Adobe, Inc.

Somdeb Sarkhel

Viswanathan Swaminathan

Saayan Mitra

Ryan A Rossi

Shivakumar Vaithyanathan

Anup Bandigadi Rao

Tung Mai

Ritwik Sinha

Xiang Chen

Thomas Greger

Alok Kothari

Kelly Quinn Nicholes

Nathan Ng

Jiatong Xie

Antonio Cuevas

William Brandon George

Handong Zhao

Stefano Petrangeli

Trevor Paulsen

Kshitiz Garg

Sheng Li

Dwight O Rodgers

David Arbour

Ali Aminian

Jason Boyer

Trisha Mittal

Wreetabrata Kar

Han Guo

Lakshmi Shivalingaiah

Shuo Yang

Swapneel Mehta

Xueyu Mao

Qi Lou

Julio Alvarez Martinez, Jr

Md Maminur Islam

Systems and methods for dynamic user profile management are provided. One aspect of the systems and methods includes receiving, by a lookup component, a request for a user profile; computing, by a profile component, a time-to-live (TTL) refresh value for the user profile based on a lookup history of the user profile; updating, by the profile component, a TTL value of the user profile based on the request and the TTL refresh value; storing, by the profile component, the user profile and the updated TTL value in the edge database; and removing, by the edge database, the user profile from the edge database based on the updated TTL value.

Dynamic user profile management

Various disclosed embodiments are directed to using one or more algorithms or models to select a suitable or optimal variation, among multiple variations, of a given content item based on feedback. Such feedback guides the algorithm or model to arrive at suitable variation result such that the variation result is produced as the output for consumption by users. Further, various embodiments resolve tedious manual user input requirements and reduce computing resource consumption, among other things, as described in more detail below.

Media content enhancement based on user feedback of multiple variations

The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating automatic suggestions to effectively modify the organization of an ingested data collection without destruction of the underlying raw data. In particular, in one or more embodiments, the disclosed systems utilize multiple machine learning models in sequence to determine likelihoods that the organizational structure of an ingested data collection should be modified in various ways. In response to generating these likelihoods, the disclosed systems generate corresponding automatic suggestions to modify the organization of the ingested data collection. In response to a detected selection of one or more of the automatic suggestions, the disclosed systems read data out of the ingested data collection in accordance with the selected automatic suggestions to effectively modify the organization of the ingested data collection.

Generating and executing automatic suggestions to modify data of ingested data collections without additional data ingestion

Techniques are disclosed for real-time bidding (e.g., for dynamic online content placement) using an optimized final bid. The final bid is determined based on a predicted clearing price and an initial bid. The initial bid represents a value to a prospective content provider, and may be computed based on campaign information. The predicted clearing price is a predicted amount paid, and may be predicted using a model trained using historical winning bids data. The clearing price may be predicted using a quantile regression model, where the quantile can be selected to control bid aggressiveness. In some cases, the quantile is determined based on pacing in an overall campaign. Once the initial bid and the predicted clearing price are calculated, the final bid is computed based on the initial bid and the predicted clearing price.

Artificial intelligence techniques for bid optimization used for generating dynamic online content

Systems and methods for dynamic user profile projection are provided. One or more aspects of the systems and methods includes computing, by a prediction component, a predicted number of lookups for a future time period based on a lookup history of a user profile using a lookup prediction model; comparing, by the prediction component, the predicted number of lookups to a lookup threshold; and transmitting, by a projection component, the user profile to an edge server based on the comparison.

Dynamic user profile projection

Techniques for recommending hashtags, including trending hashtags, are disclosed. An example method includes accessing a graph. The graph includes video nodes representing videos, historical hashtag nodes representing historical hashtags, and edges indicating associations among the video nodes and the historical hashtag nodes. A trending hashtag is identified. An edge is added to the graph between a historical hashtag node representing a historical hashtag and a trending hashtag node representing the trending hashtag, based on a semantic similarity between the historical hashtag and the trending hashtag. A new video node representing a new video is added to the video nodes of the graph. A graph neural network (GNN) is applied to the graph, and the GNN predicts a new edge between the trending hashtag node and the new video node. The trending hashtag is recommended for the new video based on prediction of the new edge.

Open-domain trending hashtag recommendations

Automatic item placement recommendation is described. An item placement configuration system receives an item for which a recommended placement is to be generated and identifies an entity associated with the item. The item placement configuration system then identifies a multi-domain taxonomy that describes relationships between different entities based on items associated with the different entities published among different domains. A representation of the entity associated with the item to be placed is then identified within the multi-domain taxonomy, along with a representation of at least one similar entity. Upon identifying a similar entity, historic item placement metrics for the similar entity are leveraged to generate a placement recommendation for the received item. In some implementations, the placement recommendation is output with a visual indication of a similar entity and associated performance metrics that were considered in generating the recommended placement.

Automatic item placement recommendations based on entity similarity

Systems and methods predict a performance metric for a video and identify key portions of the video that contribute to the performance metric, which can be used to edit the video to improve the ultimate viewer response to the video. An initial performance metric is computed for an initial video (e.g., using a neural network). A perturbed video is generated by perturbing a video portion of the initial video. A modified performance metric is computed for the perturbed video. Based on a difference between the initial and modified performance metrics, the system determines that the video portion contributed to a predicted user viewer response to the initial video. An indication of the video portion that contributed to the predicted user viewer response is provided as output, which can be used to edit the video to improve the predicted viewer response.

Explainable video performance prediction

Techniques are disclosed for training of factorization machines (FMs) using a streaming mode alternating least squares (ALS) optimization. A methodology implementing the techniques according to an embodiment includes receiving a datapoint that includes a feature vector and an associated target value. The feature vector includes user identification, subject matter identification, and a context. The target value identifies an opinion of the user relative to the subject matter. The method further includes applying an FM to the feature vector to generate an estimate of the target value, and updating parameters of the FM for training of the FM. The parameter update is based on application of a streaming mode ALS optimization to: the datapoint; the estimate of the target value; and to an updated summation of intermediate calculated terms generated by application of the streaming mode ALS optimization to previously received datapoints associated with prior parameter updates of the FM.

Online training and update of factorization machines using alternating least squares optimization

This disclosure relates to methods, non-transitory computer readable media, and systems that determine multiple personas corresponding to a user account for digital content and train a persona classifier to predict a given persona (from among the multiple personas) for content requests associated with the user account. By using the persona classifier, the disclosed methods, non-transitory computer readable media, and systems accurately detect a given persona for a content request upon initiation of the request. Based on determining the given persona, in some implementations, the methods, non-transitory computer readable media, and systems generate a digital-content recommendation for presentation on a client device associated with the user account.

Predicting digital personas for digital-content recommendations using a machine-learning-based persona classifier

The present disclosure relates to systems, methods, and non-transitory computer readable media for utilizing an artificial intelligence framework for generating enhanced digital content and improving digital content campaign design. In particular, the disclosed systems can utilize a metadata neural network, a summarizer neural network, and/or a performance neural network to generate metadata for digital content, predict future performance metrics, generate enhanced digital content, and provide recommended content changes to improve performance upon dissemination to one or more client devices.

Utilizing artificial intelligence to generate enhanced digital content and improve digital content campaign design

The present disclosure relates to systems, methods, and computer readable media that utilize a trained multi-modal combination model for content and text-based evaluation and distribution of digital video content to client devices. For example, systems described herein include training and/or utilizing a combination of trained visual and text-based prediction models to determine predicted performance metrics for a digital video. The systems described herein can further utilize a multi-modal combination model to determine a combined performance metric that considers both visual and textual performance metrics of the digital video. The systems described herein can further select one or more digital videos for distribution to one or more client devices based on combined performance metrics associated with the digital videos.

Utilizing a trained multi-modal combination model for content and text-based evaluation and distribution of digital video content to client devices

The present disclosure is directed toward systems and methods for identifying user segments. In particular, the systems and methods described herein evaluate user session logs to gather media content consumption history information associated with a plurality of users. Additionally, the systems and methods described herein analyze items of media content to identify keywords, genres, and other attributes, and further represent the items of media content as vectors. The systems and methods follow an algorithm to group items of media content into clusters and, based on the clusters of media content, further group users of media content into user clusters (e.g., user segments).

User segment identification based on similarity in content consumption

A metadata generation system utilizes machine learning techniques to accurately describe content of videos based on multi-model predictions. In some embodiments, multiple feature sets are extracted from a video, including feature sets showing correlations between additional features of the video. The feature sets are provided to a learnable pooling layer with multiple modeling techniques, which generates, for each of the feature sets, a multi-model content prediction. In some cases, the multi-model predictions are consolidated into a combined prediction. Keywords describing the content of the video are determined based on the multi-model predictions (or combined prediction). An augmented video is generated with metadata that is based on the keywords.

Growing community of inventors

San Jose, CA, United States of America

Somdeb Sarkhel