arXiv subjects

Jun Zhou

Publications and source records attributed to Jun Zhou.

At least 415 records · Page 23Linked to original sources

arXiv2021

Conditional Attention Networks for Distilling Knowledge Graphs in Recommendation

Knowledge graph is generally incorporated into recommender systems to improve overall performance. Due to the generalization and scale of the knowledge graph, most knowledge relationships are not helpful for a target user-item prediction. To exploit the knowledge graph to capture target-specific knowledge relationships in recommender systems, we need to distill the knowledge graph to reserve the useful information and refine the knowledge to capture the users' preferences. To address the issues, we propose Knowledge-aware Conditional Attention Networks (KCAN), which is an end-to-end model to incorporate knowledge graph into a recommender system. Specifically, we use a knowledge-aware attention propagation manner to obtain the node representation first, which captures the global semantic similarity on the user-item network and the knowledge graph. Then given a target, i.e., a user-item pair, we automatically distill the knowledge graph into the target-specific subgraph based on the knowledge-aware attention. Afterward, by applying a conditional attention aggregation on the subgraph, we refine the knowledge graph to obtain target-specific node representations. Therefore, we can gain both representability and personalization to achieve overall performance. Experimental results on real-world datasets demonstrate the effectiveness of our framework over the state-of-the-art algorithms.

cs.LG↗

arXiv2021

MixSeq: Connecting Macroscopic Time Series Forecasting with Microscopic Time Series Data

Time series forecasting is widely used in business intelligence, e.g., forecast stock market price, sales, and help the analysis of data trend. Most time series of interest are macroscopic time series that are aggregated from microscopic data. However, instead of directly modeling the macroscopic time series, rare literature studied the forecasting of macroscopic time series by leveraging data on the microscopic level. In this paper, we assume that the microscopic time series follow some unknown mixture probabilistic distributions. We theoretically show that as we identify the ground truth latent mixture components, the estimation of time series from each component could be improved because of lower variance, thus benefitting the estimation of macroscopic time series as well. Inspired by the power of Seq2seq and its variants on the modeling of time series data, we propose Mixture of Seq2seq (MixSeq), an end2end mixture model to cluster microscopic time series, where all the components come from a family of Seq2seq models parameterized by different parameters. Extensive experiments on both synthetic and real-world data show the superiority of our approach.

cs.LG↗

arXiv2021

Decomposing Complex Questions Makes Multi-Hop QA Easier and More Interpretable

Multi-hop QA requires the machine to answer complex questions through finding multiple clues and reasoning, and provide explanatory evidence to demonstrate the machine reasoning process. We propose Relation Extractor-Reader and Comparator (RERC), a three-stage framework based on complex question decomposition, which is the first work that the RERC model has been proposed and applied in solving the multi-hop QA challenges. The Relation Extractor decomposes the complex question, and then the Reader answers the sub-questions in turn, and finally the Comparator performs numerical comparison and summarizes all to get the final answer, where the entire process itself constitutes a complete reasoning evidence path. In the 2WikiMultiHopQA dataset, our RERC model has achieved the most advanced performance, with a winning joint F1 score of 53.58 on the leaderboard. All indicators of our RERC are close to human performance, with only 1.95 behind the human level in F1 score of support fact. At the same time, the evidence path provided by our RERC framework has excellent readability and faithfulness.

cs.CL↗

arXiv2021

A Bi-Level Framework for Learning to Solve Combinatorial Optimization on Graphs

Combinatorial Optimization (CO) has been a long-standing challenging research topic featured by its NP-hard nature. Traditionally such problems are approximately solved with heuristic algorithms which are usually fast but may sacrifice the solution quality. Currently, machine learning for combinatorial optimization (MLCO) has become a trending research topic, but most existing MLCO methods treat CO as a single-level optimization by directly learning the end-to-end solutions, which are hard to scale up and mostly limited by the capacity of ML models given the high complexity of CO. In this paper, we propose a hybrid approach to combine the best of the two worlds, in which a bi-level framework is developed with an upper-level learning method to optimize the graph (e.g. add, delete or modify edges in a graph), fused with a lower-level heuristic algorithm solving on the optimized graph. Such a bi-level approach simplifies the learning on the original hard CO and can effectively mitigate the demand for model capacity. The experiments and results on several popular CO problems like Directed Acyclic Graph scheduling, Graph Edit Distance and Hamiltonian Cycle Problem show its effectiveness over manually designed heuristics and single-level learning methods.

cs.LG↗

arXiv2021

Reminding the Incremental Language Model via Data-Free Self-Distillation

Incremental language learning with pseudo-data can alleviate catastrophic forgetting in neural networks. However, to obtain better performance, former methods have higher demands for pseudo-data of the previous tasks. The performance dramatically decreases when fewer pseudo-data are employed. In addition, the distribution of pseudo-data gradually deviates from the real data with the sequential learning of different tasks. The deviation will be greater with more tasks learned, which results in more serious catastrophic forgetting. To address these issues, we propose reminding incremental language model via data-free self-distillation (DFSD), which includes self-distillation based on the Earth Mover's Distance and hidden data augmentation. By estimating the knowledge distribution in all layers of GPT-2 and transforming it from teacher model to student model, the Self-distillation based on the Earth Mover's Distance can significantly reduce the demand for pseudo-data. Hidden data augmentation can greatly alleviate the catastrophic forgetting caused by deviations via modeling the generation of pseudo-data as a hidden data augmentation process, where each sample is a mixture of all trained task data. The experimental results demonstrate that our DFSD can exceed the previous state-of-the-art methods even if the maximum decrease in pseudo-data is 90%.

cs.CL↗

arXiv2021

Selective hydrogenation improves interface properties of high-k dielectrics on 2D semiconductors

The integration of high-k dielectrics with two-dimensional (2D) semiconductors is a critical step towards high-performance nanoelectronics, which however remains challenging due to high density of interface states and the damage to the monolayer 2D semiconductors. In this study, we propose a selective hydrogenation strategy to improve the interface properties while do not affect the 2D semiconductors. Using the interface of monolayer MoS2 and silicon nitride as an example, we show substantially improved interface properties for electronic applications after the interfacial hydrogenation, as evidenced by reduced inhomogeneous charge redistribution, increased band offset, and untouched electronic properties of MoS2. Interestingly, this hydrogenation process selectively occurs only at the silicon nitride surface and is compatible with the current semiconductor fabrication process. We further show that this strategy is general and applicable to other interfaces between high-k dielectrics and 2D semiconductors such as HfO2 on the monolayer MoS2. Our results demonstrate a simple yet viable way to improve the interfacial properties for integrating many high-k dielectrics on a broad range of two-dimensional transition metal disulfide semiconductors.

cond-mat.mtrl-sci↗

arXiv2021

PP-OCRv2: Bag of Tricks for Ultra Lightweight OCR System

Optical Character Recognition (OCR) systems have been widely used in various of application scenarios. Designing an OCR system is still a challenging task. In previous work, we proposed a practical ultra lightweight OCR system (PP-OCR) to balance the accuracy against the efficiency. In order to improve the accuracy of PP-OCR and keep high efficiency, in this paper, we propose a more robust OCR system, i.e. PP-OCRv2. We introduce bag of tricks to train a better text detector and a better text recognizer, which include Collaborative Mutual Learning (CML), CopyPaste, Lightweight CPUNetwork (LCNet), Unified-Deep Mutual Learning (U-DML) and Enhanced CTCLoss. Experiments on real data show that the precision of PP-OCRv2 is 7% higher than PP-OCR under the same inference cost. It is also comparable to the server models of the PP-OCR which uses ResNet series as backbones. All of the above mentioned models are open-sourced and the code is available in the GitHub repository PaddleOCR which is powered by PaddlePaddle.

cs.CV↗

arXiv2021

Thermal percolation and electrical insulation in composite materials with partially metallic coated fillers

We propose a type of thermal interface materials incorporating insulating nanowires with partially metallic coating in insulating polymer matrix.Large thermal conductivity can be obtained due to thermal percolation while the electrical insulation is maintained by controlling $C_{\rm M}φ< φ_{\rm c}^{\rm e}$ and $φ> φ_{\rm c}^{\rm th}$, where $φ$ is the volume fraction of filler, $C_{\rm M}$ is the metallic coating fraction, $φ_{\rm c}^{\rm e}$ and $φ_{\rm c}^{\rm th}$ are the electrical and thermal percolation thresholds, respectively. The electrical conductivity of such composite materials can further be regulated by coating configuration. In this regard, we propose the concept of "thermal-percolation electrical-insulation", providing a guide to design efficient hybrid thermal interface materials.

physics.app-ph↗

arXiv2021

Large-Scale Secure XGB for Vertical Federated Learning

Privacy-preserving machine learning has drawn increasingly attention recently, especially with kinds of privacy regulations come into force. Under such situation, Federated Learning (FL) appears to facilitate privacy-preserving joint modeling among multiple parties. Although many federated algorithms have been extensively studied, there is still a lack of secure and practical gradient tree boosting models (e.g., XGB) in literature. In this paper, we aim to build large-scale secure XGB under vertically federated learning setting. We guarantee data privacy from three aspects. Specifically, (i) we employ secure multi-party computation techniques to avoid leaking intermediate information during training, (ii) we store the output model in a distributed manner in order to minimize information release, and (iii) we provide a novel algorithm for secure XGB predict with the distributed model. Furthermore, by proposing secure permutation protocols, we can improve the training efficiency and make the framework scale to large dataset. We conduct extensive experiments on both public datasets and real-world datasets, and the results demonstrate that our proposed XGB models provide not only competitive accuracy but also practical performance.

cs.LG↗

arXiv2021

Improving Transferability of Adversarial Patches on Face Recognition with Generative Models

Face recognition is greatly improved by deep convolutional neural networks (CNNs). Recently, these face recognition models have been used for identity authentication in security sensitive applications. However, deep CNNs are vulnerable to adversarial patches, which are physically realizable and stealthy, raising new security concerns on the real-world applications of these models. In this paper, we evaluate the robustness of face recognition models using adversarial patches based on transferability, where the attacker has limited accessibility to the target models. First, we extend the existing transfer-based attack techniques to generate transferable adversarial patches. However, we observe that the transferability is sensitive to initialization and degrades when the perturbation magnitude is large, indicating the overfitting to the substitute models. Second, we propose to regularize the adversarial patches on the low dimensional data manifold. The manifold is represented by generative models pre-trained on legitimate human face images. Using face-like features as adversarial perturbations through optimization on the manifold, we show that the gaps between the responses of substitute models and the target models dramatically decrease, exhibiting a better transferability. Extensive digital world experiments are conducted to demonstrate the superiority of the proposed method in the black-box setting. We apply the proposed method in the physical world as well.

cs.CV↗

arXiv2021

A Unified Framework for Cross-Domain and Cross-System Recommendations

Cross-Domain Recommendation (CDR) and Cross-System Recommendation (CSR) have been proposed to improve the recommendation accuracy in a target dataset (domain/system) with the help of a source one with relatively richer information. However, most existing CDR and CSR approaches are single-target, namely, there is a single target dataset, which can only help the target dataset and thus cannot benefit the source dataset. In this paper, we focus on three new scenarios, i.e., Dual-Target CDR (DTCDR), Multi-Target CDR (MTCDR), and CDR+CSR, and aim to improve the recommendation accuracy in all datasets simultaneously for all scenarios. To do this, we propose a unified framework, called GA (based on Graph embedding and Attention techniques), for all three scenarios. In GA, we first construct separate heterogeneous graphs to generate more representative user and item embeddings. Then, we propose an element-wise attention mechanism to effectively combine the embeddings of common entities (users/items) learned from different datasets. Moreover, to avoid negative transfer, we further propose a Personalized training strategy to minimize the embedding difference of common entities between a richer dataset and a sparser dataset, deriving three new models, i.e., GA-DTCDR-P, GA-MTCDR-P, and GA-CDR+CSR-P, for the three scenarios respectively. Extensive experiments conducted on four real-world datasets demonstrate that our proposed GA models significantly outperform the state-of-the-art approaches.

cs.IR↗

arXiv2021

SIFN: A Sentiment-aware Interactive Fusion Network for Review-based Item Recommendation

Recent studies in recommender systems have managed to achieve significantly improved performance by leveraging reviews for rating prediction. However, despite being extensively studied, these methods still suffer from some limitations. First, previous studies either encode the document or extract latent sentiment via neural networks, which are difficult to interpret the sentiment of reviewers intuitively. Second, they neglect the personalized interaction of reviews with user/item, i.e., each review has different contributions when modeling the sentiment preference of user/item. To remedy these issues, we propose a Sentiment-aware Interactive Fusion Network (SIFN) for review-based item recommendation. Specifically, we first encode user/item reviews via BERT and propose a light-weighted sentiment learner to extract semantic features of each review. Then, we propose a sentiment prediction task that guides the sentiment learner to extract sentiment-aware features via explicit sentiment labels. Finally, we design a rating prediction task that contains a rating learner with an interactive and fusion module to fuse the identity (i.e., user and item ID) and each review representation so that various interactive features can synergistically influence the final rating score. Experimental results on five real-world datasets demonstrate that the proposed model is superior to state-of-the-art models.

cs.IR↗

arXiv2021

Structure dependent and strain tunable magnetic ordering in ultrathin chromium telluride

Two-dimensional (2D) chromium tellurides have attracted considerable research interest for their high Curie temperatures. Their magnetic properties have been found diverse in various experiments, the understanding of which however remains limited. In this work, we report that the magnetic ordering of ultrathin chromium tellurides is structure dependent and can be tuned by external strain. Based on first-principles calculations and Monte Carlo simulations, we show long-range stable magnetism with high and low Curie temperature, and short-range magnetism in 2D Cr5Te8, CrTe2, and Cr2Te3 layers, respectively. We further find that ferromagnetic-to-antiferromagnetic transition can be realized by 2% compressive strain for CrTe2 and 2% tensile strain for Cr2Te3, and their magnetic easy axis is tuned from out-of-plane to in-plane by the medium tensile and compressive strain. This strain dependent magnetic coupling is found to be related to Cr-Cr direct exchange and the change of magnetic anisotropy is understood by the atom and orbital resolved magnetic anisotropy energy and second order perturbation theory. Our results reveal the important roles of the structure and strain in determining the magnetic ordering in 2D chromium telluride, shedding light on understanding of the diverse magnetic properties observed in experiments.

cond-mat.mtrl-sci↗

arXiv2021

When Homomorphic Encryption Marries Secret Sharing: Secure Large-Scale Sparse Logistic Regression and Applications in Risk Control

Logistic Regression (LR) is the most widely used machine learning model in industry for its efficiency, robustness, and interpretability. Due to the problem of data isolation and the requirement of high model performance, many applications in industry call for building a secure and efficient LR model for multiple parties. Most existing work uses either Homomorphic Encryption (HE) or Secret Sharing (SS) to build secure LR. HE based methods can deal with high-dimensional sparse features, but they incur potential security risks. SS based methods have provable security, but they have efficiency issue under high-dimensional sparse features. In this paper, we first present CAESAR, which combines HE and SS to build secure large-scale sparse logistic regression model and achieves both efficiency and security. We then present the distributed implementation of CAESAR for scalability requirement. We have deployed CAESAR in a risk control task and conducted comprehensive experiments. Our experimental results show that CAESAR improves the state-of-the-art model by around 130 times.

cs.CR↗

arXiv2021

Improvement of Normal Estimation for PointClouds via Simplifying Surface Fitting

With the burst development of neural networks in recent years, the task of normal estimation has once again become a concern. By introducing the neural networks to classic methods based on problem-specific knowledge, the adaptability of the normal estimation algorithm to noise and scale has been greatly improved. However, the compatibility between neural networks and the traditional methods has not been considered. Similar to the principle of Occam's razor, that is, the simpler is better. We observe that a more simplified process of surface fitting can significantly improve the accuracy of the normal estimation. In this paper, two simple-yet-effective strategies are proposed to address the compatibility between the neural networks and surface fitting process to improve normal estimation. Firstly, a dynamic top-k selection strategy is introduced to better focus on the most critical points of a given patch, and the points selected by our learning method tend to fit a surface by way of a simple tangent plane, which can dramatically improve the normal estimation results of patches with sharp corners or complex patterns. Then, we propose a point update strategy before local surface fitting, which smooths the sharp boundary of the patch to simplify the surface fitting process, significantly reducing the fitting distortion and improving the accuracy of the predicted point normal. The experiments analyze the effectiveness of our proposed strategies and demonstrate that our method achieves SOTA results with the advantage of higher estimation accuracy over most existed approaches.

cs.CV↗

arXiv2021

Fast and Accurate Normal Estimation for Point Cloud via Patch Stitching

This paper presents an effective normal estimation method adopting multi-patch stitching for an unstructured point cloud. The majority of learning-based approaches encode a local patch around each point of a whole model and estimate the normals in a point-by-point manner. In contrast, we suggest a more efficient pipeline, in which we introduce a patch-level normal estimation architecture to process a series of overlapping patches. Additionally, a multi-normal selection method based on weights, dubbed as multi-patch stitching, integrates the normals from the overlapping patches. To reduce the adverse effects of sharp corners or noise in a patch, we introduce an adaptive local feature aggregation layer to focus on an anisotropic neighborhood. We then utilize a multi-branch planar experts module to break the mutual influence between underlying piecewise surfaces in a patch. At the stitching stage, we use the learned weights of multi-branch planar experts and distance weights between points to select the best normal from the overlapping parts. Furthermore, we put forward constructing a sparse matrix representation to reduce large-scale retrieval overheads for the loop iterations dramatically. Extensive experiments demonstrate that our method achieves SOTA results with the advantage of lower computational costs and higher robustness to noise over most of the existing approaches.

cs.CV↗

arXiv2021

Controllable phase transitions between multiple charge density waves in monolayer 1T-VSe$_2$ via doping and strain engineering

Two-dimensional (2D) materials are known to possess emergent properties that are not found in their bulk counterparts. Recent experiments have shown a $\sqrt7 \times \sqrt3$ charge density wave (CDW) in monolayer 1T-VSe$_2$, in contrast to the $4\times 4\times 3$ phase in bulk. Here, via first-principles calculations, we show that multiple CDW phases compete in monolayer VSe$_2$, the ground state of which can be tuned by charge doping and in-plane biaxial strain. With doping, the $\sqrt7 \times \sqrt3$ CDW of the pristine VSe$_2$ transfers to a $3 \times \sqrt3$ and $4\times 4$ phase, the latter of which is a projection of the bulk counterpart, at critical doping concentrations of around 0.2 holes per formula unit and 0.25 electrons per formula unit, respectively. The $4\times 4$ CDW phase can also be stabilized under compressive strain. Although electron-phonon coupling is prevailing in the CDW formation, we show that Fermi surface nesting is a good starting point to explain most of these transitions in monolayer 1T-VSe$_2$. These results make VSe$_2$ an appealing material for electronic devices based on controllable CDW phase transitions.

cond-mat.mtrl-sci↗

arXiv2021

Goal-Oriented Gaze Estimation for Zero-Shot Learning

Zero-shot learning (ZSL) aims to recognize novel classes by transferring semantic knowledge from seen classes to unseen classes. Since semantic knowledge is built on attributes shared between different classes, which are highly local, strong prior for localization of object attribute is beneficial for visual-semantic embedding. Interestingly, when recognizing unseen images, human would also automatically gaze at regions with certain semantic clue. Therefore, we introduce a novel goal-oriented gaze estimation module (GEM) to improve the discriminative attribute localization based on the class-level attributes for ZSL. We aim to predict the actual human gaze location to get the visual attention regions for recognizing a novel object guided by attribute description. Specifically, the task-dependent attention is learned with the goal-oriented GEM, and the global image features are simultaneously optimized with the regression of local attribute features. Experiments on three ZSL benchmarks, i.e., CUB, SUN and AWA2, show the superiority or competitiveness of our proposed method against the state-of-the-art ZSL methods. The ablation analysis on real gaze data CUB-VWSW also validates the benefits and accuracy of our gaze estimation module. This work implies the promising benefits of collecting human gaze dataset and automatic gaze estimation algorithms on high-level computer vision tasks. The code is available at https://github.com/osierboy/GEM-ZSL.

cs.CV↗