Paypal, Inc.

Karl Anton Hennig

Ajay Aswal

Bisrat Zerihun

Chun Kiat Ho

Lin Song

Kai Xie

Alexander Yuryevich Shvid

Thomas Anthony Doran

Venkata Ramana Varma Nadimpalli

Jun Zhang

Fei Li

Xiaohan Yun

Pengshan Zhang

Edwin Boaz Soenaryo

Vahini Pobbathi

Meng Zang

Quin Zuo

Rey Neo

Jerry Tso

Huynh Phi Long

Swathi Bangalore Ananthakumar

Systems and methods for automatic generation of free-form conversational interfaces are disclosed. In one embodiment, a system receives an input from a user device through a conversational graphical user interface (GUI). An intent of the user may be determined based on the received input. Based on the intent of the user, the system may identify, from a plurality of objects available to the system, one or more objects. Each of the plurality of objects has annotations corresponding to one or more elements of the object and one or more functions of the object. The one or more functions corresponding to the one or more elements are executable to perform an action upon corresponding elements. Based on the identified one or more objects and the annotations of the identified one or more objects, the system may generate a dynamic dialogue flow for the conversational GUI, where the dynamic dialogue flow is generated in real-time during a conversational GUI session.

Free-form, automatically-generated conversational graphical user interfaces

Techniques for detecting a fraudulent attempt by an adversarial user to voice verify as a user are presented. An authenticator component can determine characteristics of voice information received in connection with a user account based on analysis of the voice information. In response to determining the characteristics sufficiently match characteristics of a voice print associated with the user account, authenticator component can determine a similarity score based on comparing the characteristics of the voice information and other characteristics of a set of previously stored voice prints associated with the user account. Authenticator component can determine whether the similarity score is higher than a threshold similarity score to indicate whether the voice information is a replay of a recording or a deep fake emulation of the voice of the user. Above the threshold can indicate the voice information is fraudulent, and below the threshold can indicate the voice information is valid.

Adversarially robust voice biometrics, secure recognition, and identification

Aspects of the present disclosure involve systems, methods, devices, and the like for generating highly scalable temporal graph databases. In one embodiment, a novel architecture is presented that enables the identification of mutation or changes in graphs. For the identification a combination of graph-based modeling and journal entry is used. Events occurring are consumed and changes are ingested, transformed for use by a graph simulation system. The changes are journaled using a vertex centric temporal journaling schema.

System and method for generating highly scalable temporal graph database

Systems and methods for regulating service behavior include a system provider device where a policy is registered. The policy defines a modified service behavior for a service running one or more remote servers. In some embodiments, the registered policy is transmitted to a first satellite agent located at a first remote server. By way of example, and after transmitting the registered policy to the first satellite agent, data is received from the first satellite agent corresponding to the service having the modified service behavior running on the first remote server. Thereafter, the system provider may verify that the service having the modified service behavior running on the first remote server satisfies a metric. In various embodiments, and in response to the verifying, the registered policy is transmitted to a second satellite agent located at a second remote server.

Systems and methods for regulating service behavior

Methods and systems are presented for accessing various data values in a record stored in a data store independently from other data values in the record and without taking computationally expensive or storage-intensive procedures such as de-serializing a record based on a primary key. A data service module is provided to access each record in a data store, and generate for the records a table including a map of key-value pairs representing the data values in the records. When the data service module receives a request from an application for accessing a first data value in a first record, the data service module may retrieve a first key-value pair corresponding to the first data value in the first record without de-serializing the record. The data service module then provides a first value of the first key-value pair to the application.

Systems and methods for providing flexible data access

Methods and systems for providing an external scheduler in a stateless application framework are presented. A first instance of a stateless application transmits a job request to a data processing engine, and submits a message to a messaging queue. The message indicates a task to be performed according to a delayed schedule. The first instance of the stateless application terminates after submitting the message. The messaging queue transmits the message to the stateless application framework according to the delayed schedule. Upon receiving the message, a second instance of the stateless application is instantiated. The second instance of the stateless application performs an action based on the message and terminates.

Systems and methods for providing services in a stateless application framework

Growing community of inventors

Campbell, CA, United States of America

Karl Anton Hennig