arXiv subjects

Jun Zhou

Publications and source records attributed to Jun Zhou.

At least 451 records · Page 25Linked to original sources

arXiv2020

A Comprehensive Analysis of Information Leakage in Deep Transfer Learning

Transfer learning is widely used for transferring knowledge from a source domain to the target domain where the labeled data is scarce. Recently, deep transfer learning has achieved remarkable progress in various applications. However, the source and target datasets usually belong to two different organizations in many real-world scenarios, potential privacy issues in deep transfer learning are posed. In this study, to thoroughly analyze the potential privacy leakage in deep transfer learning, we first divide previous methods into three categories. Based on that, we demonstrate specific threats that lead to unintentional privacy leakage in each category. Additionally, we also provide some solutions to prevent these threats. To the best of our knowledge, our study is the first to provide a thorough analysis of the information leakage issues in deep transfer learning methods and provide potential solutions to the issue. Extensive experiments on two public datasets and an industry dataset are conducted to show the privacy leakage under different deep transfer learning settings and defense solution effectiveness.

cs.CL↗

arXiv2020

Heterogeneous Graph Neural Network for Recommendation

The prosperous development of e-commerce has spawned diverse recommendation systems. As a matter of fact, there exist rich and complex interactions among various types of nodes in real-world recommendation systems, which can be constructed as heterogeneous graphs. How learn representative node embedding is the basis and core of the personalized recommendation system. Meta-path is a widely used structure to capture the semantics beneath such interactions and show potential ability in improving node embedding. In this paper, we propose Heterogeneous Graph neural network for Recommendation (HGRec) which injects high-order semantic into node embedding via aggregating multi-hops meta-path based neighbors and fuses rich semantics via multiple meta-paths based on attention mechanism to get comprehensive node embedding. Experimental results demonstrate the importance of rich high-order semantics and also show the potentially good interpretability of HGRec.

cs.SI↗

arXiv2020

Bandit Samplers for Training Graph Neural Networks

Several sampling algorithms with variance reduction have been proposed for accelerating the training of Graph Convolution Networks (GCNs). However, due to the intractable computation of optimal sampling distribution, these sampling algorithms are suboptimal for GCNs and are not applicable to more general graph neural networks (GNNs) where the message aggregator contains learned weights rather than fixed weights, such as Graph Attention Networks (GAT). The fundamental reason is that the embeddings of the neighbors or learned weights involved in the optimal sampling distribution are changing during the training and not known a priori, but only partially observed when sampled, thus making the derivation of an optimal variance reduced samplers non-trivial. In this paper, we formulate the optimization of the sampling variance as an adversary bandit problem, where the rewards are related to the node embeddings and learned weights, and can vary constantly. Thus a good sampler needs to acquire variance information about more neighbors (exploration) while at the same time optimizing the immediate sampling variance (exploit). We theoretically show that our algorithm asymptotically approaches the optimal variance within a factor of 3. We show the efficiency and effectiveness of our approach on multiple datasets.

cs.LG↗

arXiv2020

Topological Hall Effect and Skyrmion-like Bubbles at a Charge-transfer Interface

Exploring exotic interface magnetism due to charge transfer and strong spin-orbit coupling has profound application in future development of spintronic memory. Here, the emergence, tuning and interpretation of hump-shape Hall Effect from a CaMnO3/CaIrO3/CaMnO3 trilayer structure are studied in detail. The hump signal can be recognized as Topological Hall Effect suggesting the presence of Skyrmion-like magnetic bubbles; but the debated alternative interpretation where the signal being an artefact between two cancelling Anomalous Hall Effect loops is also discussed. Firstly, by tilting the magnetic field direction, the evolution of Hall signal suggests transformation of the bubbles topology into a more trivial kind. Secondly, by varying the thickness of CaMnO3, the optimal thicknesses for the hump signal emergence are found, suggesting a tuning of charge transfer fraction. Using high-resolution transmission electron microscopy, a stacking fault is also identified, which distinguishes the top and bottom CaMnO3/CaIrO3 interfaces in terms of charge transfer fraction and possible interfacial Dzyaloshinskii-Moriya Interaction. Finally, a spin-transfer torque experiment revealed a low threshold current density of ~10^9 A/m^2 for initiating the motion of bubbles. This discovery opens a possible route for integrating Skyrmions with antiferromagnetic spintronics.

cond-mat.mes-hall↗

arXiv2020

Interlayer and Intralayer Scale Aggregation for Scale-invariant Crowd Counting

Crowd counting is an important vision task, which faces challenges on continuous scale variation within a given scene and huge density shift both within and across images. These challenges are typically addressed using multi-column structures in existing methods. However, such an approach does not provide consistent improvement and transferability due to limited ability in capturing multi-scale features, sensitiveness to large density shift, and difficulty in training multi-branch models. To overcome these limitations, a Single-column Scale-invariant Network (ScSiNet) is presented in this paper, which extracts sophisticated scale-invariant features via the combination of interlayer multi-scale integration and a novel intralayer scale-invariant transformation (SiT). Furthermore, in order to enlarge the diversity of densities, a randomly integrated loss is presented for training our single-branch method. Extensive experiments on public datasets demonstrate that the proposed method consistently outperforms state-of-the-art approaches in counting accuracy and achieves remarkable transferability and scale-invariant property.

cs.CV↗

arXiv2020

Learning-Based Synthetic Dual Energy CT Imaging from Single Energy CT for Stopping Power Ratio Calculation in Proton Radiation Therapy

Purpose: Dual-energy CT (DECT) has been shown to derive stopping power ratio (SPR) map with higher accuracy than conventional single energy CT (SECT) by obtaining the energy dependence of photon interactions. However, DECT is not as widely implemented as SECT in proton radiation therapy simulation. This work presents a learning-based method to synthetize DECT images from SECT for proton radiation therapy. Methods: The proposed method uses a residual attention generative adversarial network. Residual blocks with attention gates were used to force the model focus on the difference between DECT maps and SECT images. To evaluate the accuracy of the method, we retrospectively investigated 20 head-and-neck cancer patients with both DECT and SECT scans available. The high and low energy CT images acquired from DECT acted as learning targets in the training process for SECT datasets and were evaluated against results from the proposed method using a leave-one-out cross-validation strategy. To evaluate our method in the context of a practical application, we generated SPR maps from sDECT using physics-based dual-energy stoichiometric method and compared the maps to those generated from DECT. Results: The synthesized DECT images showed an average mean absolute error around 30 Hounsfield Unit (HU) across the whole-body volume. The corresponding SPR maps generated from synthetic DECT showed an average normalized mean square error of about 1% with reduced noise level and artifacts than those from original DECT. Conclusions: The accuracy of the synthesized DECT image by our machine-learning-based method was evaluated on head and neck patient, and potential feasibility for proton treatment planning and dose calculation was shown by generating SPR map using the synthesized DECT.

physics.med-ph↗

arXiv2020

Practical Privacy Preserving POI Recommendation

Point-of-Interest (POI) recommendation has been extensively studied and successfully applied in industry recently. However, most existing approaches build centralized models on the basis of collecting users' data. Both private data and models are held by the recommender, which causes serious privacy concerns. In this paper, we propose a novel Privacy preserving POI Recommendation (PriRec) framework. First, to protect data privacy, users' private data (features and actions) are kept on their own side, e.g., Cellphone or Pad. Meanwhile, the public data need to be accessed by all the users are kept by the recommender to reduce the storage costs of users' devices. Those public data include: (1) static data only related to the status of POI, such as POI categories, and (2) dynamic data depend on user-POI actions such as visited counts. The dynamic data could be sensitive, and we develop local differential privacy techniques to release such data to public with privacy guarantees. Second, PriRec follows the representations of Factorization Machine (FM) that consists of linear model and the feature interaction model. To protect the model privacy, the linear models are saved on users' side, and we propose a secure decentralized gradient descent protocol for users to learn it collaboratively. The feature interaction model is kept by the recommender since there is no privacy risk, and we adopt secure aggregation strategy in federated learning paradigm to learn it. To this end, PriRec keeps users' private raw data and models in users' own hands, and protects user privacy to a large extent. We apply PriRec in real-world datasets, and comprehensive experiments demonstrate that, compared with FM, PriRec achieves comparable or even better recommendation accuracy.

cs.CR↗

arXiv2020

Secret Sharing based Secure Regressions with Applications

Nowadays, the utilization of the ever expanding amount of data has made a huge impact on web technologies while also causing various types of security concerns. On one hand, potential gains are highly anticipated if different organizations could somehow collaboratively share their data for technological improvements. On the other hand, data security concerns may arise for both data holders and data providers due to commercial or sociological concerns. To make a balance between technical improvements and security limitations, we implement secure and scalable protocols for multiple data holders to train linear regression and logistic regression models. We build our protocols based on the secret sharing scheme, which is scalable and efficient in applications. Moreover, our proposed paradigm can be generalized to any secure multiparty training scenarios where only matrix summation and matrix multiplications are used. We demonstrate our approach by experiments which shows the scalability and efficiency of our proposed protocols, and finally present its real-world applications.

cs.LG↗

arXiv2020

RNE: A Scalable Network Embedding for Billion-scale Recommendation

Nowadays designing a real recommendation system has been a critical problem for both academic and industry. However, due to the huge number of users and items, the diversity and dynamic property of the user interest, how to design a scalable recommendation system, which is able to efficiently produce effective and diverse recommendation results on billion-scale scenarios, is still a challenging and open problem for existing methods. In this paper, given the user-item interaction graph, we propose RNE, a data-efficient Recommendation-based Network Embedding method, to give personalized and diverse items to users. Specifically, we propose a diversity- and dynamics-aware neighbor sampling method for network embedding. On the one hand, the method is able to preserve the local structure between the users and items while modeling the diversity and dynamic property of the user interest to boost the recommendation quality. On the other hand the sampling method can reduce the complexity of the whole method theoretically to make it possible for billion-scale recommendation. We also implement the designed algorithm in a distributed way to further improves its scalability. Experimentally, we deploy RNE on a recommendation scenario of Taobao, the largest E-commerce platform in China, and train it on a billion-scale user-item graph. As is shown on several online metrics on A/B testing, RNE is able to achieve both high-quality and diverse results compared with CF-based methods. We also conduct the offline experiments on Pinterest dataset comparing with several state-of-the-art recommendation methods and network embedding methods. The results demonstrate that our method is able to produce a good result while runs much faster than the baseline methods.

cs.IR↗

arXiv2020

Adapted tree boosting for Transfer Learning

Secure online transaction is an essential task for e-commerce platforms. Alipay, one of the world's leading cashless payment platform, provides the payment service to both merchants and individual customers. The fraud detection models are built to protect the customers, but stronger demands are raised by the new scenes, which are lacking in training data and labels. The proposed model makes a difference by utilizing the data under similar old scenes and the data under a new scene is treated as the target domain to be promoted. Inspired by this real case in Alipay, we view the problem as a transfer learning problem and design a set of revise strategies to transfer the source domain models to the target domain under the framework of gradient boosting tree models. This work provides an option for the cold-starting and data-sharing problems.

cs.LG↗

arXiv2020

Unpack Local Model Interpretation for GBDT

A gradient boosting decision tree (GBDT), which aggregates a collection of single weak learners (i.e. decision trees), is widely used for data mining tasks. Because GBDT inherits the good performance from its ensemble essence, much attention has been drawn to the optimization of this model. With its popularization, an increasing need for model interpretation arises. Besides the commonly used feature importance as a global interpretation, feature contribution is a local measure that reveals the relationship between a specific instance and the related output. This work focuses on the local interpretation and proposes an unified computation mechanism to get the instance-level feature contributions for GBDT in any version. Practicality of this mechanism is validated by the listed experiments as well as applications in real industry scenarios.

cs.LG↗

arXiv2020

NetDP: An Industrial-Scale Distributed Network Representation Framework for Default Prediction in Ant Credit Pay

Ant Credit Pay is a consumer credit service in Ant Financial Service Group. Similar to credit card, loan default is one of the major risks of this credit product. Hence, effective algorithm for default prediction is the key to losses reduction and profits increment for the company. However, the challenges facing in our scenario are different from those in conventional credit card service. The first one is scalability. The huge volume of users and their behaviors in Ant Financial requires the ability to process industrial-scale data and perform model training efficiently. The second challenges is the cold-start problem. Different from the manual review for credit card application in conventional banks, the credit limit of Ant Credit Pay is automatically offered to users based on the knowledge learned from big data. However, default prediction for new users is suffered from lack of enough credit behaviors. It requires that the proposal should leverage other new data source to alleviate the cold-start problem. Considering the above challenges and the special scenario in Ant Financial, we try to incorporate default prediction with network information to alleviate the cold-start problem. In this paper, we propose an industrial-scale distributed network representation framework, termed NetDP, for default prediction in Ant Credit Pay. The proposal explores network information generated by various interaction between users, and blends unsupervised and supervised network representation in a unified framework for default prediction problem. Moreover, we present a parameter-server-based distributed implement of our proposal to handle the scalability challenge. Experimental results demonstrate the effectiveness of our proposal, especially in cold-start problem, as well as the efficiency for industrial-scale dataset.

cs.LG↗

arXiv2020

A Time Attention based Fraud Transaction Detection Framework

With online payment platforms being ubiquitous and important, fraud transaction detection has become the key for such platforms, to ensure user account safety and platform security. In this work, we present a novel method for detecting fraud transactions by leveraging patterns from both users' static profiles and users' dynamic behaviors in a unified framework. To address and explore the information of users' behaviors in continuous time spaces, we propose to use \emph{time attention based recurrent layers} to embed the detailed information of the time interval, such as the durations of specific actions, time differences between different actions and sequential behavior patterns,etc., in the same latent space. We further combine the learned embeddings and users' static profiles altogether in a unified framework. Extensive experiments validate the effectiveness of our proposed methods over state-of-the-art methods on various evaluation metrics, especially on \emph{recall at top percent} which is an important metric for measuring the balance between service experiences and risk of potential losses.

cs.LG↗

arXiv2020

Learning-Based Stopping Power Mapping on Dual Energy CT for Proton Radiation Therapy

Purpose: Dual-energy CT (DECT) has been used to derive relative stopping power (RSP) map by obtaining the energy dependence of photon interactions. The DECT-derived RSP maps could potentially be compromised by image noise levels and the severity of artifacts when using physics-based mapping techniques, which would affect subsequent clinical applications. This work presents a noise-robust learning-based method to predict RSP maps from DECT for proton radiation therapy. Methods: The proposed method uses a residual attention cycle-consistent generative adversarial (CycleGAN) network. CycleGAN were used to let the DECT-to-RSP mapping be close to a one-to-one mapping by introducing an inverse RSP-to-DECT mapping. We retrospectively investigated 20 head-and-neck cancer patients with DECT scans in proton radiation therapy simulation. Ground truth RSP values were assigned by calculation based on chemical compositions, and acted as learning targets in the training process for DECT datasets, and were evaluated against results from the proposed method using a leave-one-out cross-validation strategy. Results: The predicted RSP maps showed an average normalized mean square error (NMSE) of 2.83% across the whole body volume, and average mean error (ME) less than 3% in all volumes of interest (VOIs). With additional simulated noise added in DECT datasets, the proposed method still maintained a comparable performance, while the physics-based stoichiometric method suffered degraded inaccuracy from increased noise level. The average differences in DVH metrics for clinical target volumes (CTVs) were less than 0.2 Gy for D95% and Dmax with no statistical significance. Conclusion: These results strongly indicate the high accuracy of RSP maps predicted by our machine-learning-based method and show its potential feasibility for proton treatment planning and dose calculation.

physics.med-ph↗

arXiv2020

Distributed Deep Forest and its Application to Automatic Detection of Cash-out Fraud

Internet companies are facing the need for handling large-scale machine learning applications on a daily basis and distributed implementation of machine learning algorithms which can handle extra-large scale tasks with great performance is widely needed. Deep forest is a recently proposed deep learning framework which uses tree ensembles as its building blocks and it has achieved highly competitive results on various domains of tasks. However, it has not been tested on extremely large scale tasks. In this work, based on our parameter server system, we developed the distributed version of deep forest. To meet the need for real-world tasks, many improvements are introduced to the original deep forest model, including MART (Multiple Additive Regression Tree) as base learners for efficiency and effectiveness consideration, the cost-based method for handling prevalent class-imbalanced data, MART based feature selection for high dimension data and different evaluation metrics for automatically determining of the cascade level. We tested the deep forest model on an extra-large scale task, i.e., automatic detection of cash-out fraud, with more than 100 millions of training samples. Experimental results showed that the deep forest model has the best performance according to the evaluation metrics from different perspectives even with very little effort for parameter tuning. This model can block fraud transactions in a large amount of money each day. Even compared with the best-deployed model, the deep forest model can additionally bring into a significant decrease in economic loss each day.

cs.LG↗

arXiv2020

AGL: a Scalable System for Industrial-purpose Graph Machine Learning

Machine learning over graphs have been emerging as powerful learning tools for graph data. However, it is challenging for industrial communities to leverage the techniques, such as graph neural networks (GNNs), and solve real-world problems at scale because of inherent data dependency in the graphs. As such, we cannot simply train a GNN with classic learning systems, for instance parameter server that assumes data parallel. Existing systems store the graph data in-memory for fast accesses either in a single machine or graph stores from remote. The major drawbacks are in three-fold. First, they cannot scale because of the limitations on the volume of the memory, or the bandwidth between graph stores and workers. Second, they require extra development of graph stores without well exploiting mature infrastructures such as MapReduce that guarantee good system properties. Third, they focus on training but ignore the optimization of inference over graphs, thus makes them an unintegrated system. In this paper, we design AGL, a scalable, fault-tolerance and integrated system, with fully-functional training and inference for GNNs. Our system design follows the message passing scheme underlying the computations of GNNs. We design to generate the $k$-hop neighborhood, an information-complete subgraph for each node, as well as do the inference simply by merging values from in-edge neighbors and propagating values to out-edge neighbors via MapReduce. In addition, the $k$-hop neighborhood contains information-complete subgraphs for each node, thus we simply do the training on parameter servers due to data independency. Our system AGL, implemented on mature infrastructures, can finish the training of a 2-layer graph attention network on a graph with billions of nodes and hundred billions of edges in 14 hours, and complete the inference in 1.2 hour.

cs.SI↗

arXiv2020

Privacy Preserving PCA for Multiparty Modeling

In this paper, we present a general multiparty modeling paradigm with Privacy Preserving Principal Component Analysis (PPPCA) for horizontally partitioned data. PPPCA can accomplish multiparty cooperative execution of PCA under the premise of keeping plaintext data locally. We also propose implementations using two techniques, i.e., homomorphic encryption and secret sharing. The output of PPPCA can be sent directly to data consumer to build any machine learning models. We conduct experiments on three UCI benchmark datasets and a real-world fraud detection dataset. Results show that the accuracy of the model built upon PPPCA is the same as the model with PCA that is built based on centralized plaintext data.

cs.CR↗

arXiv2020

InfDetect: a Large Scale Graph-based Fraud Detection System for E-Commerce Insurance

The insurance industry has been creating innovative products around the emerging online shopping activities. Such e-commerce insurance is designed to protect buyers from potential risks such as impulse purchases and counterfeits. Fraudulent claims towards online insurance typically involve multiple parties such as buyers, sellers, and express companies, and they could lead to heavy financial losses. In order to uncover the relations behind organized fraudsters and detect fraudulent claims, we developed a large-scale insurance fraud detection system, i.e., InfDetect, which provides interfaces for commonly used graphs, standard data processing procedures, and a uniform graph learning platform. InfDetect is able to process big graphs containing up to 100 millions of nodes and billions of edges. In this paper, we investigate different graphs to facilitate fraudster mining, such as a device-sharing graph, a transaction graph, a friendship graph, and a buyer-seller graph. These graphs are fed to a uniform graph learning platform containing supervised and unsupervised graph learning algorithms. Cases on widely applied e-commerce insurance are described to demonstrate the usage and capability of our system. InfDetect has successfully detected thousands of fraudulent claims and saved over tens of thousands of dollars daily.

cs.CR↗