公开(公告)号：US20240273339A1

公开(公告)日：2024-08-15

申请号：US18169699

申请日：2023-02-15

申请人： PAYPAL, INC.

发明人： Wei Zhao ,  Zexia Zhou ,  Jiahui Qian ,  Sizhe Wan

IPC分类号： G06N3/042 ,  G06N3/08

CPC分类号： G06N3/042 ,  G06N3/08

摘要： Methods and systems are presented for configuring, training, and utilizing a machine learning model that includes different experts corresponding to different domains, such that the machine learning model may facilitate transfer of knowledge acquired from one domain to another domain and to use different mixtures of experts to perform tasks across the different domains. The machine learning model includes individual domain experts configured to process input values corresponding to features that are unique to the corresponding domains. The machine learning model also includes a common expert configured to process input values corresponding to features that are common to the different domains. By training the machine learning model using training data associated with a first domain, both a first domain expert and the common expert are trained. The knowledge acquired by the common expert can then be utilized when processing tasks associated with a second domain.

公开(公告)号：US20200327549A1

公开(公告)日：2020-10-15

申请号：US15775980

申请日：2017-11-08

申请人： PAYPAL, INC.

发明人： Yanzan Zhou ,  Shuoyuan Wang ,  Wei Zhao ,  Ying Chen

IPC分类号： G06Q20/40 ,  G06N3/04 ,  G06N20/20

摘要： A particular machine learning architecture involving a two-part artificial intelligence (AI) model is disclosed, with one portion being trained on first data (e.g. older data) and another portion being trained on second data (e.g. newer data) in various embodiments. A robust AI model can be combined with an adaptive AI model to account for long-term trends as well as newly emerging population trends. The model architecture can be constructed using gradient boosting trees, artificial neural networks, or other machine learning models. The adaptive AI model can be re-trained on a more frequent basis than the robust AI model, and can use newer types of data in its classification techniques. The adaptive and robust AI models can be combined using logistic regression to provide unified predictions. Electronic transactions and other types of data subject to potential pattern shifts can thus be more accurately classified.