At&t Intellectual Property I, L.p.

Paul Triantafyllou

Eric Martin Zavesky

Lee Begeja

Zhu Liu

David Crawford Gibbon

Behzad Shahraray

Bernard S Renger

Tan Xu

Raghuraman Gopalan

Jean-Francois Paiement

Jeffrey Stein

Russell Fischer

Mary Keefe Hirsekorn

Ryan Cullinane

Concepts and technologies disclosed herein are directed to machine learning model understanding as-a-service. According to one aspect of the concepts and technologies disclosed herein, a model understanding as-a-service system can receive, from a user system, a service request that includes a machine learning model created for a user associated with the user system. The model understanding as-a-service system can conduct an analysis of the machine learning model in accordance with the service request. The model understanding as-a-service system can compile, for the user, results of the analysis of the machine learning model in accordance with the service request. The model understanding as-a-service system can create a service response that includes the results of the analysis. The model understanding as-a-service system can provide the service response to the user system.

Machine learning model understanding as-a-service

Aspects of the subject disclosure may include, for example, a method including obtaining an advertisement package, wherein the advertisement package defines an interactive extended reality advertisement and includes a plurality of optional features; obtaining information about a user, their equipment, and their environment; creating an interactive extended reality advertisement by choosing a selected feature of the plurality of optional features according to the user information; and presenting the interactive extended reality advertisement to the user equipment. Other embodiments are disclosed.

System for delivery of XR ad programs

A method for validating objects appearing in volumetric video presentations includes obtaining a volumetric video presentation depicting a scene, wherein the volumetric video presentation is associated with a metadata file containing identifying information for the scene, identifying user-generated content that depicts the scene, by matching metadata associated with the user-generated content to the metadata file associated with the volumetric video presentation, comparing a first object appearing in the volumetric video presentation to a corresponding second object appearing in the user-generated content, assigning a score to the first object based on the comparing, wherein the score indicates a probability that the first object has not been manipulated, and altering the volumetric video presentation to filter the first object from the volumetric video presentation when the score falls below a threshold.

Validating objects in volumetric video presentations

Methods, computer-readable media, and apparatuses for composing a video in accordance with a user goal and an audience preference are described. For example, a processing system having at least one processor may obtain a plurality of video clips of a user, determine at least one goal of the user for a production of a video from the plurality of video clips, determine at least one audience preference of an audience, and compose the video comprising at least one video clip of the plurality of video clips of the user in accordance with the at least one goal of the user and the at least one audience preference. The processing system may then upload the video to a network-based publishing platform.

Video-log production system

Aspects of the subject disclosure may include, for example, comparing an input received from a peripheral device associated with an execution of a gaming application with a threshold value, wherein the threshold value is based on a first identification of a first user, a second identification of the peripheral device, and a third identification of stimuli presented as part of the execution of the gaming application. Responsive to the comparing, a determination may be made that the input exceeds the threshold value. Responsive to the determination, a validation request may be transmitted to a user device of the first user. Other embodiments are disclosed.

Method to detect and counteract suspicious activity in an application environment

A processing system including at least one processor may receive a task request for a user from a requesting party, identify a trust profile for the requesting party, identify a first automated system to access to fulfill the task request, determine whether the trust profile for the requesting party permits an access to the first automated system, and fulfill the task request via the first automated system when it is determined that the trust profile for the requesting party permits the access to the first automated system.

Context-based automated task performance for user contacts

Aspects of the subject disclosure may include, for example, providing media content to a communication device. The communication device provides a playback of a presentation of the media content. Further embodiments can include receiving an indication from the communication device that indicates a pause in the presentation of the media content, and determining a plurality of attributes associated with the pause. Additional embodiments can include providing instructions according to the plurality of attributes associated with the pause to the communication device. Other embodiments are disclosed.

Methods, systems, and devices for providing a user experience associated with a pause in presenting media content

A processing system having at least one processor may establish a communication session between a first communication system of a first user and a second communication system of a second user, the communication session including first visual content, the first visual content including a first visual representation of the first user, and detecting a first action of the first visual representation in the first visual content in accordance with a first action detection model. The processing system may modify, in response to the detecting the first action, the first visual content in accordance with a first configuration setting of the first user for the communication session, which may include modifying the first action of the first visual representation of the first user in the first visual content. In addition, the processing system may transmit the first visual content that is modified to the second communication system of the second user.

Identifying and addressing offensive actions in visual communication sessions

An example method may include a processing system including at least one processor detecting an interaction of a first user and a second user, providing a temporary authorization to the second user to access a data set based upon an authorization of the first user to access the data set, wherein the providing the temporary authorization is in response to the detecting the interaction, generating a record of an access of the second user to the data set, wherein the record includes a notation of the temporary authorization of the second user to access the data set based upon the authorization of the first user, detecting an end to the interaction of the first user and the second user, and revoking the temporary authorization of the second user to access the data set in response to the detecting of the end of the interaction.

Proximity based data access restrictions

Methods, computer-readable media, and devices are disclosed for improving an object model based upon measurements of physical properties of an object via an unmanned vehicle using adversarial examples. For example, a method may include a processing system capturing measurements of physical properties of an object via at least one unmanned vehicle, updating an object model for the object to include the measurements of the physical properties of the object, where the object model is associated with a feature space, and generating an example from the feature space, where the example comprises an adversarial example. The processing system may further apply the object model to the example to generate a prediction, capture additional measurements of the physical properties of the object via the at least one unmanned vehicle when the prediction fails to identify that the example is an adversarial example, and update the object model to include the additional measurements.

Object modeling with adversarial learning

In one example, a method includes monitoring conditions in a real environment in which a user is present, wherein the monitoring is performed by collecting data about the conditions from a plurality of sensors located in the real environment, transmitting the data about the conditions to an extended reality device that is present in the real environment, where the extended reality device is configured to render a virtual environment, interpolating between the real environment and the virtual environment, based at least in part on the conditions, to determine an actual extended reality environment that is being presented to the user, and sending a signal to a device that is located in the real environment, based on the interpolating, wherein the signal instructs the device to take an action that modifies at least one of the conditions in the real environment.

Growing community of inventors

Summit, NJ, United States of America

Paul Triantafyllou