Microsoft Technology Licensing, LLC

Facebook, Inc.

Meta Platforms, Inc.

Linkedin Corporation

Lijun Tang

Joel Daniel Young

Xu Miao

Yitong Zhou

Anmol Bhasin

Emanuel Alexandre Strauss

Eric Huang

Xingyao Ye

Taiyuan Zhang

David Hardtke

Huihong Zhao

Yiqiu Liu

Yue Shi

An online system distributes content items describing events to one or more users of the online system. The online system receives, an event from a third-party system, the event associated with one or more content items. The online system determines a vector representation of users based on a first neural network and a vector representation of an event based on a second neural network. The online system jointly trains the first neural network and second neural network based on labels describing user entity relationships. The online system determines a likelihood of attendance of an event by a user based on a distance between the vector representation of the user and the vector representation of the entity. The online system provides the content associated with the event to users of the online system based on the likelihood of attendance of the event by the users.

Deep learning based distribution of content items describing events to users of an online system

An online system causes a graphical user interface to display at a client device. The graphical user interface includes a story field that displays ephemeral content items which are created within a threshold time period and are automatically removed after the time period. The online system uses a computer learned model to rank the selected ephemeral content items for display. The display of a set of ephemeral content items is associated with a session. The computer learned model is trained with sample sets that use an entire past session that includes a plurality of ephemeral content items. The computer model proposes a ranked order of the content items in the past session. Based on the past user actions performed on the past content items, a normalized discounted cumulative gain is determined for the past session. The computer model is trained to optimize the normalized discounted cumulative gain.

Ephemeral item ranking in a graphical user interface

A social networking system validates content items based on policies regarding use of content items. The social networking system balances the validation process of content items with at least two validation queues, each having a priority value and some number of content items. A validation queue is selected based on the priority values. From the selected validation queue, a content item is determined to be valid or invalid. A validation efficiency for each queue is calculated based on the validated content items from each queue. The social networking system dynamically adjusts the priority value of each queue based on the validation efficiencies. The dynamic adjustment may be periodic. The social networking system may withhold invalid content items from users.

Dynamic adjustment in priority values of validation queues

The disclosed embodiments provide a method and system for performing regularized model adaptation for in-session recommendations. During operation, the system obtains, from a server, a first global version of a statistical model. During a first user session with a user, the system improves a performance of the statistical model by using the first global version to output one or more recommendations to the user and using the first global version and user feedback from the user to create a first personalized version of the statistical model. At an end of the first user session, the system transmits an update containing a difference between the first personalized version and the first global version to the server for use in producing a second global version of the statistical model by the server.

Regularized model adaptation for in-session recommendations

A system and method for managing asynchronously receiving updates and merging updates into global versions of a statistical model using version control are disclosed. During operation, the system transmits a first global version of a statistical model to a set of client computer systems. Next, the system obtains, from a first subset of the client computer systems, a first set of updates to the first global version. The system then merges the first set of updates into a second global version of the statistical model. Finally, the system transmits the second global version to the client computer systems asynchronously from receiving a second set of updates to the first and/or second global versions from a second subset of the client computer systems.

Version control for asynchronous distributed machine learning

Learning to rank modeling in the context of an on-line social network is described. A learning to rank model can learn from pairwise preference (e.g., job posting A is more relevant than job posting B for a particular member profile) thus directly optimizing for the rank order of job postings for each member profile. With ranking position taken into consideration during training, top-ranked job postings may be treated by a recommendation system as being of more importance than lower-ranked job postings. In addition, a learning to rank approach may also result in an equal optimization across all member profiles and help minimize bias towards those member profiles that have been paired with a larger number of job postings.

Growing community of inventors

San Jose, CA, United States of America

Lijun Tang