Visa International Service Association

International Business Machines Corporation

Nokia Technologies Oy

Alibaba Group Holding Limited

Baidu Online Network Technology (Beijing) Co., Ltd.

Southwest Petroleum University

Google Inc.

Emc IP Holding Company LLC

Sichuan University

Shanghai Xiaodu Technology Co. Ltd.

Beijing Research Center for Information Technology in Agriculture

Visa International Service Application

Hao Yang

Wei Zhang

Yuhang Wu

Liang Gou

Mahashweta Das

Zhen Liu

Liang Wang

Minkyong Kim

Sunpreet Singh Arora

Fan Ye

Michael Yeh

Yan Zheng

Yufei Li

Yu-San Lin

Hui Lei

Junpeng Wang

Robert Brian Christensen

Yong Qing Lu

Fei Wang

Maryam Moosaei

Yanhong Wu

Huiyuan Chen

Yang Liu

Yi Bin Wang

Liangzhao Zeng

Sivanarayana Gaddam

Srinivasan Parthasarathy

Daniel Nikolaus Bauer

Luis Garcés-Erice

Atul Luykx

Yi Lin

Kristijan Dragicevic

Dhruv Gelda

Mangesh Bendre

Yu Gu

Chai Wah Wu

Dimitrios Pendarakis

Frank Andre Schaffa

Dong Hui Liu

Peter Hans Westerink

Ying Chu Wang

Julio Nogima

Anand Ranganathan

Han Chen

Igor V Karpenko

Peter A Walker

Paolo Dettori

Gidon Gershinsky

Nicholas David Butler

Cheng Song

Konstantin Shagin

Honghui Xia

Rohit Sinha

Shiquan Ye

Xiaoping Li

Xiaohua Tan

Marc-Thomas H Schmidt

Fan Zhou

Bo Dong

Shamim Samadi

Hou Gang Li

Zhenghua Fu

Kyriakos Karenos

Ablaikhan Akhazhanov

Yu Zhu

Xiaodong Yang

Yu Meng Li

Yuexi Chen

Hui Liu

Jilei Tian

Simon Hurry

Wei Wei

Francis Nicholas Parr

Zhen Wen

Haixun Wang

Chiranjeet Chetia

Mahdi Zamani

Peter Rindal

Minghua Xu

Srinivasan Raghuraman

Jiankang Sun

John Christopher Hawkins

Chunjiang Zhao

Qing Ren

Haitao Li

Shi Cao

Xiajun Luo

Andreas Aabye

Xiangtao Jiang

Jie Wen

Hua Yan

Jian Wen

Fei Shen

Hossein Hamooni

Guofan Jiang

Harishkumar Sundarji Majithiya

Roshni Ann Samuel

Xiaobo Dong

Xinxing Yang

A method includes determining a set of regions for each of a first plurality of images of a first item type, a second plurality of images of a second item type, and a third plurality of images of a third item type. The method also includes for each region in each set of regions of the images, generating, by the processing computer, a vector representing the region, and then generating a plurality of aggregated messages using the vectors corresponding to combinations of images of different types of items, the images being from the first, second, and third plurality of images. Then, unified embeddings are generated for the images in the first, second, and third plurality of images, respectively, using aggregated messages in the plurality of aggregated messages. Matching scores associated with combinations of the images are created using the unified embeddings and a machine learning model.

Scalable neural tensor network with multi-aspect feature interactions

Methods and systems for performing federated private anomaly detection are disclosed. An anomaly detector computer can collaborate with an aggregator computer and an account management computer in order to train machine learning models, which can be used to classify events as not fraudulent or fraudulent. The anomaly detector computer can obliviously use private information (e.g., account flags and account flag values) held by the account management computer to train and use the machine learning models, such that the anomaly detector does not become aware of the account flags or account flags values. Such a system can be used, for example, for the detection and prevention of financial crime. The anomaly detector computer can use the account flag information possessed by the account management computer to identify fraudulent events performed by customers of the organization operating the account management computer.

Secure anomaly detection using federated learning

Systems, methods, and computer program products are provided for saving memory during training of knowledge graph neural networks. The method includes receiving a training dataset including a first set of knowledge graph embeddings associated with a plurality of entities for a first layer of a knowledge graph, inputting the training dataset into a knowledge graph neural network to generate at least one further set of knowledge graph embeddings associated with the plurality of entities for at least one further layer of the knowledge graph, quantizing the at least one further set of knowledge graph embeddings to provide at least one set of quantized knowledge graph embeddings, storing the at least one set of quantized knowledge graph embeddings in a memory, and dequantizing the at least one set of quantized knowledge graph embeddings to provide at least one set of dequantized knowledge graph embeddings.

System, method, and computer program product for saving memory during training of knowledge graph neural networks

Provided is a method for normalizing embeddings for cross-embedding alignment. The method may include applying mean centering to the at least one embedding set, applying spectral normalization to the at least one embedding set, and/or applying length normalization to the at least one embedding set. Spectral normalization may include decomposing the at least one embedding set, determining an average singular value of the at least one embedding set, determining a respective substitute singular value for each respective singular value of a diagonal matrix, and/or replacing the at least one embedding set with a product of the at least one embedding set, a right singular vector, and an inverse of the substitute diagonal matrix. The mean centering, spectral normalization, and/or length normalization may be iteratively repeated for a configurable number of iterations. A system and computer program product are also disclosed.

Method, system, and computer program product for normalizing embeddings for cross-embedding alignment

Systems, methods, and computer program products are provided for predicting a specified geographic area of a user. An example system includes a processor configured to determine a verified geographic area associated with each user, and determine a feature vector associated with an account of each user. The processor is also configured to receive transaction data and determine a value of each parameter of the feature vector for each user based on the transaction data to produce a training matrix. The processor is further configured to train and validate a geographic area prediction model based on the training matrix. The processor is further configured to repeatedly generate a prediction that a user will conduct a transaction in a geographic area, communicate an offer to the user based on the prediction, receive new training data, and update the geographic area prediction model based on the new training data.

System, method, and computer program product for predicting a specified geographic area of a user

Methods, systems, and computer program products may formulate an iterative data mix up problem into a Markov decision process (MDP) with a tailored reward signal to guide a learning process. To solve the MDP, a deep deterministic actor-critic framework may be modified to adapt a discrete-continuous decision space for training a data augmentation policy.

Method, system, and computer program product for synthetic oversampling for boosting supervised anomaly detection

Provided is a method for detecting group activities in a network. The method may include receiving interaction data associated with a plurality of interactions. For each account identifier associated with at least one interaction, a value may be determined for each of a first set of categories, and a vector may be generated based on the value for each category. The length of each vector may be determined. At least one relational graph may be generated based on the interaction data. Each relational graph may be associated with a respective category of a second set of categories. At least one cluster of nodes may be determined based on the relational graph(s). A score for each cluster may be determined based on the length of the vector associated with the account identifier of each node of the cluster of nodes. A system and computer program product are also disclosed.

Method, system, and computer program product for detecting group activities in a network

A method includes extracting, by an analysis computer, a plurality of first datasets from a plurality of graph snapshots using a structural self-attention module. The analysis computer can then extract at least a second dataset from the plurality of first datasets using a temporal self-attention module across the plurality of graph snapshots. The analysis computer can then perform graph context prediction with at least the second dataset.

Dynamic graph representation learning via attention networks

Provided are systems, methods, and computer program products for generating node embeddings. The system includes at least one processor programmed or configured to generate a graph comprising a plurality of nodes, generate an embedding for each node of the plurality of nodes, each embedding comprising at least one polar angle and a vector length, store each embedding of a plurality of embeddings in memory, and in response to processing the graph with a machine-learning algorithm, convert at least one embedding of the plurality of embeddings to Cartesian coordinates.

Systems, methods, and computer program products for generating node embeddings

A method for dynamically assigning an inference request is disclosed. A method for dynamically assigning an inference request may include determining at least one model to process an inference request on a plurality of computing platforms, the plurality of computing platforms including at least one Central Processing Unit (CPU) and at least one Graphics Processing Unit (GPU), obtaining, with at least one processor, profile information of the at least one model, the profile information including measured characteristics of the at least one model, dynamically determining a selected computing platform from between the at least one CPU and the at least one GPU for responding to the inference request based on an optimized objective associated with a status of the computing platform and the profile information, and routing, with at least one processor, the inference request to the selected computing platform. A system and computer program product are also disclosed.

Method, system, and computer program product for dynamically assigning an inference request to a CPU or GPU

The system and method may assess the merchant risk level on a more continuous scale rather than a binary categorization. It may produce a continuous risk score proportional to the likelihood of a merchant being risky, effectively addressing the issue of shades of gray encountered by the traditional blacklisting approach. The continuous risk score feature provides greater flexibility as it allows the payment network to make dynamic pricing decisions (known as interchange optimization) based on the merchant risk level. Using collective intelligence from transactions across the payment network, the system and method may be able to assess the merchant risk level with high accuracy. The system and method may be particularly beneficial to small merchants with low transaction volume as even a few fraudulent transactions can easily put them in the high-risk merchant category. Further, the system and method may help payment processing networks make better decision on cross-border transactions.

Method and system for assessing the reputation of a merchant

Provided are systems for determining adversarial examples that include at least one processor to determine a first additional input from a plurality of additional inputs based on a proximity of the first additional input to an initial input, determine a second additional input from the plurality of additional inputs based on a proximity of the second additional input to the first additional input, generate a first vector embedding, a second vector embedding and a third vector embedding based on the second additional input, generate a first relational embedding, a second relational embedding, and a third relational embedding based on the third vector embedding and the first vector embedding, concatenate the first relational embedding, the second relational embedding, and the third relational embedding to provide a concatenated version, and determine whether the first input is an adversarial example based on the concatenated version. Methods and computer program products are also provided.

System, method, and computer program product for determining adversarial examples

A computer system can perform a semi-supervised machine learning processes to cluster a plurality of entities within a population based on their features and associated labels. The computer system can generate visualization data representing the clusters of entities and associated labels for displaying on a user interface. A user can review the clustering of entities and use the user interface to add or modify the labels associated with a particular entity or set of entities. The computer system can use the user's feedback to update the labels and then re-determine the clustering of entities using the semi-supervised machine learning process with the updated labels as input. As such, the computer system can use the user's feedback to improve the accuracy of the machine learning model without requiring a larger amount of labeled input data.

Mixed-initiative machine learning systems and methods for determining segmentations

Disclosed are a system, method, and computer program product for user network activity anomaly detection. The method includes generating a multilayer graph from network resource data, and generating an adjacency matrix associated with each layer of the multilayer graph to produce a plurality of adjacency matrices. The method further includes assigning a weight to each adjacency matrix to produce a plurality of weights, and generating a merged single layer graph by merging the plurality of layers based on a weighted sum of the plurality of adjacency matrices using the plurality of weights. The method further includes generating a set of anomaly scores by generating, for each node in the merged single layer graph, an anomaly score. The method further includes determining a set of anomalous users based on the set of anomaly scores, detecting fraudulent network activity based on the set of anomalous users, and executing a fraud mitigation process.

System, method, and computer program product for user network activity anomaly detection

Provided is a system, method, and computer program product for predicting a specified geographic area of a user. The method includes receiving transaction data associated with a plurality of transactions during a predetermined time interval. The method also includes generating a geographic area prediction model based on the transaction data by determining a verified geographic area for each user, and determining transaction data associated with a plurality of transactions involving each user for a plurality of feature vector parameters, training the geographic area prediction model based on the plurality of feature vector parameters for the verified geographic area for each user, and validating the geographic area prediction model based on the plurality of feature vector parameters for the verified geographic area for each user.

Growing community of inventors

San Jose, CA, United States of America

Hao Yang