US Patent Application 18350860. PRIVACY-SENSITIVE TRAINING OF USER INTERACTION PREDICTION MODELS simplified abstract

From WikiPatents
Jump to navigation Jump to search

PRIVACY-SENSITIVE TRAINING OF USER INTERACTION PREDICTION MODELS

Organization Name

Google LLC


Inventor(s)

Lukas Zilka of Zürich (CH)

PRIVACY-SENSITIVE TRAINING OF USER INTERACTION PREDICTION MODELS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18350860 titled 'PRIVACY-SENSITIVE TRAINING OF USER INTERACTION PREDICTION MODELS

Simplified Explanation

- This patent application describes methods, systems, and apparatus for collaboratively training an interaction prediction machine learning model. - The goal is to train the model using multiple user devices while respecting user privacy. - The machine learning model processes a search query and a data element as input. - It generates an output that represents the likelihood of a user interacting with the data element if it were presented on a webpage. - The model uses search results that are responsive to the search query to make predictions. - The innovation focuses on privacy by ensuring that user data is protected during the collaborative training process. - The patent application emphasizes the use of computer programs encoded on a computer storage medium to implement the methods and systems described. - The abstract provides a high-level overview of the patent application, highlighting the key aspects and objectives of the innovation.


Original Abstract Submitted

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for collaboratively training an interaction prediction machine learning model using a plurality of user devices in a manner that respects user privacy. In one aspect, the machine learning model is configured to process an input comprising: (i) a search query, and (ii) a data element, to generate an output which characterizes a likelihood that a given user would interact with the data element if the data element were presented to the given user on a webpage identified by a search result responsive to the search query.