study to what extent users' "wow" and click behavior can be predicted from their social connections. To address this problem, we present a hierarchical graph representation learning based model DiffuseGNN, which is capable of capturing the structure-based social observations discovered above. Our experiments show that the proposed method can significantly improve the prediction performance compared with alternative methods.

cs.SI↗

arXiv2021

Koopmans' theorem as the mechanism of nearly gapless surface states in self-doped magnetic topological insulators

The magnetization-induced gap at the surface state is widely believed as the kernel of magnetic topological insulators (MTIs) because of its relevance to various topological phenomena, such as the quantum anomalous Hall effect and the axion insulator phase. However, if the magnetic gap exists in an intrinsic MTI, such as MnBi$_{2}$Te$_{4}$, still remains elusive, with significant discrepancies between theoretical predictions and various experimental observations. Here, including the previously overlooked self-doping in real MTIs, we find that in general a doped MTI prefers a ground state with a gapless surface state. We use a simple model based on Koopmans' theorem to elucidate the mechanism and further demonstrated it in self-doped MnBi$_{2}$Te$_{4}$/(Bi$_{2}$Te$_{3}$)$_{n}$ family through first-principles calculations. Our work shed light on the design principles of MTIs with magnetic gaps by revealing the critical role of doping effects in understanding the delicate interplay between magnetism and topology.

cond-mat.mtrl-sci↗

arXiv2021

LineaRE: Simple but Powerful Knowledge Graph Embedding for Link Prediction

The task of link prediction for knowledge graphs is to predict missing relationships between entities. Knowledge graph embedding, which aims to represent entities and relations of a knowledge graph as low dimensional vectors in a continuous vector space, has achieved promising predictive performance. If an embedding model can cover different types of connectivity patterns and mapping properties of relations as many as possible, it will potentially bring more benefits for link prediction tasks. In this paper, we propose a novel embedding model, namely LineaRE, which is capable of modeling four connectivity patterns (i.e., symmetry, antisymmetry, inversion, and composition) and four mapping properties (i.e., one-to-one, one-to-many, many-to-one, and many-to-many) of relations. Specifically, we regard knowledge graph embedding as a simple linear regression task, where a relation is modeled as a linear function of two low-dimensional vector-presented entities with two weight vectors and a bias vector. Since the vectors are defined in a real number space and the scoring function of the model is linear, our model is simple and scalable to large knowledge graphs. Experimental results on multiple widely used real-world datasets show that the proposed LineaRE model significantly outperforms existing state-of-the-art models for link prediction tasks.

cs.AI↗

arXiv2021

Multi-Stage Transmission Line Flow Control Using Centralized and Decentralized Reinforcement Learning Agents

Planning future operational scenarios of bulk power systems that meet security and economic constraints typically requires intensive labor efforts in performing massive simulations. To automate this process and relieve engineers' burden, a novel multi-stage control approach is presented in this paper to train centralized and decentralized reinforcement learning agents that can automatically adjust grid controllers for regulating transmission line flows at normal condition and under contingencies. The power grid flow control problem is formulated as Markov Decision Process (MDP). At stage one, centralized soft actor-critic (SAC) agent is trained to control generator active power outputs in a wide area to control transmission line flows against specified security limits. If line overloading issues remain unresolved, stage two is used to train decentralized SAC agent via load throw-over at local substations. The effectiveness of the proposed approach is verified on a series of actual planning cases used for operating the power grid of SGCC Zhejiang Electric Power Company.

cs.LG↗

arXiv2021

Atomic-scale investigation of the irradiation-resistant effect of symmetric tilt grain boundaries of Fe-Ni-Cr alloy

In this paper, the Fe-20Ni-25Cr alloy that is used for fuel cladding or pressure vessels with various grain boundaries (GBs) was investigated by employing molecular dynamics simulations. The bi-crystals comprised of Σ3(111), Σ3(112), Σ9(114), Σ11(113), Σ19(116), and Σ17(223) types GBs were considered to systematically examine the interplay between irradiation defects, irradiation microstructure evolution under stress, and irradiation mechanical properties with irradiation intensity, coincidence site lattice parameter, tilt angle, and GB thickness. It is found that irradiated vacancies and interstitials are annihilated by competitive GB absorption and recombination. Bias absorption of interstitials is observed for most bi-crystals except Σ3(111) and Σ11(113) at 15 keV incident energy, and results in abundant residual vacancies clusters in grain interior. In addition, different GBs exhibit quite diverse irradiation defect sink ability, and the number of residual vacancies is inversely related to the GB thickness, where Σ3(111) and Σ11(113) GBs with narrow GB thickness are weak in defect absorption and the others are strong. Furthermore, uniaxial tensile simulations perpendicular to the GB reveal that all of the mechanical performance of bi-crystals deteriorates after irradiation, which originates from dislocation propagation facilitated by irradiation defect clusters. In particular, regardless of whether the irradiation is applied, the maximum tensile strain, toughness, and Youngs modulus are monotonically correlated with GB tilt angle, while the ultimate tensile strength is stable for larger GB CSL parameter. Finally, on the basis of the evolution of the irradiation defects, microstructures, and mechanical performances, we proposed guidelines of rational design of irradiation-resistant Fe-Ni-Cr alloy.

cond-mat.mtrl-sci↗

arXiv2021

Tunable Flux through a Synthetic Hall Tube of Neutral Fermions

Hall tube with a tunable flux is an important geometry for studying quantum Hall physics, but its experimental realization in real space is still challenging. Here, we propose to realize a synthetic Hall tube with tunable flux in a one-dimensional optical lattice with the synthetic ring dimension defined by atomic hyperfine states. We investigate the effects of the flux on the system topology and study its quench dynamics. Utilizing the tunable flux, we show how to realize topological charge pumping, where interesting charge flow and transport are observed in rotated spin basis. Finally, we show that the recently observed quench dynamics in a synthetic Hall tube can be explained by the random flux existing in the experiment.

cond-mat.quant-gas↗

arXiv2021

Recent Advances on π-Conjugated Polymers as Active Elements in High Performance Organic Field-Effect Transistors

As high-performance organic semiconductors, π-conjugated polymers have attracted much attention due to their charming advantages including low-cost, solution processability, mechanical flexibility, and tunable optoelectronic properties. During the past several decades, the great advances have been made in polymers-based OFETs with p-type, n-type or even ambipolar characterics. Through chemical modification and alignment optimization, lots of conjugated polymers exhibited superior mobilities, and some mobilities are even larger than 10 cm2 V-1 s-1 in OFETs, which makes them very promising for the applications in organic electronic devices. This review describes the recent progress of the high performance polymers used in OFETs from the aspects of molecular design and assembly strategy. Furthermore, the current challenges and outlook in the design and development of conjugated polymers are also mentioned.

cond-mat.mtrl-sci↗

arXiv2021

Transformations of the transfinite plane

We study the existence of transformations of the transfinite plane that allow one to reduce Ramsey-theoretic statements concerning uncountable Abelian groups into classical partition relations for uncountable cardinals. To exemplify: we prove that for every inaccessible cardinal $κ$, if $κ$ admits a stationary set that does not reflect at inaccessibles, then the classical negative partition relation $κ\nrightarrow[κ]^2_κ$ implies that for every Abelian group $(G,+)$ of size $κ$, there exists a map $f:G\rightarrow G$ such that, for every $X\subseteq G$ of size $κ$ and every $g\in G$, there exist $x\neq y$ in $X$ such that $f(x+y)=g$.

math.LO↗